:mindblown: a new reading of the Mettasutta that will change everything

This is part of a series of posts where I am posting drafts of various discussions in my Introduction to the Suttanipāta.

Shifting to a positive mood, we next find the famous Mettasutta (“The Discourse on Love” [snp1.8]). Rather than a series of verses on a theme, this is a coherent poem divided roughly into three sections: verses 1–3ab lay the foundation for virtuous conduct; 4cd–6 describe the meditation on love; while verses 7–9 depict the exultant state of liberation that results. Verse 10 is in a different metre and appears to be a later addition, describing in a highly compressed form the development of insight for attaining first stream-entry, then non-return.

I now translate mettā as “love” rather than the Buddhist neologism “loving-kindness”. The latter has become widely accepted, and is justified by arguing that “love” has too much of a sensual connotation. And it is true that Pali distinguishes sensual love (kāma, pema, etc.) from spiritual love (mettā), much like the Greek eros and agape. I once asked a Catholic contemplative monk about this. His native language was Italian. He said there is no equivalent distinction in modern Italian; they just use amore in both cases and let the context make the meaning clear. I adopted the same approach, and it seems to work fine.

I suspect that the real reason for the rendering of “loving-kindness” is that we can be uncomfortable around expressing emotions. “Loving-kindness” is a more distancing word; it’s emotionally cooler than “love”. I prefer the more direct, ordinary language expression.

The opening lines pose something of an interpretive problem, evidenced by the complex discussion in the commentary. The second line refers to someone who has realized the “peaceful state” of Nibbāna (santaṁ padaṁ). The term for “realized” (abhisamecca) is usually reserved for the breakthrough to the four noble truths, eg. [sn56.4:1.1]). Such a person must then be a stream-enterer at least ([sn56.49:1.7], see also āgataphalā abhisametāvinī viññātasāsanā at [pli-tv-bu-vb-ay1:2.1.26]). Is the poem then restricted to the practice for a stream-enterer? Given the universal nature of its themes, this seems unlikely.

Jayawickrama suggests that “the peaceful state” need not mean Nibbāna, but this does not seem to be supported in early Pali. Bodhi addresses this issue in his note 685, where he confirms that a stream-enterer is meant, rejecting the commentary’s alternative explanations. This is clearly the most straightforward reading of the lines.

There remains, however, the problem of how this fits with the remainder of the text. The stream-enterer is “perfected in ethics”. Yet the subject of the Mettasutta is exhorted to have moral integrity, to be admonishable, to not be overly demanding of donor families, and to not act in a blameworthy way, all of which are things that would surely come naturally to a stream-enterer. Further, as we shall see below, the original poem merely taught as far as rebirth in the Brahmā realm, and it is unlikely that the Buddha should teach this to a stream-enterer. Overall, the poem has a high degree of unity and purpose, and this all feels ill-fitting.

Revisiting the commentary, it offers several different approaches. Among them, it suggests the absolutive may be read in an infinitive sense. That is, instead of “what has been realized”, what it means is “in order to realize” (abhisamecca viharitukāmo). While the primary gloss on abhisamecca confirms that it is an absolutive (abhisameccāti abhisamāgantvā), it also explains it in terms of an infinitive (adhigantukāmena). Further, it speaks of a mendicant who “is practicing to attain that state” (tadadhigamāya paṭipajjamāno).

If we are to adopt this reading, we are left with a further puzzle in the term atthakusala. The word attha means many things, but here it is interpreted by the commentary as what is good or beneficial, and this is followed by most or all modern interpreters. It does, however, create a similar problem to that discussed in the previous paragraph, because it is generally considered that only a stream-enterer is accomplished in ethics. Those who have not seen the four noble truths can, of course, live a good life, but they are not yet “experts”. The commentary appears to be aware of this problem, as it offers an infinitive reading for atthakusala, that is, “one who wishes to dwell in the fourfold ethical purity” (catupārisuddhisīle patiṭṭhātukāmo).

However, such a reading is not necessary here, for atthakusala occurs in the Suttas, where it refers to expertise in the scriptures ([an5.169:3.1]):

a mendicant is skilled in the meaning (atthakusala), skilled in the teaching, skilled in terminology, skilled in phrasing, and skilled in sequence.

Now we have a much more satisfying reading to the opening lines. They are addressed to someone who has already mastered the textual teachings and wishes to realize them in practice. This follows exactly the same pattern as practiced by the bodhisatta under his former teachers, where he first learned the scriptures, then undertook meditation. It is also the fundamental framework for monastic practice, the Gradual Training, where the mendicant learns the Dhamma then goes off to meditate.

I therefore translate the opening lines:

Those who are skilled in the meaning of scripture
should practice like this so as to realize the state of peace.

This also allows us to situate the opening of the Mettasutta more precisely with the added final verse, which begins with the words “avoiding views”. This is a major theme of the Suttanipāta, especially the Aṭṭhakavagga, which formed one of the kernels around which the Suttanipāta formed. The redactors who added this verse must have known the Aṭṭhakavagga, with its oft-repeated admonishment to avoid the trap of heated and angry debating on views. It is quite possible, likely even, that this verse was added when the Mettasutta was included in the Suttanipāta, creating a connection with the themes of the Aṭṭhakavagga, offering a meditation path especially suited to those with a tendency to become heated in their arguments.

