Notes:

[1] NC Buddhist Vihara's chanting of the Karaniya Metta Sutta is excerpted from http://www.employees.org/~sammuthu/public/pirith1.wma. Bodhi Lankarama's chanting of the Karaniya Metta sutta is excerpted from http://www.ancient-buddhist-texts.net/Texts-and-Translations/Safeguard/index.htm. For more information, see the "Acknowledgments."

For web browsers that do not embed mp3 players, a clickable blue "play" icon () is displayed; click on this icon to popup an mp3 player that will play the selected chant. Note that this page's "LOOP" feature does not work with these browsers.

Note for Firefox users: During the testing of this page with different browsers, Firefox appeared to have difficulty with consistently displaying the 104 embedded MP3 files on this page. If instead of an MP3 play button you see a white box with a red "X" in it, you can still play the embedded MP3 recording by right-clicking on the white box and then left-clicking on "Play/Pause."

[2] The primary source of this page's Pali text (found twice in the Pali canon: at Khuddakapāṭha [Khp] 9; and, at Sutta Nipāta [Sn] 1.8, verses 143-152) is the Sri Lanka Tripitaka Project (SLTP) redaction, obtained from Bodhgaya News, Gunaratana (2008), and Metta Net. In addition, in these end notes, primarily for the purpose of metrical analysis, occasional reference is made to the Chaṭṭha Saṅgāyana CD (CSCD) redaction, obtained from VRI.

[3] The invaluable metrical analysis of this sutta in Ānandajoti (2007) was repeatedly referenced to confirm or correct this web page's own scansion of this sutta.

For a few notes regarding the canonical context, commentarial backstory and modern metrical analysis concerning this sutta, see "Dating the Metta Sutta."

[4] The five measure (Pali: gaṇa) bars that vertically transverse this entire sutta are meant to support apprehension of this sutta's regular meter. For instance, if you scroll down the page of Pali text, you will notice that, in general (except for verse 10), every other line is the same length:

• the first and third half lines (half lines a. and c.) are the same length, three measures and one syllable;
• the second and fourth half lines (half lines b. and d.) are also the same length, four measures and one syllable.

Similarly, within a specific measure on alternate lines, the sequence of heavy and light syllables is similar. For instance, in the second and fourth half lines of a verse, the second measure is usually composed of two heavy syllables, and the fourth measure usually contains two light syllables followed by one heavy syllable. Recognizing these rhythmic regularities helps one anticipate and memorize this sutta's chanting.

For more information on pertinent Pali prosody, see the general "Primer on Pali versification" and the "Primer on Metta Sutta versification." For an on-line text that clearly marks out this entire sutta in terms of heavy and light syllables, see Ānandajoti's invaluable "Catubhāṇavārapāḷi, A New Edition" (last update: June 2009), "Karaṇāyamettasuttaṃ." Also see Ānandajoti, 2004, & 2006; and, Warder, 1967 & 1970.

[5] Both the SLTP and CSCD redactions identify the first three Pali words of this sutta as being a single compound: Karaṇīyamatthakusalena. On this page, to increase comprehensibility, the three words are separated by hyphens: Karaṇīyam–attha–kusalena. However, in terms of chanting, the final "m" (or, if the first word were uncompounded, "ṃ") of "karaṇīyam" and the opening "at" of "attha" are chanted as one syllable: "mat."

[6] In the SLTP redaction of this text, half-line "2d" must have one non-standard five-mattā gaṇa. Thus, for instance, on this page, the third gaṇa has five mattās:

gaṇa:	1	2	3	4
accents:	— o	— —	o — o o	o o —	—
2d.	ap • pa •	gab • bho	ku • le • su   a •	na • nu • gid •	dho

Alternate solutions can be found for making this half-line more consistent with the sutta's regular meter.

Ānandajoti (2007), sutta 9, represents "kulesu" as "kulĕsu" and states in n. 04: "Metre: sometimes a syllable has been taken as light [metri causa], against its normal weight, this is indciated by a breve ... above the syllable in question." (In addition, Ānandajoti adds a ligature above the first two syllables of this word to indicate that the two initial light stresses could be scanned as one heavy syllable, to be more consistent with other gaṇas in this meter.) The resultant third gaṇa would appear metrically as:

gaṇa:	1	2	3	4
accents:	— o	— —	o o o o	o o —	—
2d.	ap • pa •	gab • bho	ku • lĕ • su   a •	na • nu • gid •	dho

Alternately, the CSCD redaction (at http://www.tipitaka.org/romn/cscd/s0501m.mul8.xml) combines the last two words, eliding an intermediary syllable: kulesvananugiddho. This latter solution creates a third gaṇa consistent with most other even-numbered half-lines in this sutta:

gaṇa:	1	2	3	4
accents:	— o	— —	o — o	o o —	—
2d.	ap • pa •	gab • bho	ku • le • sva •	na • nu • gid •	dho

Nonetheless, each of the recorded chants of this sutta used on this web page articulates "kulesu" as identified in the SLTP redaction without substitution or elision.

