Pali Text Society Translation Series No.

46

The Word of the Doctrine

(Dhammapada),

translated

with an introduction and notes

by

K.R. Norman

Published by
The Pali Text Society
Oxford
2000
 First published 1997
Reprinted with corrections 2JOO

©Pali Text Society 1997

11 ooi; : 1·~;;-1·ri:.11-::i~! 1
· --ISBN----0 86013-335 4- hardbacki----
086013 384 2 paperback

All righls reserved. No part of this publication may be reproduced

or transmitted in any form or by.any means, electronic or
mechanical, including photocopying. recording or any information
storage and retrieval system, without prior permission in writing
frpm I.he Pali Text Society.

Printed in Greal Britain by

Anlony Rowe Lid. Chippenham, Wiltshire
 CONTENTS

Preface vii
Bibliography IX
Abbreviations xv
Introduction XlJ(

I. The Dhammapada genre XlJ(

II. Parallel versions xx

III. The arrangement of the verses xxi
IV. The commentary xxiii
V. The translation XXJV
VI. The title of the translation xxv
vn. The metres ofDhammapada xxv
VIII. Metrical licence xxxi
IX. Orthography and phonology xxxviii
Translation
I. The Pairs (Yamakavagga)
II. Carefulness(Appamadavagga) 4
Ill. Thought (Cittavagga) 6
IV. Flowers (Pupphavagga) 8
V. The Fool (Biilavagga) IO
VI. The Learned Man (PaiJc;litavagga) 12
VIL The Arahant (Arahantavagga) 14
VIII. Thousands (Sahassavagga) 16
IX. Evil (Papavagga) 18
X. Violence (Dai:i<lavagga) 20
XL Old Age (Jariivagga) 22
XII. The Self(Attavagga) . 24
XIII. The World (Lokavagga) - 26
XIV. A wakened (Buddhavagga) 28
XV. Happiness (Sukbavagga) 30
XVI. Pleasant (Piyavagga) 32
JNTI. Anger (Kodhavagga) 34
XVIII. Impurity (M?Javagga) 36
XIX. The Righteous (DhammaWiavagga) 39
XX. The Path (Maggavagga) 41
XX!. Miscellaneous (Pakiiµ:iakavagga) 43
XXII. Hell (Nirayavagga) 45
VI 17ie Word ofrhe Doclrine

XX!ll. The Elephant (Nagavagga) 47

XXIV. Cr::iving (Tanh:ivagga) 49
XXV. The Bhikkhu (Bhikkhuvagga) 52
XXVI. The Brahman (Braiuror;avagga) 55
Notes 61
Index of words discussed or quoted in the notes 167
Inde~ of grammatical points discussed in the notes 173
 ABBREVIATIONS

Editions ofDhammapada and commentary:

OvH/KRN 0. von Hiniiber and KR. Norman, PTS Oxford,

1994
The basis of OvHIKRN 1994 is V. Fausboll, The Dhammapada,
London I 900, with its metrical "corrections"
removed, except where there is manuscript authority
for them.
Ch Cha\lhasangayana ed.. Rangoon I 96 I [ l 962]
C&P J.R. Carter, & M. Palihawadana, I 987
Dhp-a H.C. Norman, 1906-14.
J Buddha Jayanti Tripitalca Series, Vol XXIV, I 960
SST PTS edition by SUriyagO<,ia Suman gala Thera, 1914

Manuscripts:

A: no. 51, ca. AD 1500, verses 60-67: Dhp-a ll 6,7-39,3

B: no. 52, AD 1582, verses 93-99, 412: Dhp-a II 173,3-208,4
C: no. 53, ca. AD 1500, verses 136-51, 233: Dhpca III 61,11-!22,16
D: no. 54, AD 1521, verses 203-22; 325: Dhp-a Ill 261,19-301,19
E: no. 55, early 16th century, verses 348--02, 397: Dhp-a JV 61,5-
91,21
F: no. 56, ca. AD 15000verses412-23: Dhp-aJV 187,I0-236,t3*
H: manuscript written AD 1786 and kept at Vat Laqhivan, Ban
Tai/Hot (north Thailand), folios ka-kah, kha-khah, ga-gu;
folio khal'f! containing verses 329d-343a is lost. The
manuscript has been microfilmed by the Social Research
Institute, University of Chiang Mai, no. 84.134.01 H. 043-
043.
L: complete text (with gaps) written AD 1611 and kept at Vat Lai
~in: 0. v.on Hiniiber: Die Piili-Handschrijien des K/osters
Lai Hin bei Lampang.VOifJ), Supplementa (forthcoming),
manuscript no. 56.

The "no." refers to the "Lai Hin Kata!og" (quoted under "L")
XVI The Word of the Doctrine

Other Texts:

GD hp (=Brough 1962)
PD hp (=Cone 1989)
Udiina-v (B =Bernhard 1965; N = Nakatani 1987)

The abbreviations of the titles of Pali texts are those. adopted by CPD

Translations:

EV I Norman, 1969
EV II Norman, 1971
GD I Norman, 1982
GD II Norman, 1992

Dictionaries & Concordances:

BHSD F. Edgerton, 1954A

CDIAL R.L Turner, 1962
CPD Critical Pali Dictionary
DPPN Dictionary of Pali Proper Names (=
G.P. Malalasekera, 1937-38)
MW Sanskrit-English Dictionary(= M. Monier-Williams,
1899)
PED PTS's Pali-English Dictionary
PSM Paia-sadda-mahai:n:iavo (= H.D.T. Sheth, 1963)
PTC Pali Tipi!akam Concordance

Grammars

BHSG F. Edgerton, 19548

Geiger W. Geig<;r, 1994
Pischel R. Pischel, 1900
Oberbl 0. von Hiniiber, 1986
Whitney W.D. Whitney, 1889
 Abbreviations XVil

Periodicals, Collections and Series:

AO Acta Oricntalia
BDCRI Bulletin of the Deccan College Research Institute
BEi Bulletin d'Etudes lndiennes
BSO(A)S Bulletin of the School of Oriental (and African)
Studies
BSR Buddhist Studies Review
CPI -VJ KRN, Collected Papers I-VJ
HOS Harvard Oriental Series
llJ lndo-Iranian Journal
IL Indian Linguistics
IT Jndologica Taurinensia ·
JA Journal Asiatique
JBRS Journal of the Bihar Research Society
JOl(B) Journal of the Oriental Institute (Baroda)
JPTS Journal of the Pali Text Society
JRAS Journal of the Royal Asiatic Society
OlZ Orientalistische Literaturzeitung
SBB Sacred Books of the Buddhists
SBE Sacred Books of the East
SP OvH, Selected Papers
Stll Studien zur lndol_ogie und lranistik

General:

se Burmese edition
BHS Buddhist Hybrid Sanskrit
ce Sinhalese edition
E< European (romanised)edition
Brough J. Brough, 1962
KRN K.R. Norman
OvH
Pkt
• Oskar von Hiniiber
Prakrit
PTS Pali Text Society
Rau W. Rall, 1959
se Siamese (Thai) edition
Skt Sanskrit
 TRll'IT\K.-\ BOOKS I" l:"CLISll
llO" \TED .\S DIIA\l\l.\D.-\NA
By Layman i\lAO VAN, Counsellor of the Royal Embassy of
Cambodia in the Union of Myanmar and Lay-woman THON
SAMBO altogether with children l\fao Samborath, Mao Monirath,
Mao Samborith, Mao Sophearath, Mao Borei Kanhchana and Mao
Borom Panh-nia (called Sarpejn) and all children-in-law.
We take this opportunity to wish that the BUDDHISM will be widely
practiced in the world, and to share merit achieved from the donation first to
the Most Venerable Samdech CHUON NAT, former Supreme Patriarch of
Cambodia, the most of eminent TRIPITAKA translators from Pali into Khmer
language and to the Esteemed Guru BUTH SAVONG, an eminent Buddhist
teacher who so far has preached the Dhamma, based on TRIPITAKA Books,
more than 2,000 recorder tapes(60m) of Buddhist priests and teaching lessons;
secondly to father MAO SOAN and mother SAN SEN. to father THON HAM
and mother TUY \10M, to grand-fathers and grand-mothers, to great grand- ,
fathers and great grandmothers, to uncles and aunts, to brothers and sisters ·
such as brother Thon Hon (called Chea) and brother Mao Vat, and other relatives
of the two sides in this endless cycle of births and rebirths and to step-mother
Ham and step-father Tap Ho, and step-grandfather Meas Sum, and friends
such as Dr. Aine Kim Long, as well .as all Buddhists of all over the world
especially to th;se who check i::a~'ti~usly their Buddhist knowledge with
TRIPITAKA Books.
Yangon, February 28, 2007
i ". -11ic111 \J\IJ \_-\'.\and Lt\- worn'rn TllO'.\ SA\IBO
 PREFACE:

The Dhammapada is perhaps the Buddhist text which has been

translated into English more often than any other, and some may
doubt the need for yet another translation, but the publication by the
Pali Text Society of a new edition by Oskar von Hiniiber and myself
in 1994 (reprinted with corrections in 1995), replacing the 1914
edition, seemed to be an appropriate time to produce a new annotated
translation replacing that made for the PTS in 1931 by Mrs Rhys
Davids.

In making the edition, great attention was paid to metre, and the
results of the analysis which was made are categorised in the section
of the introduction entitled "Metres of the Dhammapada", and are
included in the Notes. I have also included there, in as much detail as
seemed necessary, the grammatical, syntactical and lexic0graphical
information which I found of help when making the translation.

Over the years I have read many translations of the Dhammapada. in

whole. or in part, and since I see no sense in rejecting a good
translation simply because someone else has-already used it, my
translation 9wes much, consciously or unconsciously, to my
predecessors. They are too many to name, but I acknowledge a
special debt to Radhakrishn;rn, whose edition and translation first
introduced me, as a student, to the Dhamrnapada, and to Max Muller
whose translation I compared with R2dhakrishnan's at that time; to
Carter and Palihawadana and to Kalupahana. whose translations I
read attentively when I reviewed them; to John Brough 's edition of
the Gandhari Dharmapada; to students over the years with whom I
h;:ve read the Gandhari Dharmapada; but most especially to Dr
Margaret Cone for many discussions while she was working on her
dissertation on the Patna Dharmapada and for the numerous
suggesvons she.has made in the dissertation itself.

John Brough is reported as saying, when asked if he would produce

a new transla!ion of the Dhammapada for the PTS, that he could no~
because "it was too difficult". I regret to say that I must agree with
VIII The Word of the Doctrine

him. My notes reveal how of\en I was quite unsure about the
meaning of a verse.

On behalf of the Pali Text Society I should like to express tha.,ks to

Kyung-Jun Lee and Sun•Kyong Lee for their very generous
contribution to the costs of publishing this translation.
 BIBLIOGRAPHY

Alsdorf, L., 1962, "UttJ.rajjhaya Studies'', !!JV!, pp. 110-36 ·

- - , 1975, "Pali Miscellanies: uddhaT"{lSara', St!/ I, pp. I 09-17
Andersen, D., 1904-5, A Pali Glossary, Copenhagen
Bailey, H.W., I 960, "Indagatio lndo-lranica", TPS, pp. 62-86
Balk, M., I 988, Untersuchungen zum Udiinavarga, Bonn
Berglie, Per-Arne, 1995, "The Brahman who tapped on the skulls of
the dead", in Sauhrdyamailgalam (Studies in Honour of
Siegfried Lienhard on his 10th Birthday), Stockholm, pp. 41-
53
Bernhard, F., 1965, Udiina-varga, Vol. I, Gottingen
- - , 1968, Udiina-varga, Vol. II, Gottingen
Bollee, W.B., 1983, Reverse Index of the Dhammapada,
Suttanipiita, Thera- and Ther!gathii Piidas with Parallels
from the Ayiirariga, Suyagacfa, Ullarajjhiiyii, Dasaveyiiliya
and lsibhiisiyiiil[l. Studien zur lndologie und Iranistik.
· Monographie 8. Reinbek 1983 (rev.: K.R. Norman, OLZ
1988, 341-44; O.v. Hiniiber, WZKS 31, 1987, 201 foll.)
Brough, J., 1962, The Giindhiiri Dharmapada, London
Burlingame, E.W., I 921, Buddhist Legends, 3 Volumes, HOS
Cai Ila~ C., 1991, (ed.) Middle Jndo-Aryan and Jaina Studies, Leiden
- - , 1993, "Words for violence in the 'Seniors' of the Jaina
canon", in Jain S1udies in Honour of Jozef Deleu, Tokyo,
pp.207-36
Carter, J.R. & Palihawaciena, M., 1987, The Dhammapada, New
York
Chakravarti, N.P., 1930, L "Udanavarga sanskril, Tome premier
(Chapitres I a XXI), Paris
Collins, S., 1982, Selfless Persons, Cambridge
- - , 1987, "KalyaQamitta and Kalyfu)amittat:l'', JPTS XI, pp. 51-
72
Cone, }.;., 1986, The Parna Dhannapada, unpublished Ph.D.
dissertation, Cambridge
- - , 1989, "The Patna Dharmapada. Part I: Text", JPTS XIII,
pp. 101--217
Edgerton, F., 1953A, Buddhist Hybrid Sanskrir Dictionary, New
Haven
 x The Word oflhe Doclrine