The Mettasutta urges us to spread love to all beings, including those who are “born or to be born” (bhūtā vā sambhavesī vā). The latter phrase, which occurs only here and in a stock phrase at eg. [sn12.11:1.3], evidently refers to beings who are in the process of taking a new life, an idea that is evidently at odds with the early Theravadin insistence that one life follows immediately after another, with no “in-between state” ([kv8.2]). A majority of early Buddhist schools accepted this state, and this phrase in the Mettasutta joins a list of other contexts in the early texts that show with reasonable certainty that the early Buddhists did too.

This doctrinal point speaks to the earliness of the poem, as does the metre (old Ārya), the coherent flow of ideas, and the focus on the universal subject of mettā without a forced “Buddhistic” perspective, the lack of which evidently prompted the later addition of the final verse.

It is assumed in the Buddhist tradition that mettā and the other brahmavihāras (karuṇā or compassion, muditā or rejoicing, and upekkhā or equanimity) are pre-Buddhistic, and belong with the very many ideas that the Buddha happily adopted from his religious surroundings. While they are not, to my knowledge, attested in any surviving pre-Buddhist texts, they are found in the Yogasūtra (1.33), several later Upaniṣads, the Jain Tattvarthasūtra (7.11), and even a Tibetan Bon text of the eleventh century, “A Cavern of Treasures” (mDzod-phug).

If we look at the way the word brahmavihāra is used in the suttas, we find that apart from the Mettasutta, there is one other verse where the term particularly refers to the practice of mettā ([thag14.1:5.3]). The term is sometimes used for mindfulness of breathing, along with ariyavihāra and tathāgatavihāra, in which case the term brahmā must refer to the being of that name, rather than an abstract sense such as “divine” or “holy” ([sn54.11:3.2], [sn54.12:8.4]).

While the group of four qualities are commonly taught, they are referred to as brahmavihāras specifically to indicate that such a practice leads to rebirth in the Brahmā realm ([mn83:6.2], [an5.192:6.8], [dn17:2.13.8]). Elsewhere the Buddha clarifies that such practices, which he did in his past life, do not lead to Nibbāna, unlike his own eightfold path ([dn19:61.4]). When someone who is a Buddhist disciple practices them, then if they do not realize full enlightenment in this life, they will do so after being reborn in the Brahmā realm ([an4.125], [an4.126])

This lends a greater specificity to the line “this is a meditation of Brahmā in this life”, which would have been the final line of the original poem. The sense is that through this meditation, one can live like the god Brahmā in this life; and such a life leads to being reborn as a Brahmā in the next. While it is unusual to find a Buddhist text that ends with such a rebirth, it is not unique. In the Tevijjasutta the Buddha teaches this path to some brahmin questioners ([dn13]). Sāriputta did the same in [mn97], although the Buddha evidently felt he should have gone further. What is unusual, however, is that such suttas are taught to those who were not committed Buddhists, while the introductory passages of the Mettasutta evidently address Buddhist mendicants.

As with other didactic texts of the early period, the poem is noteworthy for its plain style and lack of metaphors. But this restraint shows no lack on behalf of the poet, for when the metaphor of a mother’s love for her child is introduced, it comes at the climax of the poem. The restraint creates a heightened emotion here, which is one of the secrets behind this poem’s enduring popularity.


Nice! :astonished:

It’s funny because I’ve often thought of the opening “knows the path” (in the old, chanting translation)
as referring to those with doctrinal knowledge. :sweat_smile: Glad there’s some justification for my “mis” reading!


I’m curious of your opinion on the line about the mother protecting her child, bhante.

This is often equated to someone loving and protecting all beings via mettā as a mother would to her child. However, it seems, as some have pointed out, that perhaps this is saying that one protects their mind of mettā in the same way a mother protects her only child, not the beings themselves. Of course, the mettā would entail feelings of kindness and friendliness / well-being to all beings, but one would not be treating them as a mother does. On the other hand, that very mind of mettā must be protected at all costs, always maintained and practiced.


1 Like

I believe we’ve discussed this before on the forum? Try searching and see if it turns up.

1 Like

Here are some relevant threads regarding the mother and child simile:


They are Greek, not Latin. The Latin word for eros is “amor”, for agape I don’t know; not sure if there is one.

Wow. It seems I should perhaps also revisit my translation. :mindblown:

Thanks for the essay. I love the metta sutta even more now! :smiley:


OMG I am revealing my ignorance here! I’ll fix it.


Sometimes diligentia

ἀγαπᾶτε τοὺς ἐχθροὺς
diligite inimicos vestros.
Love your enemies.

and sometimes caritas

μείζων δὲ τούτων ἡ ἀγάπη.
major autem horum est caritas.
but the greatest of these is love.


Well I always thought caritas to be the active side of love, helping people who are suffering, which would be more in line with karuna (and maybe even comes from the same root).