[7] In the SLTP redaction of half-line "3a," there are 15 mattās — two more mattās than typically found before the caesura in this meter. As a result, Ānandajoti marks this line as "hypermetric" and includes the six-mattā phrase na ca khuddaṃ in one gaṇa (which is followed on this web page). In contrast to the SLTP's use of the ten-mattā phrase khuddaṃ samācare, the CSCD text uses the eight-mattā term khuddamācare (where ācare is perhaps a conjugate of ācarati?), thus resolving this half-line to the standard 15 mattās.

[8] In terms of scansion, Ānandajoti identifies the first syllable of khemino as "khĕ" (with a short ĕ) and associates it with a light (o) accent. In terms of the two recordings embedded on this page, the first syllable ("khe") appears to be heavy (with a long e). Thus, here, the syllable is represented as heavy (—) and the associated gaṇa is irregular ( — — o ).

[9] In the SLTP redaction of this text, half-line "4b" has 19 mattās, creating an irregular even-numbered half-line. Alternate solutions can be found for making this half-line more consistent with the sutta's regular meter.

Ānandajoti (2007), sutta 9, n. 06, states: "Metre: we need to read short a in vā twice in this line [metri causa]." (Such a substitution is identified as generally acceptable in the PTS PED, p. 606, definition of "vā", where it states: "In verse vā is sometimes shortened to va.") The resultant 17 mattās would appear metrically as:

gaṇa:	1	2	3	4
accents:	o —	o — o	— o o	o o —	—
4b.	ta • sā	va   thā • va •	rā   va   a •	na • va • se •	sā

Alternately, the CSCD redaction (at http://www.tipitaka.org/romn/cscd/s0501m.mul8.xml) simply combines the last two words, eliding the long a: v'anavasesā. This latter solution creates a second and third gaṇa consistent with most other even-numbered half-lines in this sutta:

gaṇa:	1	2	3	4
accents:	o —	— —	— o o	o — o	—
4b.	ta • sā	vā   thā •	va • rā   v'a •	na • va • se •	sā

Nonetheless, each of the recorded chants of this sutta used on this web page articulates all 19 mattās identified in the SLTP redaction without substitution or elision.

[10] The SLTP redaction of half-line "4c" has 15 mattās (see, e.g., http://www.bodhgayanews.net/tipitaka.php?title=&record=8041, verse 146, second line), one or two more than the standard odd-numbered half-line for this sutta's standard meter. Ānandajoti (2007) notes: "Metre: in the first half of the pādayuga we need to read short a in va [metri causa] or possibly yĕ...." (Ānandajoti's first suggestion is followed here in terms of the grouping of mattās into gaṇas.)

The CSCD redaction (at http://www.tipitaka.org/romn/cscd/s0501m.mul8.xml) replaces the SLTP's ye ... vā with yeva: Dīghā vā yeva mahantā. The resultant half-line could thus be scanned in regular meter as:

gaṇa:	1	2	3
accents:	— —	— —	o o —	—
4c.	Dī • ghā	vā   ye •	va   ma • han •	tā

This web page's recorded chants use the irregular meter established in the SLTP text.

[11] Bodhgaya News' SLTP Khp 9 text spells this word as majjhamā while the SLTP redaction of Khp 9 provided by MettaNet, Gunaratana (2008), and Ānandajoti (2007) as well as VRI's CSCD text all use majjhimā. This web page follows the latter spelling.

[12] Unusual for lines of irregular meter in this sutta, the SLTP redaction's line "4d" is actually short one mattā. As a result, in terms of the meter, Ānandajoti suggests: "we could read rassakāṇukkathūlā, with a doubling of the consonant [i.e., the second 'k']...." The CSCD redaction leaves rassakā uncompounded: majjhimā rassakā aṇukathūlā.

[13] The syllables represented here as yeva are represented differently by different sources. As is done here, Gunaratana (2008, p. 46), the BodhgayaNews SLTP redaction, and the CSCD redaction represent the syllables as yeva; the MettaNet SLTP redaction represents them as ye va; the PTS PED (1921-25, p. 320) represents them as ye vā; and, Ānandajoti (2007) represents them as ye ca.

[14] As with line "4c" (see note [10] above), for the sake of metrical conformity, Ānandajoti suggests here that vā could be represented as va, or that ye could be represented as yĕ. Thus, for example, following up on the first of these suggestions, this half-line could be represented in standard meter as:

gaṇa:	1	2	3
accents:	— —	o — o	— —	—
5a.	Diṭ • ṭhā	va   ye • va	ad • diṭ •	ṭhā

Alternately, instead of addiṭṭha (which is consistent with this word's spelling in the PTS PED), the CSCD redaction has adiṭṭha, thus allowing for the following standard meter:

gaṇa:	1	2	3
accents:	— —	— —	o o —	—
5a.	Diṭ • ṭhā	vā   ye •	va   a • diṭ •	ṭhā

[15] Consistent with the standard meter for this sutta (and allowed by the PTS PED), the CSCD redaction uses va where here (in accord with the SLTP text) vā is used.