- - , 19538, Buddhisl Hybrid Samkril Grammar, New Haven

Fausboll, V., 1900, The Dhammapada, London
Geiger, W., i 994, Pali Grammar, PTS Oxford
Gonda, J., 1968, "Abbreviated and inverted nominal compounds in
Sanskrit'', in Pratidiinam: Indian, Iranian and fndo-
European Studies presented to F.B.J. Kuiper, The Hague,
pp. 221--46
Hara, M., 1992, "A note on Dhanrmapada 9T', !IJ 35, 213, pp. 179-
91
- - , 1995, "A note on the phrase lqfo dhamani-samtala",
Asialische Studien/Etudes Asiatiques, XUX, 2, pp. 377-89
Hi_niiber, 0. von, 1968, Studien zur Kasussyntax des Pali,
besonders des Vinaya-Pi!aka, Miinehen
- - , 1977-78, "Notes on nominal composition in.Pali 1: Three
abbreviated compounds", in LN. Mishra Volume, JBRS
63/64, pp. 817-21
- - , 1981, "A Vedic verb in Pali: udajita", in S1ernbach
Felicitation Volume, Lucknow, pp. 819-22
- - , 1986, Das altere Mittelindisch im Oberblick, Wien
- - , 1994, Selected Papers in Pali Studies, PTS Oxford
Hiniiber, 0. van and Norman, K.R. {eds), 1994, Dhammapada,
PTS Oxford
Kalupahana, D.J., 1986, Dhammapada. The Palh of Righleousness,
Lanham
·La Vallee Poussin, L de, 1912, "Fragments de l'Udinavarga de
Dharrnatrata", IRAS, 1, pp. 355-77
Lesny, V., 1924-27, "A new reading of Dhammapada 20T', JPTS,
pp. 235-36
Liiders, H., 1940, Philologia fndica, Giittingen
- - , 1954, Beobachtungen uber die Sprache des buddhis1ischen
Urkanons, Berlin
Malalasekera, G.P., 19.37-38, Dictionary of Pali Proper Names,
London
Mehendale, M.A., 1955-56A, Review ofliiders 1954, BDCR! 17,
pp. 53-75
- - , 1955--568, "Some remarks on the language of the original
Buddhist canon", BDCR! 17, pp. 157-71
 Bibliography XI

Monier-Williams, Sir Monier, 1899, Sanskrit-English Dictionary,

new edition, Ox ford
Muller, Max, 1881, 71ie Dhammapada, SEE X, Oxford
Nakatani, H_, 1987, Udiinavarga de SubaJi, 2 p;:rts, Paris
- - , 1988, (Japanese edition)
NaQamoli, Bhikkhu, 1994, A Pali-English Glossary of Buddhist
Technical Terms, Kandy
Norman, H.C., l 906-l 4, Dhammapada11hakathii. The commentary
on the Dhammapada, 4 volumes, PTS London
Norman, KR:, 1960, "Middle Inda-Aryan Studies (I)", JO!(B), IX,
pp. 268-73 (=CPI, pp. 1'-20)
- - , 1961, "Middle Inda-Aryan Stutes II",JO!(B) X, pp. 348-52
(= CP I, pp. 25-29)
- - , 1962, "Middle Inda-Aryan Studies Ill", JO!(B) XI, PP- 322-
27 (=CPI, pp. 30-35)
- - , 1967, "Notes on the Afokan Rock Edicts", IIJX, PP- 16G--70
(=CPI, pp. 47-58)
- - , 1969, Elders' Verses, Vol. I, PTS London
- - , l97IA, Elders' Verses, Vol. II, PTS London
- - , 197 IB, ''Notes on Loe Gandhari Dhamapada", IL 32, pp. 213-
20 (=CPI, pp. l 13-21)
- - , 1974, "The Gandhari version of the Dhammapada", in
Buddhist Studies in honour of l.B. Horner, Dordrecht,
PP· 111-80 (= er r. pp. 110-80)
- - , I976A, "Middle Indo-Aryan Studies XIII", JO!(B) XXV,
pp. 328-42 (= CP l, pp. 220-37)
- - , l976B, "Labialisation of vowels in Middle Indo-Aryan'',St!I
II, PP- 41-58 (=CPI, pp. 247---0l)
- - , J979A, "Dhammapada 97: a misunderstood paradox", /TVll,
l 979, pp. 325-31 (= CP II, pp. I 87-93).
- - , 19798, "Two Pali etymologies", BSOAS 42, pp. 321-28 (=
CP II, PP- 71-83)
- - , i979C, ''.Middle lndo-Aryan Studies XIV", JO!(B) XXlX,
pp. 37-41 (= CP II, PP- 113-18)
- - , l 979D, "Middle Inda-Aryan Studies XV", JOI(B) XXIX,
P!'· 42-49 (= CP II, pp. 119-,-27)
- - , 1979E, Review of A Critical Piili Dictionary, Vol. II, 7-8, in
OLZ 74, 4, pp. 391-94 (= CP VJ, PP- 146-49)
Xll The Word of the Doctrine

- - , 1980, "The dialects in which the Buddha preached", in The

language of the earliest Buddhist tradition, Gottingcn,
pp. 61-77 (= CP II, pp. 128-47)
- - , 1981, "A not.eon attain Loe Alagaddi1pama-sutta", in Studies
in Indian Philosophy (Memorial volume for Pandit Sukhlaji
Sanghvz), Alunedabad, pp. 19-29 (= CP II, pp. 201-09)
- - , 1984, Group of Discourses, Vol. I, PTS London
- - , 1985, "The influence of the Pali commentators and
grammarians upon the Theravadin tradition", Buddhist Studies
(Bukkyo KenkyU), XV, pp. 109-23 (= CP Ill, pp. 95-107).
- - , 1987, ''Pali Lexicographical Studies IV", JPT'S XI, pp. 33-49
(= CP IR pp. 157-72)
-"-, 1989A, "Pali Lexicographical Studies VJ'', JPTS XIII,
pp. 219-27 (= CP IV, pp. 72-79)
- - , 19898, "On translating Llie Dhanunapada'', BSK6, 2, pp. 153--
65 (= CP VI, pp. 156-70)
- - , 1989C, "Dialect forms in Villi", in Dialectes dans /es
/itteratures indo-aryennes, Paris, pp. 369-92 (= CP IV,
pp. 46-71)
- - , l 990A, Collected Papers, Vol. I, PTS Oxford
- - , 19908, "Pali Lexicographical Studies VII", JPTS XIV,
pp. 219-25 (= CP IV, pp. 139-45)
- - , 1991A, Collected Papers, Vol. II, PTS Oxford
- - , l 991B, "Syntactical compounds in Middle lndo-Aryan", in
Middle Inda-Aryan and Jaina Studies, Leiden, pp. 3-9 (= CP
IV, pp. 218-25)
- - , l 992A, Collected Papers, Vol. lIT, PTS Oxford
- - , 19928, Group of Discourses, Vol. II, PTS Oxford
- - , l 992C, "The nasalisation of vowels in Middle lndo-Aryan",
in Philosophy, Grammar and lndalogy (Festschrift G. Roth),
Delhi, pp. 331-38 (= CP V, pp. 107-18)
- - , 19920, "Pali Lexicographical Studies X", JPTS XVII,
pp. 215-18 (= CP V, pp. 80-83)
- - , l 993A, Collected Papers, Vol. IV, PTS Oxford
- - , 1993B, "Pali Lexicographical Studies XI", JPTS XV!Il,
pp. 149-04 (= CP V, pp. 84-99)
- - , 1993C, "Exemal sandhi in Pali'', JPTS XIX, pp. 203-13 (=
CP V, pp. 168-79)
 Bibliography x Ill

- - , 1994A, "Mistaken ideas about nibbiina", Buddhist Forum Ill,

London, pp. 211-25 (= CP Vl,pp. 9-30)
- - , 19948, Collected Papers, Vol. V, PTS Oxford
- - , 1994C,. "Pali Lexicographical Studies XII", JPTS XX,
pp. 211-30 (= CP VI, pp. 47-{;7)
- - , 1996, Collected Papers, Vol. VI, PTS Oxford
- - . 1997, "Adopting the domestic way of life", in Bauddha-
vidyiisudhiikara~ (Studies in honour of Heinz Bechert on the
occasion of his 65th birthday), Gottingen
Pischel, R., 1900, Grammatik der Priif-it-Sprachen, Strassburg
- - , 1957, Comparative Grammar of the Prakrit Languages,
Benares (=English translation of 1900)
Radhakrishnan, S .. 1950, The Dhammapada, London
Rau, W., 1959, "Bennerkungen und nicht-buddhistische Sanskrit-
Parallelen zum Piili-Dhammapada", in Claus Vogel (ed.):
Jiiiinamuhiivali. Commemoration Volume in Honour of
Johannes Nobel on the Occasion of His ?Oth birthday offered
by Pupils and Colleagues, New Delhi, pp. 159-75
Rhys Davids, C.A.F., 193 I, Minor Anthologies, Vol.
(Dhammapada and Khuddakapii[ha), SBB, London
Roth, G., 1980, "The Patna Dhammapada", in The Language af the
Earliest Buddhist Tradition, Giittingen, pp. 93-135)
Sasaki, Genjun H., 1986, Linguistic approach to Buddhist thought,
Delhi
Schubring, W., 1910, Aciiriinga-siitra (Erstcr Srutaskandha),
Leipzig,
Sheth, H,D,T,. 1963, Paia-sadda-mahannam, second edition,
Bena res
Shukla, N.S. , I 979, The Buddhist Hybrid Sanskrit Dharmapada,
Patna
Smith, H., 1925, Dhammapadatthakathii. The commentary on the
Dhfimmapada, Vol. !, part I (second edition)
- - , 1928-{;6, Saddaniti, 5 volumes, Lund
- - , 1938, letter to J.Bloch, in Caillat, 1991, p. 2
Thieme, P., 1975, " Kranich und Reiher im Sanskrit", St/I I, pp. 3-
36
Turner, R.L., 193 I, Dictionary of the Nepali Lnni,n,age, London
XIV The Word of the Doctrine

- - , 1966, Comparative Dictiondry of the Inda-Aryan Languages,

London
Warder, A.K., 1963, Introduction to Pali, PTS London
- - , 1967, Pali Metre, PTS London
Watanabe, K_, 1995, "Avoiding all acts by both Buddha and
Mahavira", BE! 11-12 (1993-94), PP- 29--32
Whitney, W.D., 1889, Sanskrit Grammar, Cambridge, Mass_
Wintemitz, M., 1933, A history of Indian literanue, Vol_ II
(Buddhist literature and Jaina literature), Calcutt;i
Wrigh~ J.C., 1995, review of Hinuber, O_ von and Nonnan, K.R.,
1994, BSOAS 58, 2, PP- 437-38
--., 1996, "Sitliiia, kathii and other current topics in Pali", BSOAS
59, I, PP- 44-62
Yamazaki, M., Ousaka, Y., & Miyao, M., 1995, Indexes lo the
Dhammapada, PTS Oxford
 The Word of the Doctrine

•
 ~ J e ~

~~iGigE
co::~
~~!.s~~g-;
~
~fu~~i*~mG26~~S~:?~~
1V ..,11;1

ruN ~s~:co ';ic' ~;c '1iil!'\?~eseiF1jid

*' I !

~~i fi.5~~~ ~$; G~!~ ~~ ~100Jm.!

~~ ~'~~~ c~~:hn~3t:0 2e ~~~ ~i
uun~ euruG '§ms1sui~ 1wU~l"its~ ttTib~'1Jtl q§wf
lilWE\tn: "1f:go)S•S.o3>11-
~ ~ ~-
S~ Ql1\lG1mfUb1n~'
... t.: :J ;'J
"ii
"

~a- Stlq'§fUtJsnmnUJGBJblrJ ~a tiSmej E1m~

I
-··
\lli1NllN\llS1illilllSS9lbmn\llffl\lffi9lbfUb31bn:1mi~
• - " I
- , ... i
tlnn
'
iii iii.? SbHNITTUl •s.1>
- <>
'"JI> lmlfiilnN
'
el? J, I
I
HMlUl lf<!l ~~ llmlltls llm9nlltls9n ilnNtlmrud
lflln~'!Plfl•UlOJ~l!ls•igmlii~iismsmoi - " - • • • ' " I
' •
on~l!'llllllltqrntiliiuii <i's c.;:,;; tlirn: ~\lb \lb~SUlfULhlITTS\lb el? l511>rnn1 C:~l ~s !tij
u11~2onoim n: '.fo'O'.f-G<:e. stl 1!1'"\i(Lu !f.ei librl)l~ITTbr.J\)lbtll9l9Jcqbrui1J11l~ 1im1:B~
~1n1mt111 oE1rn:HnJ!N1L~i§lfiiTI!:n~ ffitl fi~ti ~fl!! ~ Sb~ruuB ts':~ a9fiGij en3 ~ 8~
... """ "" 0 d "'" ... • ,i
BU'illnmITTt~uariri ~e ~~~~e sti~go.nfUBiil.1B~mi,
• o ••• ' , •• _I
"fimmiN thmruru9gu1rum8rm~jb\h~Lmmi "Ln•aj
1'e"B'isrn~ti9Ul~ifi
Ll •• ~
itlrumsiub9n1siiibUti:"turillclii
' ~
1 Ii

9try~oiibt{!S 1~9 1£Jcl fary)l: lji boo~

[\\:ilrllfi '"' 50S S~ q\:ilrtlITI 1'~ >3'!'