And I also don’t think “love” is in the original spectrum of meaning for diligentia either. The original meaning is rather similar to “carefulness”.

I think these Bible quotes that you mention are rather a later application of Latin, could that be? Or maybe if there are different words used to translate the Greek agape this means this distinction between eros and agape does not originally exist in Latin, just as is now the case in Italian?


I’m not sure. It’s from an online edition of the Vulgate, but the website doesn’t state whether it’s St Jerome’s original one (late 4th century) or from the Nova Vulgata (1979).

1 Like

Well when I learned Latin at high school (long time ago) we would not study Bible texts (strangely enough actually, as it was a Christian school, led by nuns). We studied Julius Caesar for example, and compared to that, Bible translations are late. :smile:

Christianity arrived rather late in the history of Latin language, that’s what I was going to say. I think we didn’t study Bible texts because they wanted to teach us “original” Latin, not later versions. For, like all languages, Latin developed over time before eventually dying out.

But I am not an expert. I forgot most of what I’ve learned.


Also AN 11.15.

Bhante. Could you kindly explain the Pali word “punaretī” and how it is formed; given the multitude of different translations. Thank you :dizzy:

1 Like

It seems to me that all those who have opted to translate the sutta’s last line (as opposed to those who’ve opted to paraphrase it) offer substantially similar renderings:

one never again comes back to the bed of a womb.

The same translator’s rendering of the commentary:

“One absolutely will not come back again to the bed of a womb. Having been reborn in the pure abodes, one reaches arahantship and attains final nibbāna right there.”

And some others:

assuredly does not come to lie again in a womb.
K.R. Norman

He surely comes no more to any womb.

Never in a womb is one born again.

they never return to a womb again.

he will never come to lie in a womb again.

will never again go to a mother’s womb.
Max Müller

Verily never again will he return to conceive in a womb.

one never again will lie in the womb.

Und geht nicht ein mehr in den Mutterschoß.


As for the paraphrases, I don’t have any to hand, but the ones that I recall seeing were mostly by American lay vipassanā teachers and seemed to be aimed at eliminating the idea that the last line has something to do with rebirth.


na hi jātu gabbhaseyyaṃ punaretīti

na - not

hi - indeed, surely.

jātu = jātuṃ, to be born. Final consonant dropped to fit the metre.

gabbhaseyyaṃ - bed which is a womb, bed of the womb, womb which is like unto a bed, lying/sleeping in a womb.

punaretīti = puna + r + eti + iti

puna - again
r - consonant inserted for euphony.
eti: go, come, arrive at
iti: particle showing the end of the speech.


Thank you Venerable. I didn’t realize the meaning of “jātu” was “jati”. Its not what the Pali-English word look up offers.


1. at all; ever; undoubtedly

Thank you. I didn’t know there were two words above. Seyya (“a bed”; per SN 31.13–22 for example) or seyyaka (“lying” prone; per MN 64).

The above is too difficult for the unlearned to work out. Thank you :pray:t2:

1 Like

If I am not mistaken, this may be the very first time I’ve seen you make a mistake in Pali. Amazing!

Jātu is an indeclinable particle of emphasis. It’s listed as such in Pali and Sanskrit dictionaries.

Na hi jātu is an idiom found in eg. mn86:18.23:

Na hi jātu so mamaṁ hiṁse,
aññaṁ vā pana kiñci naṁ;
For then they’d surely wish no harm
upon myself or others.

The commentary glosses it as such also:

ekaṃseneva puna gabbhaseyyaṃ na eti


Whoops. :blush:

Thanks for the correction, bhante.


Though it has the intimidating name of epenthetic excrescence (or āgama in Pali), the insertion of an r (or some other consonant) between two vowels to avoid a clumsy-sounding hiatus is quite an easy feature of Pali sandhi to understand, for we do the very same thing in spoken English.

When Pali speakers make changes like these…

puna + eti = punareti
ni + antaraṃ = nirantaraṃ.
ni + ojaṃ = nirojaṃ.
du + atikkamo = duratikkamo.
du + ājāno = durājāno.
pātu + ahosi = pāturahosi.
catu + ārakkhā = caturārakkhā.

… they are simply doing what Margaret Thatcher did when she pronounced law and order as if it were a woman called Laura Norder, or what John Lennon did when he sang, “I saw a film today, oh boy” as “I sorer film today, oh boy”, or what most English speakers do when they pronounce Kafkaesque as Kafkaresque. The only difference is that in Pali the inserted consonant is always recorded in the spelling, while in English it most often isn’t.

In all Pali has 10 consonants that can be used in this way: t, d, n, m, y, r, l, ḷ, v and h.


Being derived from epi-thesis, “epenthetic” is satisfyingly autological

1 Like

It should be “3cd–6 describe the meditation on love”.

I noticed that you haven’t updated the translation for this on suttacentral, and perhaps not for the other translations you’ve suggested so far in these posts either.

Do you plan on updating the translations after the posts on the Sutta Nipāta are finished? Or do you see the translations in these posts as explaining the detailed implications and underlying meaning of the translation already up on the site?

With mettā