[16] The line as represented here is consistent with both the SLTP redaction and the recorded chants embedded in this page. However, as represented here, the line's fourth gaṇa is irregular. To make the text more consistent with the standard meter, Ānandajoti (2007) replaces the heavy naṃ with a light năṃ. Similarly, the CSCD redaction uses na instead of naṃ.

Moreover, the CSCD redaction indicates that the last two words of this line are represented variously by different redactions with this square-bracketed note:

na kañci [naṃ kañci (sī. pī.), naṃ kiñci (syā.), na kiñci (ka.)]

Relatedly, the na(ṃ) in this line appears to possibly be open to different translations, as indicated by the PTS PED (1921-25), p. 344, where it tentatively states in its first of three definitions of "Na": "Perhaps at Sn 148."

[17] In his translation, Ñanamoli (1987) reverses the order of what are here represented as Pali text lines 6c and 6d. However, to facilitate readers' understanding of this sutta's meaning in English, on this web page Ñanamoli's translations of line 6c and 6d have been inverted (indicated by the introduction of ellipses).

[18] Bodhgaya News' SLTP Khp 9 text spells this word as aparimānaṃ while the SLTP redaction of Khp 9 provided by MettaNet, Gunaratana (2008), and Ānandajoti (2007) as well as VRI's CSCD text all use aparimāṇaṃ. This page follows the latter spelling.

[19] The line as represented here is consistent with both the SLTP redaction and the recorded chants embedded in this page. However, as represented here, the line's third gaṇa is irregular. To make the text more consistent with the standard meter, Ānandajoti (2007) replaces averaṃ's heavy ve with a light vĕ. The CSCD redaction compounds the last two words, averamasapattaṃ, thus eliding a mattā.

[20] The SLTP redaction of half-line "9b" has an excess of mattās. Ānandajoti (2007, n. 10) notes: "Metre: the second half of the pādayuga is very irregular, and just how we should scan it is not clear." The CSCD redaction maintains consistency with the presumed Old Gīti meter by eliding the vā and using vitamiddho (instead of vigatamiddho) for the last word. Given the atypical meter, this web page preserves all the syllables of the SLTP redaction while mimicking the implicit scansion of the CSCD redaction (which, for instance, allows for a typical amphibrach in the third gaṇa and a near anapest in the fourth gaṇa).

[21] The SLTP redaction of half-line "9d" has an extra mattā, represented on this web page by a five-mattā third gaṇa. To resolve this with the standard meter, Ānandajoti (2007) lightens vihāraṃ's third syllable by shortening the last vowel: vihārăṃ. The CSCD redaction also lightens vihāraṃ's last syllable by compounding this word with the term that follows it, creating vihāramidhamāhu; thus, metrically, changing the trochaic -raṃ i- into the pyrrhic -ra-mi-.

[22] Given the metrical anomalies in this last verse, Warder (1970), p. 228, n. 1, suggests that this last verse is "a later addition."

Echoing Warder's observation, regarding half-line "10a," Ānandajoti (2007), n. 11, writes: "Metre: it may be we should take the first half of the pādayuga as a Siloka line showing the savipula. If it is Old Gīti it is very irregular." The CSCD redaction partially ameliorates the irregularity by doubling the g of half-line 10a's last word, anupaggamma.

The intrinsic anomalies are significantly compounded by there being a difference in the division of words between half-lines "10a" and "10b" depending on whether one is parsing the line for pure versification or for chanting. Thus, for example, Ānandajoti (2004, p. 18; similarly, 2007), who is primarily interested in the metric integrity of the verse, parses this line as follows:

Diṭṭhiñ-ca anupagamma,
sīlavā dassanena sampanno

Conversely, Gunaratana (2008, p. 48), who is primarily interested in representing the verse for devotional practices, parses this line this way:

Diṭṭhiñ ca anupagamma sīlavā
Dassanena sampanno

The recordings embedded in this web page appear to follow Gunaratana's parsing. Given that this web page is primarily meant to be an aid to Pali chanting, it is this latter parsing of line 10a/10b that is represented in this page's Pali text.

[23] In his versification of this verse, Ñanamoli (1984) follows the traditional written parsing of this line, before the Pali term sīlavā (English: "or is virtuous"). However, as explained in the immediately preceding note, the traditional chanting of this line breaks after the phrase sīlavā. Thus, Ñanamoli's line break is modified here in the service of clarifying the chanted Pali phrasing.

[24] Regarding the last full line of this sutta (half lines "10c" and "10d"), Ānandajoti (2007), n. 12, writes: "Metre: this is a Classical Gīti line as it stands, with the caesura coinciding with the end of the first half of the pādayuga. But the jagaṇa (o — o) in the second half of the pādayuga is irregular, being in an odd gaṇa instead of an even one." See also Warder (1970), p. 228, n. 1, referenced in the immediately preceding endnote.