UUl"tIIGUJtillJ ~ri t.!'~t>e~ Nit3tlfflJGUt9WS1

Wl1uUr1ij1 :r-;i:Ji'3ii!'> UB9guiWg 911 nEtl&

qiwtiu~uuGYtnem9intytutNn~tl Sll ~ro
tttU]iMlMBl ~tllSJbtijll tsin~tri utiro~mltlt

Sl1tf']m1Bn1hnsun1tnUtln tsit~ Bl1isi~: .,

 INTRODUCTION

L THE DHAMMAPADA GENRE

§ 1. The Pali Dhammapada is perhaps the best known of all

Buddhist texts. We know, however, of other examples of this type of
collection or anthology, which we might describe as the Dhanmpada
genre of literature, belonging to other schools of Buddhism, and it is
likely that a text of this type was included in the canons of all the
Hinayana schools of Buddhism.

§ 2. Winternitz suggested (1933, II, pp. 83-84) that, since more than
a half of all the verses of the Dhammapada have also been traced in
other texts of the Pali canon, there was scarcely any doubt that,
generally speaking, the compiler of the anthology took them from the
setting in which we still find them today. He did, however, note that
the collection has come to include some sayings which were
originally not Buddhist at all, but rather drawn from that
inexhaustible source of Indian gnomic wisdom, from which they
also found their way into Manu's law book, into the Mahabharata,
the texts of the Jains, and into narrative works such as the
P~catantra. Since Wintemitz \vrote, investigations have suggested
that in many cases the Dhanunapada did not borrow from elsewhere
in the canon, but that in the canon as a whole borro\ving took place
from a store of verses which in all probability pre-dates the canon in
its present form. Although we talk about Dharrnapada literature, the
Pali parallels of verses in the other Dharmapada texts are sometimes
to be found, not in the Dhanunapada. but in the SuttanipatJ., or the
Samyutta-nik:.aya, or the Jataka, or occasionally in other canonical
lCX ts.

§ 3. We can be certain that all versions we have of the

Dh>trrnapada/Dhamrnapada are translations from earlier versions, all
going back ultimately to the basic store of Dharrnapada verses,
which Brough called "abody of floating verses". Even if we could
date the versions which we hve, we should be dating only the
translation from an earlier version. Many of the verses in the store
had no specific Buddhist flavour, which explains why parallel
xx The Word of 1he Doclrine

versions arc found in borh Jain and Brahmanical texts. The

Buddhists began to add to and draw upon this store at the time of the
Buddha, although it is very likely that additions from other sources
were also made to the corpus af\cr that time. If we look at any one of
triis group of texts we will find that each one of them has some
features which might reasonably be surmised to be old. and yet at the
same time each one has features which are manifestly incorrect or
late. ..

§ 4. The relationship between the various Buddhist Dharmapada

texts is very complicated.with patterns of equivalence between them
<.varying from verse to verse, and sometimes even fi.{,-n pada to pada.
The fact that any two or more of them agree in some feature tells us
only that in some way, in the history of the texts, they were
dependent upon a common source for that particular feature. The
number of verses each redactor selected, the number ofvargas into
which they were sorted and the way in which the verses were
apportioned to each varga, give us no information whatsoever about
the date at which each selection was made.

IL PARALLEL VERSIONS

§ 5. Beside the Theravadin version in P5li (OvH/KRN, 1994), we

have a version in the G5ndhari Pra.lait (Brough, 1962), which has on
good grounds been assigned to the Dharmaguptakas. There is a
version in a highly Sanskitiscd Prakrit, the so-called Patna
Dharmapada (Shukla, 1979;. Roth, 1980; Cone, 1989), which may
belong to the Mahasanghikas, and there are several closely related
versions of the Udana-varga in Buddhist Hybrid Sanskrit, probably
belonoina
0 0
to . the Sarvastiv5dins or l\1Ulasarv:istiv5dins. Bernhard
(1965) sometimes prints two versions side by side when they are
irreconcilable. If we compare Nakatani's edition (1987) with
Bernhard's we sometimes find evidence of yet another tradition.
Brough noted several places where the Tibetan translation of the
Udana-varga seems to follow yet another tradition.
 § 6. There are also portions of Dharmapada material in the
Mahavastu, which belongs to the Lokotiaravadin Mahasailghikas.
The Sahasra-varga is quoted by name (bhagaviil[I Dharmapade~u
Sa!zasravargarp bh~ati, Mvu III 434,12) at the beginning <if the
veiscs at Mvu III 434-36, and is probably complete. The versei; at
Mvn III 421-23 are not specifically ascribed to the Dharmapada but
from their cont~! they are almost certainly a Bh~u-varga, which is
possil>ly not complete. Individual stanzas are cited (Dharmapade or
Dhannapad~) at II 212,18, III 91,18 and 156,15.

'nL THE ARRANGEMENT OF THE VERSES

§ 7_ In all the versions of the Dharmapada genre, we find the same

pallttn of named vargas consisting, for the most part, of verses
appropriate to the title of each varga. Although they clearly go back
to a common body of verses, each text contains verses not found in
the other versions. These may be taken from the common fund, and
their absence may be due simply to editorial choice, or they may be
i~rted from a source known only to the individual editor or his
successor, or they may be the invention of the editor or of someone
!ala" in the train of transmisson_ It has been suggested, for example,
that 109 "was inserted in its position merely to provide an
introduction to 110-15. It is found in thaf position only in
Dhammapada. ,

§ ll Despite the resemblances, it is easy to show that not one of these

anihologies is directly dependent upon any other. Some verses in the -
Pama Dharmapada are so similar. in form to verses in the .
Dhammapada that it is tempting to think that the former is simply a
more Sanskritised version of the latter, but the nature of the
language, and the number of the vargas and the distribution of the
vexses in the vargas shows clearly that this is not so. When we
__ .compJe theDharmapadas belongingiolhe.various traditions, we
fuid that no two versions agree for more thana few verses, at most,
at a time, although clearly all access to the basic
store of material.
XXll 71ie Word of the Doctrine

§ 9. n,ere is no total agreement about varga nomes. There appears to

be no consistency in the ordering of the vargas, or of the verses in
the vargas. Brough suggested that there was some significance in the
fact that the Brahmana-varga is the fim in the Giindhari
Dhannapada, but the last (of26 vaggas) in die Dharnmapada and (of
33) in the Udanavarga. He m~de his .suggestion before the
publication of the Patna Dharmapada, where it is third (of22), which
would seem to indicate that the section had no particular significance
for the composer of that collectiori..

§ 10. There are 423 verses in the Dbarnmapada, arranged in 26

vaggas, according to their subject matter, although a verse inay
sometimes seem appropriate to more than one vaiga.., so that it
appears twice. So &O appears in the Pai:idita-vagga, but reappears as
145 in the Dai:i<)a-vagga. In the Puppha-vagga (44-)9) each verse
contains either the word puppha or the name of a specific flower,
except for 50 and 57, as Rau points out Sometimes it is possible to
see why such anomalous insertions have occurred. We can surmise
that 57 was inserted because sampannasiliinOl?J seemed appropriate
in view of silavataT?J gandho in 56, in just the same way as 58 and
59 make a pair, with yathii sahkiiradhiinasmil!' in 58 parallel to
eval[l saizkiirabhutesu in 59.

§ 1 I. Similarly, every verse in the Brahma0 a-vagga (383-423)

contains- the word briihmana (very frequently in the refrain tam
ahal[l bnlmi briihmanam), except for 394. This undoubtedly was
inserted af\er 393 because, like 393, it contained the wordjatiihi
and in its rejection of non-Buddhist ascetic practices it seemed to
echo the condemnation of the idea that a briihmaQa was such
because of the caste he was born into.

§ 12. In the same way, in the Naga-vagga (320-33), we find not

necessarily the word riaga, but nevertheless some other word for
"elephant" in every verse, except 321 323. 325 328 .331 332
333. We can see that sometimes a verse is included because it makes
a pair with another verse which is appropriate, e.g. 321 and 323 are
included because of the word danta which they have in common
with 322. There seems no reason for 325, as Rau points out, except
 Introduction xx iii

for the fact that it is concerned with a large animal. It is interesting to

note the relationship between 32 8-33, and the train of associations
which explains to some extent why they are in the Naga-vagga.
There is no mention of niiga in 328, but it forms a pair with 329,
where naga is mentioned: Pada d of 330 matches pada d of 329. Jn
328-30 there are references to sahtiya(ttl), which leads on to 331,
which includes sahiiya, but not niiga. It also includes a reference to
suklia, which leads to 332 and 333, neither of which refer to niiga.

IV. THE COMMENTARY

§ 13. There is a corrunentary, which is.said to have been written at

the request of a thera named Kumiirakassapa, and to be based upon
corrunentarial material in the language of the island. At the end of the
work it is said that it was composed by Buddhaghosa while-residing
in a residence built by king Sirik:Uia Doubt has been cast upon this
ascription because the style of the commentary is so different from
Buddhaghosa's other cties, but it is possible that Buddhaghosa
collected together a number of traditional stories and his role was
raL'1er that of an editor.

§ 14. Each verse, or group of verses, is ascribed to the Buddha, and

the commentary includes a narrative portion explaining the
circumstances in which the verses \Vere uttered. The information
given in the story is frequently helpful in understanding the verses.
Sometimes the story is based upon an event which is mentioned
elsewhere in the canon, e.g. the story which accompanies 266,
although some additional details are given. The source of many of
the stories is unknown, and we cannot tell if they represent the
genuine background to the utterance, or whether they have been
manufactured at a later date to suit the circumstances. A..11 the stories
end wtth the Buddha uttering the· verse or verses and, since the
authorship of those verses which' are also found in non-Buddhist
sources must be uncertain, the cir~umstances narrated in the stories
must also be uncertain. As in the case ofthelataka stori~s. it would
appear that occasionally a verse has been married to a story because
of a supposed connection which may not be there. In other cases the
X..,"tJV The Word of the Doctrine

verse is of such general application that it could have been uttered in

any circumstances.

§ 15. There is also a grammatical section explaining the meaning of

the ~ds of the verse(s). I quote extensively from this porti.on of the
commentary, to explain why I tranSlate in the way I do, although it is
0

clear that the commentary or the' tradition it was following did not
understand the construction of some verses, e.g. 42, or >ome
grammatical forms, e.g. pahiitave 34, or the meaning of some
words, e.g. loka-vaefefhana 167, ma/ta 290, veyyaggha 295,
paribbiija 313.

§ ·16. Some readers of my earlier translations have regretted the fact

that I have also quoted copiously from the comn:ientaries there,
without giving any translation_ The need to do so in the case of the
commentarial portions of the Dhammapada-a[[hakathii has now
been met in an exemplary manner by C&P, whose work
complements the translation of the stories by Burlingame.

V. THE TRANSLATION

§ 17. This translation is based upon OvHJKRN, except where

otherwise indicated, even where a comparison with other versions of
the text in Prakrit and Sanskrit suggests that the Pali tradition has
misunderstood or changed the text in some places, e.g. 26d 369b. In
such cases I have discussed the alternative versions in the notes, and
have tried to explain why the Pali tradition shows variations. Very
occasionally, however, where I think the Pa.Ji tradition is incorrect, I
have translated what I think is the correct reading, e.g. 259d. I have
discussed such points in the notes.

§ 18. The notes and the frequent quotations from the commentary
they contain are intended to be a guide to the way in which I arrived
at my translation. Where I differ from the commentary's explanation,
I give my reasons for doing so. I have made frequent references to
parallel versions where those texts help to throw light upon the
meaning ofDharnmapada. I have not tried to give a reconstruction of
 Introduction xxv

the "original" !onn of the Dhammapada - fascinating though that

would be.

VI. THE TITLE OF THE TRANSLATION

§ 19. I think it is important, in order to avoid confusing potential

readers, to have different titles for a text and for its translation. The
word dhammapada, like Sanskrit dharmapada (see BHSD, s.v.
dhannapada) is used in two ways: as the title of a text and in the
sense of "religious c.xtrine". I assume that the title has the same
meaning as the occurrences of the word in the text itself;
dhammapadarrr sudesitam 44-45 and ekam dhammapadam seyyo
102. Kalupahana entitles his version "A Path of Righteousness",
doubtless relying on 44-45 where he translates dhammapadarrr
sudesitam as "the well-taught path of righteousness", presumably
takingpada in the sense ofpatha, although in the notes (p. 164) he
includes a reference to "the well-taught verses of the doctrine", and
he translates ekam dhammapadam in 102 as "one statement of the
doctrine" (p. 12)

§ 20. If we accepted the meaning "path" for pada, then we could

translate "the path of the doctrine". If we prefer "place" for pada,
then we can translate "the place of the doctrine". I prefer the equation
with 44-45, and follow the translation I give there and in l 02. I
have therefore entitled the translation "The Word of the Doctrine"
(abbreviated as \VD).

VII. THE METRES OF DHAMMAPADA

§ 2L The following metres are found in Dhp:

Tri;!Jlj>h: 19 20.46 54 83 108 127 128 141142143B 151 177 221
306 309 325 328 329 331 345 346 347 353 354
Jagati: 144 326 338 '
Vaitiiliy~: 15 16 17 18 24 44 !\5 80 95 145 180 235 236 237 238
240 284 285 324 334 341 343 348 349 350 388
Aupacchandasaka: 184
XXVI The Word of the Doctrine

Anu,tUbh (Sloka ): the reinainder

Mixed metres:
Tri~fubh (listed padas)/Jogati: 40abc 84ab 94abc 125bcd 208ac
280abc 281 b 31 Ocd 390abd
Tr~fubh (listed padas)!An~fubh (Sloka) 2 ! 8c 330cd
Vaitaliya (listed padas)!Aupacchandasaka: ! 79acd 342abc 344bcd
3~2abc 371a .
Vaitaliza (listed piidas)!Anu,[ubh (Sloka): 69d 1 l9d 120d

§ 22. The following analyses are based upon the readings in

OvWKRN, and represent an attempt to define the metre of each
pa'da. An asterisk(*) signifies that an alternative reading is suggested
in the notes. Many padas can be scanned in two or mQre ways, but I
have listed such padas under one category only. Decisions have been
made quite subjectively, and there are certain to be mistakes,
particularly in the case of padas where emendations produce a
different metre. I have discussed these alternatives in the notes. I
have been quite arbitrary in determining the position of the caesura in
doubtful cases in TriHubh and Jagatf piidas. Resolved syllables are
ignored in these analyses, but lists of such syllables are added in the
case of Tr~[ubh, Jagatf and Anu,[ubh (Sloka) padas. The question
of svarabhakti vowe!S is discussed at the end of the introduction
(§ 40).

§ 23. Tri~[ubh padas:

(a) Openings:
19abd 20f 46bd 54abc 83bc 84ab 94c 125c 127abc
!28abc 14lac !42a !Slbc 177a 22\d 280l:'306acd
309c 325cd 328ad 345ab 347b 353d 390abd
l 9c 20abcde 40abc 46ac 54d 83a 94ab I OSabcd
125bd. 127d 128d 14\bd 142cd 143Bab !Siad
l 77bcd 208ac 22labc 280ac 306b 309abd 325ab
328bc 329abcd 330c 33Jabcd 345cd 346abcd
347acd 353abc 354bcd
[-J--·- 83d*
142b 330d*
 Introduction xxvii

218c' 28Ib'
354a

(b) Breaks:
19a 20a !25d 142b 309b 330c 346b
- -' - 19b 20c 40ab 46ac 94ac 108a 125b 142ac 151b
208a 221c 280a 306c 309a 325c 328c 33lad 345d
346c 347c 353bd
- -- 19c 20de 46b 54ac 83bc 84b 127abc 128abc l4lad
'
142d 151 c I 77c 306abd 309cd 329c 345c 390b
19d 20f
- --'
20b 84a IO& 328a 329a
40ab 54d 83a 94b !08d 128d 14lb 142d 15\ad
177bd 208c 218c 22\abd 280b 325abd 328bd 329b
331 be 345a 346ad 347b 353a 354bc 390ad
40c 125c
46d 83d 345b 347d• 354ad•
54b281b
108b
127d
-.[-]-- 14\c
143Ba
l43Bb 353c
-,[-]·- 177a
280c 347a
- [-] ' - - 329d 330d
345d

(c) Cadences:
all

(d) (i) Resolution. Long syllables are resolved in the following

piif's:- .

First syllable: 20e 40b 108d 125d 328c 346d 347d

(ii) Redundant syllables. There are redundant syllables in the

following padas:-
X.XVlll The Word of the Doctrine

Fir;t sylbble: 8Jd•

Fi f\h syllable: 14 l c l 77a 329d 330d

(iii) Replacement There is the replacement of short 6th and 7th

syllables by a long syllal>le in the following padas:~

l 9d 20f 40c 125c

§ 24. Jagati padas:

(a) Openings:
40d'84d 144b 20&\ 3 !0a 326ac 338a 390c
[-1--n-- 84c
94d 125a 144ac208b280d28la_cd 310b 326bd
338bc
144d• 338d*

(b) Breaks:
40d 208b 280d 281 a 338ab
84c 125a l44ad28ld310ab326a338d
84d
94d
144b 281c 390c
144c 208d 338c
326bc
326d

(c) Cadences:
144a•
the rest

(d) Redundant syllaples. There are redundant syllables in the

following padas:-

First syllable: 84c•

Third_syllable: 84c•
 Introduction XXIX

§ 25. Vaitiiliya padas:

(a) Openings:
(i) Prior padas:
!Sa !6a 17a 18a 80a 95c 145a 237a 240a 284c
334a 34Ja 342a 343a
15c 16c 24a 44c 45a 80c !45c 179c l 80c 235c
236c 238c 240c 285a 324c 34!c 342c 343c 344c
348c 362c 388c
l 7c• !&•
24c• 44a
45c•
95a• 324a•
l 79a 235a 348a 388a
J80a 236a 238a 285c 37la
237c•
284a•
334c•
349a•
349c
350a•
350c•
362a•
(ii) Posterior padas
- - - - - - !Sb• 17b* 24d* 237b* 341 b* 343d*
!5d 16d J7d 18d 80d 95d 145d 235d 236b 237d
238b 284d 285d 334b 34 ld 344d 348b 350d 388d
69d ll 9d J20d
- - - - - - 80b !45b 179d !80d 238d 240d 343b 350b
------ 95b*
- - - - - - - 235b 324d
236d·
- 240b
- 1 - - ---324b•
- - - - - - 349b -
349d·
362b•
xxx The Word of the Docrrine

(b) Cadences:
all
\
' Syncopation.
(q. S There are syncopated openings in the following
. 1iadas: .
l5b* 17b* l79b !80a 236a 233a 285c 349a*d* 350a•c•
37la

§ 26. Aupacchandasaka padas:

(a) Openings:
(i) Prior Vidas
l 84a 344a 3 71 c
184c
(ii) Posterior padas
179b
l84b 362d 37lb
184d
342d
37ld*

(b) Cadences
all ,

(c) Syncopation. There is a syncopated opening in the following

pada:
l 79b

§ 27. Anusnibh (Siok.a) podas:

(a) Openings: Short second and third syllables are found in the
follo\ving p3das: ..
13b* l4b' 148c* 363a420a

(b) Cadences.
(i) Prior: there is the cadence· - ·with no caesura after the
fifth syllable in 7c
(ii) Posterior: There is the cadence···· in 92c.
 Introduction XXX!

There is the prior cadence - - - - in I 82d'

(c) (i) Resolution. There are resolved syllabics in the following

pad as:
First syllable: 8a 5!d 52d 73a 99a 126d 153c 183b l85a
227e 27 lc 302f307d 3 79b 411 c
Third syllable: 210c•
Founh syllable: 8e 9b l4bd 27a I 31 d 228a 231 d 23 2d 233J
24&a 302[ 333d 364d 4 l 4a
Sixth syllable: 7a 8a 47a 48a 181 c l 82a l 83a l 85a 274c 275a
302a 333c 389a418a
Seventh syllable: 246c 387c 409c
Doubtful: 2la
(ii) Redundant syllables. There are redundant syllables in the
following padas:
Nine syllable padas: J26a• 137c* J54c• 162b 210c• 222d
274c (after resolution assumed) 277a 291 b•
Ten syllable pada: 274c
Thirteen syllable piida: 352f
(iii) Omitted syllables. The following padas have less than
eight syllables:
Seven syllable padas: 3c• 260a• 26 l d' 266a*

A number of padas listed as showing resolution of syllabics

other than the first or sixth are probably to be regarded as hyper-
metric. Some of these hyper-metric padas result from faulty readings.
Others can be normalised by contraction or by elision. Sorr.e seem to
be genuinely hyper-metric.

VIII. METRlCAL LICENCE

§ 28' Mai;y verses scan only because of cer8in changes which have
been introduced into them metri causa. Such changes usually
·involve the writing of a long syllable for a short, or vice versa, but
other forms of metrical licence are also employed. More verses could
be cegularised if the necessary changes were made m.c. The
following lists are examples of the changes which occur in
 XXX!l The JVord of the Doctrine

OvH/KRN, and suggest places where similar chongcs could be made

to improve the metre.

§, 29. The unhistoric doubling of consonants. Examples of

thls are found in the following padas:

(a) Vaitiiliya!Aupacchandasaka padas:

suggalim l 8d

(b) Anuetubh (Sloka) padas:

-pphalaT[I 66d
upassaggaT[I J 39a
anibbisaT[I 153b
suggaliT[I 3 l 9d

§ 30. The restoration of doubled consonants rn initial

position. Examples are found in the following padas:

(a) Vaitiiliya!Aupacchandasaka padas:

ppacessati 44d 45<l
cchecchati 350d

§31. The lengthening of vowels.· The metre is frequently

corrected by lengthening a vowel which is normally short. Examples
'are fauna in the following padas:

(a) T~tubh/Jagati padas:

vijjati l 27c 128c
iipajjati309b
tatiyam 309d (?)
gati310a
· rati310b
satimii 328d
aniipalitto 353b
-rati354c

(b) Vaitiiliya!Aupacchandasaka padas:

satimato 24a
 Introduction xxxiii
khan ti l 84a
palavati 334c
bhiivayati 350b

(c) Anustubh (Sloka) padas:

bhiisati l d 2d
-ppasahati &
muni 49d 268a 269b
maiiiiati 63a 69a
paccari 69c l I 9c I 20c
jutimanto 89e
satimanto 9la
ramati 99b I l 6d
passati l 19a l20a
aggi140b
hiri- 143Aa
kasiim I 43Ad
icchati 162d
satimalfll!l l81d
jiiyari 193c 212ab 2 l3ab 214ab 215ab 2 l 6ab 283b
ajjataniim 227b
ahin7cena 244a
hirimatii 245a
nayati257b
nibbindati 277c 278c 279c
sahati 335a 336a
charti1{1Sati339a
-siiriro 352e
rati 373c
labhari374c
khanti- 399c
-sGrira0 400c
nandi-413c
kiimii-415c
niriipadhiJ!i 4 I Sb ·

§ 32. The metre can sometimes be corrected by lengthening vowels:

XXXlV Tne Word of the Doctrine

(a) Trisrubh/Jagatipadas:
-he!U 84a
viriyena ·l 44a
pahascathii l44d
ca 218c 28\b
nibbartarf 338d

(b) Vairaliya!Aupacchandasaka piidas:

ariya- 236d
palavan-334c

(c) AntLS[Ubh (Sloka) padas:

VU{{hi \3b \4b
bhijjarf l48c
jayari 282a

§ 33. The simplification of consonant gronps. Examples of

this are found in the followingpadas:

(a) Tris[Ubh/Jagarfpadas:
dukhena 83c
c' for cc' 306b

(b) Anustubh (Sloka) padas:

dukha 186c 203b 302b

The metre is improved if groups are simplified in the following

p:idas:

(a) Tristubh!Jagatipad:as:
appossu[k]ko 330d
-du[k]kha'!I 347d 354d
-[k]khaye 353c

(b) Vaitaliya!Aupacchandasaka padas:

-sa[n]natassa 24c
a[p]pamattassa 24d
-sa[n]nato 362a
 ln1roduc1ion xx xv

sa[tijiia1o J62b
du[kjklwn 371 d

§ 34. Shortening of-..iwels. The metre is frequently corrected by

shortening a vowel 'l&iich is naturally long. Examples of this are
found in the followingpiidas:

(a) Trfytubh!JOf}l!ipiidas:
va foooi 14 lc
(b) An!J1fllbh {Sloka) piidas:
· va for,..13&d 139abcd 195b 409a
pani""'!' 135d
afta~355d

The metre can be conmed by shortening vowels:

(a) Vaitaliyalhpacchandasaka padas:

me J7c l8c
-sam395a
-khiliipllmi! 95 b
vii.so me
1e 237c
-palali324a
-pabhdano 324b
somanasslini 34 I b
kho 34'd
-upasame 350a
-saiiiilitii 362a
saiiiia» 362b
kandi311d
jb) An1J1;ubh{!Toka) padas:
uppadol82d

§ 35. The shortening of nasalised vowels. In a number of words a

nasalised vowel is to lie scanned as short. This is shown in the text
by the omission ofanasvara, although this is, of course, no guide to
the actual pronunciation of a short nasalised vowel.
xx.xvi The Word of the Doctrine

(i) examples of the loss of final -m:

\
(a) Vaita/iya!Aupacchandasaka padas:
akal[lkha 343d
vanasml 334d

(b) Anw;!LJbh (Sloka) padas:

pa[ivadeyyu !33b
phuseyyu 133d
lokasmi 143Ab
maccana 182b
buddhana 183d 185d
addhana 207b
appasmi 224b
deviina 224d

(ii) The metre is improved if-'!' is omitted in:

(a) Vaitiilrya/Aupacchandasaka padas:

yiiva['!'] 284a

(b) Anw;tubh (Sloka) padas:

yesa['!'] 92c

(iii) The-metre is improved if -m- is orrjttcd in:

(a) Vaitiiliya!Aupacchandasaka padas:

sa[m]payiito 237b

(iv) Sometimes a vowel is shortened by changing-'!' to -m before a

VO\Vel:

(a) Anw;!LJbh (Sloka) padas:

pamiidam 27a

(v) The metre is improved if -m is changed to -m before a vowel:

 Introduction xxxvii

(a) Anusrubh (Sloka) padas:

buddhiinal[I (>-am) 194a
piyiinal[I (>-am) ~!Oc

§ 36. Removal of syllables. Hyper-metric padas can sometimes

be improved by the re!TIQva] ofa syllable:

(a) T~rubh!Jagatipiidas:
n[a] 83d 84c
icche[>YaJ84c

(b) Vaitiiliya/Aupacchandasaka padas:

-[pa]mathitassa 349a
[mii] 37ld

§ 37. Exceptions to the two-morae rule. In a number of places a long

vowel is found before a double consonant or consonant group, in a
way which goes against the normal pattern of Middle Indo-Aryan.
They are possibly the result ofSanskritisation, e.g.

niigghati 70d
bhasmiicchanno 7ld
miippa- 12la 122a
niissa l 24a 389b
niibbw:iam I 24c
-dussilyam l 62a
niiiiiio l 65f
iirogya 204a
pitvii 205a
piyiippiya0 211 d
viikkarar;a- 262a
(a)tiviikya'!' 320c 32ld
i
(yathiiUanii 323c)__
niiiiiiesam.. 365b .·
XX..XVllt The Word of the Doctrine

IX. ORTHOGRAPHY AND PHONOLOGY

§ 38. Consonant groups not making position· in initial

position. In general the language ofDhp follcws the usual rule that
a naurally short vowel is scanned' asJong if it is followed by a
consonant group, but there are exceptions to this, where certain
groups seem not to make pcsition; i.e. a short vowel before them is
still to be scanned as short.

(a) br : in positions where we should expect a short vowel

before br, the position is:
Brahmunii does not make pcsitiorl in 105b
briihmana does not in 294d 295d 383bd 384b 385d 386d
3S7d 38Sa 389b 39ld 392d 393bd 396f 397d-4 I 3d
4l 4f 4 l 5d-422d 423f
brilhaya does not in 285c (nor does anubriihaye in 75f)

(b) The other occurrences of conjunct consonants after short

vowels are in positions where the metre is not f1-.:ed.

§ 39. Consonant groups. A number of consonant groups occur

which go against the normal pattern of Middle Indo-Aryan. They
possibly represent the result ofSanskritisation, e.g.

am·eti I e 2e 7 lc l 24c
")'0.satfa- 47b 48b
bolyam 63a
bhasmiicchanno 7ld
kismici 74d
-klesehi 88d
->}'Oyam I !3bd 374b
bhadra l !9a 120abcd l43Ad !43Ba
0

-asmi(m) (loc. sg.) 143Ab 22lc 224b 247c 367a 395c

-dussilyal!J l 62a
asmil!J 168d I 69d 242d 41 Ob
pathavyii !78a
-cetyiini l 88c
iirogya- 204a
 Introduction xxxix
rasmi- 222d
byiisalla- 287b
(a)tiviikyarp 320c 321 d
vyanti- 350c
asmi 353a
tasmii 357c 380c
kalyii!Je 375f

§ 40. Svarabhakti vowels. In many words a vowel which can be

shown on historic grounds to be epenthetic, evolved to resolve a
consonant group, must be disregarded for the purpose of scansion.
In most cases this probably results from the fact that the epenthetic
vowel had not yet been developed at the time the verse was
composed. Jn the case of sul.Jiumo (J 25d) the phenomenon is
probably that of the resolution of a long syllable rather than the
disregard of a svarabhakti vowel, since resolution at the beginning
of a Tristubh pada is commoIL If a svarabhakti vowel occurs in a
Vaitiiliya or Aupacchandasaka p3da in a position where either· or - -
is possible, it is impossible to decide whether the vowel should be
disregarded or not

(a) Triefubh padas:

-cariyii 14la
lwdariyii I 77a
kayirii 281 b 330c

(b) Anustubh (S!oka) padas:

arahati 9d I Od 230b
ariya22d 79c 164b !90c 19lc 206a
kayiriitha 25c l l 7c l 18b 211 a
garahita 30d
kayirii 42a 43a 53b61c 105c l 17ab I l 8a I 59a
• -cariya 61 c
payirupiisati 64b 6Sb
issariya 73c ·
rahada 82a
pariyodapeyya 88c
arahanto 98c
 771e Word of1he Doctrine

viriya ! l 2d
vajira !6ld
pariyodapana ! 83c
kadariya 223 c
kayirati 292b
.. kayira 313a

The metre can be improved by inserting a svarabhakti wwel:

(b) Vai1ii.liya!Aupacchandasaka padas:

viyanli- 350c

•
 THE WORD OF THE DOCTRJNE

I. THE PAJRS (Yamaka)

1. Mental phenomena are preceded by mind, have mind as their

leader, are made by mind. If one acts or speaks with an evil mind,
from that sorrow follows him, as the wheel follows the foot of the
ox.

2. Mental phenomena are preceded by mind, have mind as their

leader, are made by mind. If one acts or speaks ·ith a pure mind,
from that happiness follows him, like a_ shadow not going a\vay.

3. "He abused me, he struck me, he overcame me, he robbed me". Of

those who wrap themselves up in it hatred is not quenched.

4. "He abused me, he struck me, he overcame me, he robbed me". Of

those who do not wrap themselves up in it hatred is quenched.

5. For not by hatred are hatreds ever quenched here, but they are
quenched by non-hatred. This is the ancient rule.

6. Others too do not know that we should restrain ourselves here,

but those who have knowledge in this respect, thereby their quarrels
CC2Se.

7. Contemplating pleasant things, living uncontrolled in his senses,

and not kno\ving moderation in eating, slack, lacking in energ}'. him
indeed M5.ra overcomes, as the \Vind-overcomes a \Veak. tree.

8. Contemplating unpleasant things, living well controlled in his

senses, and kno\ving moder.::nion in eating, having faith. putting forth
energy_. him indeed Mara docs not overcome, as the wind does not
overcome a rocky moun~in.

9. The impure person who will.put on a yellow robe, without sclf-

control and truth, he is not wor1hy of the yellow robe.
2 T1ie Word of the Doctrine

10. llut whoever would be devoid of impurity, well concentrated in

virtues, possessed of self-control and truth, he indeed is worthy of
the yellow robe.

I I. Those who think there is essence in non-essence, and see non-

essence in essence, they do not· attain the essence, having false
thoughts as their realm.

12. But those whu kno\v essence as essence, and non-essence as

non-essence, they anain the essence, having right thoughts as their
realm.

13. ·just as rain penetrates a badly thatched house, so passion

penetrates an undeveloped mind_
~ ~

14. Just as rain does not penetrate a well thatched house, so passion
does not penetrate a well developed mind.

15. Here he grieves, having passed away he grieves; the evil-doer

grieves in both places. He grieves, he is tormented, seeing his own
defiled action.

16. Here he rejoices, having passed away he rejoices; the one who
)las done merit rejoices in both places. He rejoices, he exults seeing
ihe purity" of his own action.

17. Here he suffers, having passed away he suffers; the evil-doer

suffers in both places. He suffers, (thinking) "I have done evil"; he
suffers all the more gone to a bad rebirLh.

18. Here he delights, having passed away he deligh<s; the one who
has done merit rejoices iq both places. He delights (thinking) "I have
done merit"; he rejoices all the more gone to a good rebirth.

19. Although reciting a large number of scriptural texts, if being

careless he does not act accordingly, like a cowherd counting the
cows of others, he has no share in the ascetic's life.
 I. The Pairs (Yamaka) 3

20. Although reciting a small number of scriptural texts, if he octs

righteously in accordance with the doctrine, abandoning passion and
hatred and delusion, knowing rightly, with mind well released, rot
grasping in this world or the next, he has a share in·the ascetic's life .

•
 II. CAREFULNESS (Appamidi)

21. Carefulness is the place of the death-free; carelessness is the

place of death. The careful do not die; the careless are a-,, though
(already) dead.

22. Learned men, knowing this especially in respect of carefulness,

rejoice in carefulness, delighting in the realm of the noble ones.

23. Meditating, persevering, constantly making a firm effort, those

~wise ones attain nibbina, suprc1nc rest from exertion.
"
24. Of one who exens himself. 1s mindful, docs p~re deeds, acts
considerately, is restrained, Jives according to the J~nv. is careful, the
fame increases.

25. By exertion, by carefulness. by restraint and self-control, a wise

man would make an island, which a Ooo<l does not overwhelm.

26. Fools, stupid people, apply themselves to carelessness; but a

\vise man guards his carefulness J.S his best treasure.

, 27. You should not apply yourselves to carelessness, nor to

·acquaintance with delight in sensual p!e~1sures. For being careful
{and) meditating one obtains great hJpplncss.

28. \Vhen the learned m3.n thrusts J\vJy c.Jrclcssncss by carefulness,

climbing on to the palace of kno\vledgc, he gazes gricfless on the
grieving people. The \vise man gJzcs do'.>:n on fools as one sranding
on a mountain gazes do\vn on those st~nding on the ground (bclo\v)_

29. Careful among the careless, \vide a\.vll...l.:e Jrnong those asleep, the
\vise man ·goes leaving (rhCm) behind, -as a S\vifr horse: leaves behind
a weak horse.

30. By carefulness Maghava (Indra) went to supremacy among the

gods. They praise carefulness. ·carelessness is always blamed.
 IL Carefulness (Appamada) 5

31. A bhikkhu who delights in carefulness, who sees danger in

carelessness, goes about like a fire consuming his fetter, small or
large.

32. A bhikkhu who delights in carefulness, who sees danger in

carelessness, cannot fall away but is indeed n= to nibbana .

•
 III. TI!OUGHT (Citta)

33. The wise man makes straight his trembling. uusfeady thought
which is difficult to guard and difficult to restrain, just as a fletcher
makes an arrow straight.

34. Like a fish taken from its watery home and thrown on the
ground, this thought quivers all over in order to escape the dominion
of Mara

35. Good is the taming of thought vihich is hard to grasp, light,

alighting where it will. Tamed thought is the bringer ofhappin=.

36. A wise man should guard his thought, which is difficult to see,
extremely subtle, alighting where it will. Guarded thought is the
bringer of happiness.

37. Those who will restrain their thought, which travels far, alone,
incorporeal, lying in the cave (of the heart), will be freed from
Mara's fetter.

38. The wisdom of one whose thought is unsteady, who does not
.,know the true doctrine, whose serenity of mind is troubled, is not
"perfected

39. There is no fcor for one whose thought is untroubled (by faults),
whose thought is unagitated, who is freed from good and evil, who
is awake.

40. Knowing that this body is (fragile) like ajar, making this thought
(firm) like a city, one sh9uld fight Mara with the weapon of wisdom,
Olle should gu2.r<l \vh2t has been conquered; one should take no

Before long, alas, this body will lie on the earth, rejected, without
/5:on!SCl.ousm:ss, like a useless log of wood.
 Ill. Thought (Ci11a) 7

42. Whatever an enemy may do to an enemy or, again, whatever a

hater may do to a hater, a wrongly directed mind would do worse
than that to him.

43. A mother or father or any other relative would not do that

(good); a rightly directed mind would do better than that to ~im.
 IV. FLOWERS (Puppha)

44. Who will conquer this earth and this world of Yama with its
gods? Who will gather the well-taught word of the doctrine, as a
skilful man gathers a flower?

45_ The learner will conquer the earth and this wodd ofYama with
its gods_ The learner will gather the well-taught word of the doctrine,
as a ski! ful man gathers a flower.

46. Knowing that this body is like foam, knowing that it has the
nalllre of a mirage, cutting off the flowery blossoms of Mara, one
should go where the king of death cannot see him_

47. Death canies away a man even as he is gathering flowers, while

his mind is distracted, just as a great flood carries away a sleeping
village.

48_ Death overpo\vers a man even as he is gathering Oo\vers, \vhile

his mind is distracted, not having had his fill of sensual pleasures.

49. As a bee takes nectar and goes away without hurting the colour
or the.smell ofa flower, so should a sage wander in a village.

50. One-should not have regard for the bad deeds of others, nor the
things done and lei\ undone by others, but only for the things done
and left undone by oneself_

51. Like a beautiful flower, full of colour but without scent, so is a

\vell-spoken utterance fruitless for one \Vho docs not Jct.

52. Like a beautiful flower, full of colour and scent, so is a well-

- spoken utterance fruitful for one who acts .

. 53. Just as one might make many kinds of garlands from a heap of
flowers, so much good is to be done by i mortal when born.
 IV. Flowers (Puppha) 9

54. The smell of flowers does not go against the wind, nor (the smell
of) jasmine nor incense nor sandalwood, but the smell of the good
d~ g~~ag~i:s.t the wind. A good man perfumes all directions.

55. Sandalwood or incense, lotus or jasmine, among these k.i nds of

perfimle, the perfum.e of virtue is supreme.
56. Of limited extent is this perfume, which is (of) incense and
sandalwood; but the perfume of virtuous men blows among the gods
as best

57. Of those who are possessed of virtue, live without carelessness,

and are freed by proper knowledge, Mara does not find the way.

58. Just as on a· heap of rubbish thrown on the high road, a lotus

might grow there with sweet smell, delighting the mind,

59. similarly among common people who are blind and like rubbish,
the disciple of the fully awakened one shines forth by reason of his
wisdom.
 V. THE FOOL (Bala)

60. Long is the night for one who is awake. Long is a yojana for one
who is weary. Long is journeying-on for fools who do not know the
tfue doctrin~.

61. If when wandering one does not meet one's better or one's
equal, one should certainly make a solitary journey. There is no
companionship with fools.

62. (Thinldng) "! have sons, I have wealth • the fool is tormented.
He bas indeed no self of his own, how much Jess sons? How much
less wealth ?

63. Whatever fool realises his folly, he is on that account a learned

man indeed. But a fool thinking he is learned, he indeed is called
.. fool".

64. Even if a fool associates with a learned man all his life, he docs
not know the doctrine, just as a spoon docs not know the taste of
soup.

65. If a wise ma..'1 associ2.tes with a learneJ man even for a moment,
be quickly knows the doctrine, just as the tongue knows the taste or
soup.

66. Stupid fools wander about with a self which is like an enemy,
doing evil action which has bitter fruit.

67. That deed \vhen done is not good \vhich one regrets \Vhcn onl:
has done ir> the resulr of\vhich one experiences \vceping \vith tearful
face.

68. But that deed when done is good which one does not regret
. when one has done it, the result of which one experiences pleased
and happy.
 V. The Fool (Bii/a) 11

69. A fool thinks it is like honey, as long as evil docs not mature.
But when evil matures, then the fool goes to misery.

70. A fool might eat his food month by month with the tip ofkusa
grass; he is not worth one sixteenth part of those who have perfected
the doctrine.

71. For an evil deed when done does·not curdle the same day, as
milk does. Burning like fire covered in ashes, it follows the fool.

72. A reputation for skill arises for a fool merely to his disadvantage;
it destroys the fool's good share (of merit), splitting his head.

73. He might wish for respect from bad people, and pre-eminence
among bhikkhus, and authority in dwelling places, and honours
among the families of others.

74. "Let both householders and wanderers know that this was done
by me. Let them be under my control alone in anything concerning
what is to be done and not done." Such is the fool's thought. His
desire and his pride increase. ·

75. There is one means for getting gai;, another means for going to
nibbiina. Thus having learned this, let the bhikkhu, the follower of
the Buddha, not rejoice in honour. Let him practise scciusion_
 VL THE LEARNED MAN (Pal)<)ita)

76. If one should see one who sees faults and speaks rebukingly,
wise, one should follow such a learned man as one would the
reporter of treasures. For one following such a one it becomes better,
not worse.

77. He would advise. he would admonish, he would restrain from

the impure. He becomes beloved of the good; he becomes unbcloved
.of the bad.
I
78. One should not associate with evil friends; one should not
associate with the lowest of men. One should associate with good
friends; one should associate with the best of men.
to J.. f

79. He who drinks of the doctrine sleeps happily, with a clear mind.
The learned man always rejoices in 1he doctrine taught by 1he noble
ones.

80. Truly canal-makers lead water; lletchers bend the arrow

(straight); carpenters bend wood; learned men tame the scl(

81. As a solid rock is not moved by the wind, so learned men are not
=oved amid praise and blame.

82. Just as a deep pool is calm and clear, so, hearing' the teachings,
learned men are calm.

83. Good men indeed go everywhere. The good do not boast from
desire for sensual pleasures. Touched by happiness or misery,
learned men do not show variation.

84. \Vho neither for his own sake nor for 1h_e_sake of another would
wish for a son, or wealth, or a kingdom, nor would wish prosperity
for himself by unla,vful means, he would be virtuous, wise and
lawful.
 VI. The Learned Man (PaQdi1a) 13

85. Few are those persons among men who go to the far shore. 13ur
those other people only run along the bank. .

86. But those people. who, when the doctrine has been properly
preached, follow the doctrine, will go to the far shore_ The dominion
of death is very hard to cross over.

87. Leaving the black(= bad) characteristics, a learned man should

develop the bright(= good) (characteristics), corning from his home
to the homeless state, in seclusion where it is hard to find enjoyment.

88. There he should seek enjoyment, leaving sensual pleasures,

having nothing. The learned man should cleanse himself from all
defilements of the mind.

89. Of whom the mind is properly developed in the elements of

enlightenment, who delight in the abandonment of attachment
without clinging, they with asavas extinguished, full of brightness,
have gained nibbana in the world.
 Vil. THE ARAHANT (Arahanta)

90. For one who has completed his journey, is without sorrow, is
freed in very way, has left behind every fetter, there is no suffering.

91. The mindful exert themselves. They do not delight in a house.

Like geese leaving a pool, they leave their various homes.

92. Of whom there is no accumulation, who have knowledge of(and

have renounced) food, whose realm is empty ar.d unconditioned
release, their going is hard to follow, like that of birds in the sky.

9f Whose asavas are destroyed, and who is not dependent upon

food, whose realm is empty and unconditioned release, his track is
hard to follow, like that of birds in the sky. · .

94. Whose senses have gone to rest, like hor~es well tamed by a
charioteer, who has abandoned pride and has no asavas, him, being
of such a kind, even the gods envy.

95. Like the earth he is not hostile, he is like a locking post,

venerable, with good vows, like a pool without mud. For one of
such.a kind t~ere are no journeyings-on_

· 96. Of him properly released by knowledge, calm, of such a kind,

the mind is calm, the voice is c2Jn1 Jnd also the deed.

97. The man who is without desire (without faith), knows the
uncreated (is ungrateful), cuts off rebirth (is a housebreaker), who_
has got rid of occasions (for qu<!rrc!s or rebirth) (has destroyed his
opportunity); has abandoned Jcsirc (is an cater of vomit, i.e. what
has been abandoned by .others), is the best person (is one of extreme
audacity).

98. Whether in a village or in a foresr, \Vhether on lo\V land or on

high ground, \Vherever arahar:ts live, thJ.t is a delightful place.
 VII. The Arahanr (Arohanro) 15

99. Forests are delightful. Where ordinary people do not delight,

those devoid of passion will delight. They do not seek sensual
pleasures .

•
 VIIL THOUSANDS (Sahassa)

l 00. If there were a thousand utterances made up of meaningless

\vords. better is one \Verd of meaning. which hearing Dnc becomes
calm.

l 0 l. If there were a thousand verses made up of meaningless words,

better is one '-vord of a verse. \vhich hearing one becomes calm.

l 02. And if anyone were to speak one hundred verses made up of

meai1ingless words, better is one word of the doctrine, which hearing
one ·becomes calm.

103. !fa man were to conquer in battle a thousand times a rhousand

men, but conquer Oile, himself, he indeed is rhe best conqueror in
battle.

104. The self as a conquest is indeed better than these other people
(as a conquest) for a man who has tamed himself, who constantly
wanders self-controlled.

l 05. Not a god nor a gandhabba, nor Mara with Bralun:i, could make
,conquest into defeat for a person of such a kind. ·

l 06. If anyone were to sacrifice with a thousand month by month for

a hundred years, but were to honour even for a moment someone
with developed self - that very homage would be better than
sacrificing for a hundred years.

I 07. And if any person were ro attend the fire for a hundred years in
the forest, but were to honour even for a moment someone \Vith
developed self- that very homage would be better than sacrificing
for a h_undred years.

I 08. If a man were to sacrifice in this world any sacrifice or offering

for a year, looking for merit, all that does not come to a quarter (of
the better offering). Respectful ·salutation to the righteous is a belier
thing.
 VIII. Thousands (Sahassa) 17

!09. To him who practises respectful salutation. constantly

respecting his elders, four things increase: - age, beauty, happiness,
strength.

110. And if anyone were to live for one hu_ndred years, wicked and
unconcentrated, living for one day would be better for him being
virtuous and meditacing.

111. And if anyone were to live for one hundred years, ignorant and
unconcentrated, ].,ing for one day would be better for him having
knowledge and meditating.

112. And if anyone were to live for one hundred years, lazy and
lacking in energy, living for one day would be better for him
strenuously putting forth energy.

113. And if anyone were to live one hundred years, not seeing
arising and passing away, living for one day would be better for him
seeing arising and passing away.

114. And if anyone were to live for one hundred years, not seeing
the place which is death-free, living for one day would be better for
him seeing the place which is death-free. ·

115. And if anyone were to live one hundred years, not seeing the
highest doctrine, living for one day would better for him seeing the
highest doctrine.
 IX. EVIL (Papa)

l 16_ One should make haste in doing good; one shou l<l restrain
one's thought from evil. For of one performing merit slowly tho
mind delights in evil.

l 17 _If a man were to do evil, he should not do it again an<l again_ He

should not set his heart on it The accumulation of misery is
sorrowful.

l 18. If a man were to do good, he should do it again and again_ He

should set his heart on it The accumulation of good is happy_

119. Even an evil person sees good as long as his evil does not
mature_ But when his evil matures the evil person sees evil.

120. Even a good person sees evil as long as his good does not
mature_ But when his goqd matures the good person secs good_

12 L One should not think little of evil, thinking "That will not come
to me"_ Even a waterpot is filled by the falling of drops of water. A
foolish man is filled with evil, even practising it little by little_

1'22_ One should not think little of merit, thinking "That will not
come to me". Even a waterpot is filled by the falling of drops of
water. A wise man is filled with merit, even practising it little by
Iink

123_ As a merchant with a small caravan and much wealth would

avoid a dangerous road, as one desirous of life \Vould avoid poison.
so one should avoid evils .
.
124. If there were no wound on one's hand; one could pick up
poison with one's hand. Poison does not enter one \vho has no
wound_ There is no evil for one who does not do (evil)_
 IX. Evil (Papa) 19

J 25. Whoever docs:barm to an innocent man, a pure man who is

without blemish, evil comes back to that very fool like fine dust
thrown against the wind.

J 26. ·some are rebom in a (human) womb; evil doers are reborn in
hell; those with a good rebirth go to heaven; those without asavas
gain nibbiina.

127. Not in the sky.not in the middle of the sea, not entering an
opening in the mounbins is there that place on earth where sta;oding
one might be freed furn evil "\Clion.

128. Not in the sky.not in the middle of the sea, not entering an
opening in the moll!l£Zins is there that place on earth where standing
death would not ovc:srome one.
 X. VIOLENCE (DaDc;la)

129. All tremble at violence; all fear death. Comparing (others) with
oneself one should not kill or cause to kill.

130. All tremble at violence; to all life is dear. Comparing (others)

with oneself, one should not kill or cause to kill.

131. Whoever injures with violence creatures desiring happiness,

seeking his own happiness he does not gain happiness when he has
passed away.

132. Whoever does not injure \vlth violence creatures desiring

happiness, seeking his own happiness he gains happiness when he
has passed a\~ay. ·

I 33. Do not say anyihing harsh to anyone. Those spoken to \l:ould

answer you back. For ;irrogant talk is painful. Retaliation(s) would
assail you.

134. If you do not make yourself utter a sound, like a broken gong,
you have attained nibb5na. 1\rrogance is not follnd i~ you .

... 135. Just as a cowherd \virh a stick drives the CO\VS to pasture, so old
: ::tge and death bring t~c l i fc of creatures to an end.

136. But a fool doiog evil deeds does not know (this). The stupid
man is burned by his own deeds, like one burned by fire.

137. \Vhoevcr docs h~irin \vith violence to non-violent innocent

people, goes very soon :ndecc.I to one of the ten states:

138. He may arrive at harsh suffering, Joss, or breaking of the body,

or severe illness, or arLxicty of mind,

139. or trouble from the king, or a dreadful accusation, or the loss of

relatives, or the destruction of \Veal th,
 X. Violence (DDl;ufa) 21

140. or purifying fire burns down his houses. After the break-up of
the body; the fool is reborn in hell.

141. Not nakedness, nor matted hair, nor mud, nor fasting, nor lying
on the ground, nor dust. and dirt, nor exertion in a squatting position
pu.rify a mortal who has not passed beyond doubt.

142. If although adorned (with fine clothes) he practices equanimity,

is calm, controlled, restrained, living the holy life, having laid aside
violence with regard to all living creatures, he is a brahman, an
ascetic, a bhikkhu.

143A. Is a man restrained by modesty found anywhere in the world,

who thinks little of censure, as a well-bred horse thinks little of the
whip?

l43B. Like a well-bred horse touched by the whip, be eager and

swift. .

144. By faith, and by virtuous conduct, and by energy, by

concentration, and by discernment of the doctrine, endowed with
knowledge and good conduct, mindful, you will abandon this not-
insignificant suffering. •

145. Truly canal-makers lead water; fletchers bend the arrow

(straight); carpenters bend wood; men of good vows tame the self.
 XI. OLD AGE (Jara)

146. What is this laughter, why is there joy when (the world) is
constantly burning? When bound by darkness do you not seek a
lamp?

147. See the painted ;mage, a heap of sores, a compounded body,

diseased, with many (bad) thoughts, of which there is no permanent
stability.

148. This body is worn out, a nest of diseases and very frail This
heap.of corruption will break to pieces, for life has death as its end.

149. What delight is there seeing these white bones, which are
scattered like gourds in autumn ?

150. A city is made of bones, with a plaster of flesh and blood,

wherein old age and death and pride and hypocrisy are deposited

151. The decorated chariots of kings grow old indeed, and the body
too comes to old age; but the doctrine of the good does not comes to
. old age. The good indeed teach it to the good.

t52. This man of little learning grows old like an ox. His flesh
increases; his knowledge does not increase.

153. I have run through th~ journeying-on of numerous births,

without respite, seeking the house-maker, birth again and again is
painful.

l 54. 0 house-maker, you are seen. You will not make the house
again. All these rafters arc! broken, the house-ridge is destroyed. The
mind, set on the destruction (of material things), has attained the
termination of the cravings. -- - -· . - - -

l 55. Those who have not lived the holy life, who have not acquired
wealth in youth, pine away like old herons in a lake without fish.
 XL Old Age (Jarii) 23

156. Those who have not lived the holy life, who have not acquired
wealth in youth, lie like (arrows) scattered from a bow, sighing after
things of the past
 XII. THE SELF (Atta)

157. If one knew the self to be dear, one would guard it well. A
learned man would stay awake for one of the three watches.

158. One should first establish oneself in what is proper, then one
should advise another. (Doing this) a lca.'Iled man would not be
defiled ·

159. If a man should so do to himself as he advises another, then

well tamed indeed he would tame, for the self they s·ay is hard to
tame.

160. The self indeed is the lord of self; who else indeed could be
lord? By the self indeed, when well tamed, one obtains a-lord who is
hard to obtain.

161. By the self alone is evil done; it is born of self. has its origin in
self. It crushes the stupid one as diamond crushes a jewel made of
stone.

162. He whose excessive bad conduct is spread out (over him), like
a miiluvii creeper spread out over a sal tree, does to himself just as
his enemy wishes to do to pjm_
'
163. Evil deeds, and those harmful to oneself, are easily done. What
indeed is both beneficial and good, that indeed is very difficult to do.

164. The foolish man who, following a wrong view, scoffs at the
teaching of the arahants, the noble ones who live in accordance with
the law, bears fruit to his own destruction, like the fruits of the
bamboo.

165. By the self alone is evil done; by the self i~ one defiled. By the
self is evil not done; by the self alone is one purified. Purity and
impurity concern the individual. One man may not purify another.
 XII. 1he Self(Atta).

166. One should never give up one's own interests for another's ~-
interests, even if great. Knowing one's own interests, one should be
intent upon one's own interests .

•
-' _, ----..._
.-~ ..
- -- ._;_
i: ! •

(-
 XIIL TIIB WORLD (Loka)

i 67. One should not follow an inferior practice. One should not
dwell with carelessness. One should not follow a wrong view. One
should not be worldly.

168. One should exert oneself; one should not be careless. Clnc
should practise the doctrine well. He who practises the doctrine
sleeps happily in this world and the next

I (<l. One should practise the doctrine well; one should not practise it
badly. He who practises the dottrine sleeps happily in this world and
thenht

170. One should see it as a bubble; one should see it as a mirage.

The king of death does not see one regarding the world in this way.

171_. Come, see this world like a king's painted chariot Where fools
sink down, there is no attachment for those who know.

172. But whoever formerly being careless afterwards is not careless,

he illuminates this world like the moon freed from a cloud.

173. Whose evil deed when done is covered up by a good one, he

illuminates this world like the moon released from a cloud.

174. This world has become blind; few see here. A few go to heaven
like a bird released from a net

175. The geese go on the path of the sun. They go through the sky
by their supernormal power. The wise go forth from the world,
having conquered Mara and
, his army.
176. There is no evil which cannot be done by a creature who has
transgressed the unique law, speaks falsely, has abandoned the other
world.
 LT.B.M.L•
XIII. The World (Loka) Library 27

177. The miserly ones certainly do not go to the world of the gods.
Fools indeed do not praise giving. But a wise man, rcJOicing rn
giving, for that very reason is happy in the next world.

J 78. The fruit of the attainment of the stream is better than sole
sovereignty on earth, or going to heaven, (or) lordship over the
whole world.
 cL I.1
·:• .. 1di.J

XIV. AWAKENED (Buddha)

179. Whose conquest is not lost, whose conquest does not go

anywhere in the world, him awakened, with endless realm, leaving
no track, by what track will you lead him?

180. Of whom there is no attachment with a net (or) craving to lead

him anywhere, him awakened, with endless realm, ·leaving no track,
by what track will you lead him ?

181. Even the gods envy those awakened and mindful ones who are
intent on meditation, wise, delighting in the peace of the absence of
desire.

182. It is difficult to obtain birth as a man; difficult is the life of

mortals; difficult is the hearing of the true doctrine; difficult is the
arising of the awakened ones.

183. The avoidance of all evil; the undertaking of good; the cleansing
of one's mind; this is the teaching of the awakened ones.

184. Patience (and) forbearance are the highest austerity. The

awakened ones say nibbana is the highest. One is certainly not a
wanderer if one injures others; one is not an ascetic if one hanns
another. -

185. Not abusing, not injuring, and restraint under the rules of
discipline, and knowing moderation in eating, and secluded lodgings,
and exei:tion in respect of higher thought, this is the teaching ofth.e
awakened ones.

186. No satisfaction is found in respect of sensual pleasures (even)

by a rain of gold pieces. Knowing that sensual pleasures are oflittle
--- .taste and painful, the learned one_

1_87. obtains no delight, even in the sensual pleasures .of the gods.
The follower of the fully awakened one delights in the destruction of
craving.
 XIV. Awakened (Buddha) 29

188. Men who are terrified by danger go to many a refuge, to

mountains, and woods, to parks, trees and shrines.

189. Tnat indeed is not a secure refuge, that is not the best refuge.
Having come to that refuge one is not released from all suffering.

190. But if any one goes to the Buddha, the Doctrine and the Order
as a refuge, he perceives with proper knowledge the four noble
truths:

191. Suffering, the arising of suffering, and the overcoming of

suffering, and the noble eight-fold path leading to the cessation of
suffering.

192. That indeed is a secure refuge, that is the best refuge. Having
come to that refuge one is released from all suffering.

193. A noble (thoroughbred) man is hard to find. He is not born

everywhere. That family in which that wise one is born prospers
happily.

194. Happy is the arising of awakened ones; happy is the teaching of

the good doctrine; happy is unity in the Order; happy is the austerity
of those who are united.

195. Of one honouring those who deserve honour, awakened ones

or their followers, those who have gone beyond the diversified
world, those who have passed beyond grief and affliction,

196. of one honouring them, of such a kind, who have gained

nibbiina. having fear of nothing, it is not possible for this merit - so
great- to be measured by anyone.
 XV. HAPPINESS (Sukha)

197. We live very happily indeed without hate among those who
hate; among men who hate we dwell without hate.

198. We live very happily indeed without disease among those who
are diseased; among men who are diseased We dwell without
disease.

199. We live. very happily indeed without care among those who are
full of care; among men who are full of care we dwell without c2re.

200.-We live very happily indeed, for whom there is nothing.

Feeding on joy we shall be like Abhassara gods.

201. Being victorious one produces enmity. The conquered one

sleeps unhappily. The one at peace, giving up victory and defeat,
sleeps happily.

202. There is no fire like passion; there is no ill like hatred; there is
no suffering like the constituents of personality; there is no
happiness greater than peace.

403. Greed is the worst of diseases; conditioned things are the worst
~£orrow; (for one) knowing this as it really is, nibbana is the best
happiness.

204. Health is the best possession; contentment is the best wealth;

confidence is the best relative; nibbana is the best happiness.

205. Having drunk the flavour of seclusion, and the flavour of peace,
one becomes free from.distress and free from evil, drinking the
flavour of joy in the doctrine.

206. Sight of the noble ones is good; living with (them) is always
happiness; because of not seeing fools one would constantly be
happy.
 XV. Happiness (Sukha) 31

207. For he who consorts with fools suffers for a long time.
Associating with fools is grievous, like associating always with an
enemy. But a wise man is good to associate with, like meeting with
reiatives.

208. Firm, and wise, and learned, much enduring, dutiful, noble,
him, such a one, a good man, wise, one should follow as the moon
follows the path of the constellations .

•
 XVI. PLEASANT (Piya)

209. Jo.lning ·oneself to distraction, not joining oneself to

concentration, abandoning one's goal and seizing what is p 1easant,
one envies the one who applies himself to the goaL

210. Do not at any time associate with pleasant.(or) unpleasant

things. Not seeing pleasant things is painful, and also seeing
unpleasant things.

· 211. Therefore one should not treat anything as pleasant; for the loss
of the pleasant is evil. There are no bonds for those for whom there
is neither pleasant nor unpleasanl

212. From what is pleasant comes grief, from what is ple'a.sant comes
fear. For one who is freed from what is pleasant there is no grief,
much less fear.

213. From love comes grief, from love comes fear. For one who is
freed from love there is no grief, much less fear.

214. From delight come_s grief, from delight comes fear. For one
who is freed from delight there is no grief, much less fear.

:iJ5. From sensual pleasure comes grief, from sensual pleasure

comes fear. For one who is freed from sensual pleasure there is no
grief, much less fear.

216. From craving comes grief, from craving comes fear. For one
who is freed from craving there is no grief, much less fear.

217. One who has virtue and insight, is righteous, speaking the truth,
doing his own tasks, him People hold dear.

218. One should have a desire for the undefined, and be suffused
with mind, and have one's thought not ti.ed to sensual pleasures.
(Such a one) is called an Up-streamer.
 XVI. Pleasant (Piya) 33

219. A man who has been abroad for a long time, (and) has come
back to safety from afar, him kinsmen, friends and companions
welcome when he has come back.

220. In the same way his merits also receive one who has performed
merits, gone from this world to the next world, as lcinsmen receive a
dear one who has come back.
 XVII. ANGER (Ko<lha)

221. One should abandon anger, one should give up pride; one
should pass beyond every attachment Sufferings do not befall one
who is not attached to name-and-form. possessing nothmg.

222. One who indeed could control his anger when it arises, like a
chariot gone astray, him I call a charioteer. The other people are
merely rein-holders.

223. One should conquer anger by non-anger; one should conquer

bad by good; one should conquer miserliness by giving, hnd one
speaking falsehood by truth.

224. One should speak what is pleasant; one should noi be angry; if
asked one should give, even if there is only a little. By these three
means one would go to the presence of the gods.

225. Sages who do no harm, constantly restrained in body, they go

to the unshakable place, where having gone they do not grieve.

226. Of those who are always awake, learning by day and night,
· intent upon nibbana, the asavas disappear.

227. Thi> is an old thing, Atula, this is not of today only. They blame
one sitting silent; they blame one speaking much; they even blame
one speaking little. There is no one unblarned in the world.

228. There was not, and there will not be, and there is not now, a
man who is wholly blamed, or wholly praised.

229. But the one whom the wise praise, having observed carefully
day by day, as being faultless in conduct, intelligent, concentrated on
wisdom and virtue,

230. like a ring of gold, who ought to blame him? Even the gods
praise him. He is praised even by Brahma.
 XVIL Anger~odha) 35

231. One should beware of bodily "'l&Cr. one should be restrained in

body. Having abandoned misconduct of body, one should practise
good conduct with the body.

232. One should beware of angcr:ia one's speech; one should be

restrained in speech, Having abandcmed misconduct of speech, one
should practise good conduct with spttch_

233. One should beware of angcrin one's mind; one should be

restrained in mind.. Having abarulmed misconduct of mind, one
should practise good conduct with lhemind.

234. The wise are restrained in bodJ;and restrained in speech, the

wise are restrained in mind. They ind!:cd are well restrained.
 XVIIL IMPURITY (Mala)

235. You are now like a yellow leaf, and Yama's men too stand near
you. You stand at the beginning of an unc1ertaking, and you do not
even have provisions for the journey.

236. Make an island for yourself, strive quickly, l>e learned. With
impurity blown away, without blemish, you will go to the heavenly
land of the noble ones.

237. You are now advanced in age indeed. You have come to the
presence of Yama_ There is not even a stopping place for you in
between. and you do not even have provisions for the journey.

238. Make an island for yourself, strive quickly, be learned. With

impurity blown away, without blemish, you will not come again to
birth and old age.

239. A wise man would blow away his own impurity, gradually,
little by little, at every opportunity, as a smith blows away the
impurity of silver. ·

240. Just as the impurity_arising from iron, rising from it eats that
very thing, in the same way his own deeds lead one excessively
'9evoted to ascetic practices to a bad rebirth.

241. Prayers have non-recitation as their impurity; families have lack

of exertion as their impurity; idleness is the impurity of (physical)
appearance; carelessness is the impurity of one who guards.

242. Bad conduct is the impurity of a woman; miserliness is the

impurity of one giving; evil characteristics are impurities indeed in
this world and the next •

243. There is a greater impurity than this impurity. Ignorance is the

. greatest impurity. Giving up this impurity, be free from impurities, o
bhikkhus.
 XVIII. lmpurity(Mala) 37

244_ Life is easy to live for a shameless person, a crow-like hero,

importunate, a braggart, arrogant, defiled_

245. But it is difficult to live for a modest person, constantly seeking

the pure, unattached, not anogant, living a pure life, discerning.

246. Whoever destroys a living crea!ure, and speaks untruth, takes

what is not given in the world, and goes to another's wife,

247_ and whatever man applies himself to drinking liquor and

intoxicants, t/iat person digs up his own root here in this verv world_

248_ Thus know, o man, that evil chiracteristics are uncontrolled_

May lust and the unlawful not deliver you over to misery for a long
time.

249. Men give indeed according to their faith and according to their
inclination. In this matter one who is upset about the food and drink
(given) to others does not attain concmtration by day or by night

250. But he for whom this (upset stale} is cut out, .removed root and
all, destroyed, he indeed attains concentration by day and by night.

25 I. There is no fire like passion; there is no seizer like hatred; there

is no net like delusion; there is no river like craving_

252_ The fault of others is easily seen, but one's own is hard to see_
One winnows the faults of others iruleed like chaff, but hides one's
own as a crafty gambler hides a bad throw.

253. Of one who sees the faults of others, who is constantly

censorious, the asavas increase. He is far from the destruction of the
asavas . •

254. But there is no footprint in the sky; there is no ascetic outside

(our Order); ordinary people rejoice in the diversified world;
tathagatas are free from the diversified world.
38 The Word ofthe Doctrine

255. But there is no footprint in the sk)r, there is no ascetic outside

(our Order); eternal conditioned things do not exist; there is no
agitation in the awakened ones.

•
 XIX. THE RIGHTEOUS (Dhammanha)

256. One who would conduct a case hastily is not thereby righteous.
But the learned one, who having detennined bo'.h what is the case
· and what is not the case, ·

257. conducts others without hastiness, with righteousness and

impartiality, protected by the law and wise, is called righteous.

258. A man is not learned because he talks much. One who is

secure, without hate, without fear '' called learned.

259. A man is not an expert in the doctrine simply because he talks

much. But he who having heard even a little doctrine practises it with
his body, he indeed is an expert in the doctrine who is not careless
about the doctrine.

260. He is not an elder simply because his head is grey. He has

reached a ripe old age, he is called "old in vain".

26L In whom there is truth and righteousnes, non-killing, self-

reslraint, taming, he indeed free from impurity and firm is called an
elder. •

262. Not by mere talk or beauty of complexion does a man who is

emious, stingy and crafty become good.

263. He for whom this is cut out, removed root and all, destroyed,
he, free from fault, wise, is called good.

264. Not by tonsure does one who is without vows and who speaks
UDIIllth become an ascetic. How can one who is full of desire and
grttd beco;ne an ljScetic ?

265.But he who quietens ·evils, small or large, in every way, because

ofthe quieting of evils.is called an ascetic.
40 The Word of /he Doctrine

266. One is not a bhikkhu simply because one begs others for alms:
having adopted the domestic way of life, thereby one is not a
bhikkhu.

267. But whoever has put aside merit and evil, a liver of the holy life,
wanders the world carefully, he indeed is called a bhikkhu.

268. Not by silence does a foolish, ignorant person become a sage.

But whatever learned man, holding a balance, as it were, and taking
the better choice,

2b9. avoids evils, he is a sage. Thereby he is a sage. Whoever knows

both in the world, he is called a sage for that reason.

270. One is not noble because one injures living creatures. Because
of the non-injury of all living creatures one is called noble.

271. Not merely by virtuous conduct and vows nor, again, by much
learning, nor by the attainment of meditation, nor by sleeping in
seclusion,

272. do I att.ain the happiness of the absence of desire, not attained

by worldlings. Nor has a bhikkhu obtained confidence, as long as he
has not attained the destniction of the iisavas.
 XX. THE PATH (Magga)

273. Of paths the eight-fold is best; of truths the four sayings; of

states freedom from passion; of two-footed ones the one with eyes.

274. Only that one is the path; there is none other for the purification
of insight. Enter on this one. This causes confusion for Mara

275. Entering on this one you will put an end to suffering. This path
was preached to you by me, knowing the destruction of the dart.

276. You must show energy. The tathagatas are (only) teachers.
Those who have entered (on the path}, meditative, will be released
from Mara's fetter.

2n. "All conditioned things are impermanent". When one-sees this

by wisdom, then one is wearied of suffering. This is the path to
purity.

278. "All conditioned things are suffering". When one sees this by
wisdom, then one is wearied of suffering. This is the path to purity.

279. "All phenomena are non-self'. When one sees this by wisdom,
then one is wearied of suffering. This is the path to purity.

280. He who does not exert himself at the time of exertion, who
(though) young and strong has come to sloth, whose thoughts and
mind are depressed, indolent, that lazy one does not find the road by
wisdom.

281. Guarding one's speech, well restrained in mind and body, one
should not do evil. Purifying these three paths of action, one would
attain the path ta\]ght by the sages .
•
- -282.Truly-frorii meditation arise5 wisdom; from non-meditation the
loss of wisdom. Knowing this two-fold path to existence and non-
existence, one should so conduct oneself that one's wisdom
increases.
42 The Word of the Doctrine

283. Cut down the forest (desire), not the tree; danger comes from
the forest (desire). Having cut down forest and desire, be without
desire, o bhikkhus.

284. For as Jong as the desire, even small, of a mm for women is not
cut off, so Jong has he a mind atiached indeed, like a suckling calf to
its mother.

285. Cut out the Jove of self, as you would an autumn lily by the
hand. Promote the path to peace, the nibbiina taught by Llie well-farer.

286. "Here I shall dwell for the rainy season, here I shall dwell in
winter and summer". So thinks the fool; he does not know the
obstacle.

287. As a great flood carries oIT a sleeping village, death carries off
that man who is intoxicated by sons and cattle, whose mind is
distracted.

288. Children are not a protection, nor a father, nor even relatives.
For one who is overcome by the end-maker there is no protection
among kinsmen.

289. Knowing this fact; a learned man, restrained by virtuous

conduct, would very quickly clear the path leading to nibbana
 XXL MJSCELLANEOUS (Pakinnaka)

290. If by the abandonment of happiness coming from material

. thingS one· might see abundant h'!Jlpiness, a wise _man would
abandon happiness coming from material things, seeing abundant
happiness. .,

291. Whoever desires his own happiness by infiicting suffering on

others, he~joined ·in close association with hatred, is not freed from
hatred.

292. For what is to be done is thrown away: but what' is not to be

done is done. Of those who are unseemly and careless the iisavas
m=se.
L

293. But those whose mindfulness of the body is constan!ly fully

undertaken, they do not practise what is not to be done, persevering
at what is to be done. Of those who are mindful and knowing the
iisavlls come to an end.

294. Having killed mother and father, and two khattiya kings, having
kiiled a nation together with the attendants, a brahman goes
scatheless.

295. Having ki!Jed mother and father, and two learned kings, having
killed a tiger among men as the fifth, a brahman goes scatheless.

296. The followers of Gotama are always well awakened, in whom

day and night_ there is constantly mindfulness of the Buddha.

297. The followers of Gotama are always well awakened, in whom

day and night there is constantly mindfulness of the Doctrine .

•
298. The followers of Gotama are always well awakened, in whom
day and night there is constantly mindfulness of the Order.

299. The followers of Gotama are always well awakened, in whom

day and night there is constantly mindfulness of the body.
44 lhe Word oflhe Doclrine

300. The followers of Gotama are always well awakened: whose

minds day and night delight in non-harming.

301. The fol!owcrs of Gotama are always well awakened: whose

minds day and nighr delight in mental exercise.

302. ll is hard to go forth; it is hard to be delighted; houses arc hard

to live in and miserable; living with those wh0 are different 1s
difficult; a traveller is beset with misery. Therefore one should not be
a traveller and one would not be beset with misery.
'
303'. Whatever region one with faith, endowed with virtue,
possessing fame ond prosperity visits, there he is honoured.

304. Good men shine from afar, like the Himavat mountain. [lad
men are not seen here, like arrows shot at nighl

305. Sitting alone, sleeping alone, wandering alone unwearied, alone

taming the self, one would be delighted in the forest.
 XXIL HELL (Niraya)

306. One who speaks of things that never were goes to hell; or the
one who having done something says be did not do it (goes· :oo):
both of these, when passed away, become the same - men with
contemptible deeds in the next world.

307. There are many with saffron around the neck, of evil nature,
unrestrained. Being evil, because of their evil deeds they are reborn
in hell.

308. It is better that an iron ball heated like flames of fire be eaten
than that an immoral unrestrained man should eat the country's alms
food.

309. A careless man who courts another's wife gains four things:
acquisition of demerit, an uncomfortable bed, third blame and fourth
hell.

310. Acquisition of demerit and an.evil state of rebirth, and a brief

delight for the frightened man with the frightened woman, and the
king imposes a heavy punishment. Therefore a man should not court
another's wife.

311. Just as kusa grass wrongly grasped cuts the hand, the ascetic's
life wrongly practised drags one to hell.

312. Whatever action is slack, and whatever vow ts defiled,

(whatever) holy living is vile, that is not very fruitful.

313. If there is something to be done one should do it; one should

make an effort firmly. For a slack wanderer scatters himself with
dust all the more..
•
314. A bad deed is better undone; afterwards one repents ofa bad
deed. But a good deed is better when done, which one does not
repent having done.
46 The Word of the Doctrine

315. Just as a border ciry is protected inside and our, so protect

yourselves. Do not let the moment pass you by, for those who have
been passed by the moment grieve when they are consigned to hell.

316. They are ashamed of what is not shameful; not ashamed of

what is shameful. Because of taking up wrong views, beings go to a
bad state of rebirth.

317. Seeing danger where there is no danger, and not seeing danger
where there is danger, because of taking up wrong views, beings go
to a bad state of rebirth.

31 g. Thinking there is a fault where there is no fault, and not seeing a

fault where there is a fault, because of taking up wrong views,
beings go to a bad state of rebirth. -

319. But knowing fault as fault, and non-fault as non-fault, because

of taking up right views, beings go to a good state of rebirth.
 XXIII. THE ELEPHANT (Niiga)

320. I shall endure abuse as a naga elephant in battle endures arrows

shot from the ~w. Ill-natured i_ndeed are the majority of people.

32 l. They lead a tamed one to war, the king mounts a tamed one.
The tamed one, who endures abuse, is the best among men.

322. Excellent are mules when tamed and Sindh thoroughbreds and
great kuiijara naga elephants: better than these is a tamed self.

323. For not by these vehicles would a man go to the region where
there is no rebirth, as by means ofa well-tamed self. The tamed one
goes by means ofa tamed (vehicle).

324. The kuiljara elephant called Dhanapalaka is hard to- control

when he is oozing with rul When tied up the kuiijara elephant does
not eat a mouthful offood, (but) remembers the naga elephant forest.

325. When be is lazy and a glutton, a sleeper who rolls as he lies,

like a great hog fed on grain, the fool comes to the womb again and
again_ .

326. Formerly this mind wandered where it wished, where it liked,

as it pleased. Today I shall control it properly, as a hook-holder
controls an elephant in rul

327. Be rejoicing in carefulness; protect your own mind; save

yourselves from the difficult way like a kunjara elephant sunk-in
mud.

328. If one finds a zealous companion, a co-wanderer, a good-liver,

wise, overcoming all dangers one should wander with him, happy
(and)
-- -
minMitl.
- --- __

329. If one does not find a zealous companion, a co-wanderer, a

good-liver, wise, like a king leaving a conquered kingdom one
should wander alone like a miitailga naga elephant in the forest
48 The Word of the Doctrine

330. Wandering for a solitary person is better; there is no

companionship with fools. One should wander alone and one should
not do evil deeds, desirous of little, like a mataliga n5ga elephant in
the forest

331. Companions are pleasant when a need arises; contentment with

this or that (i.e. with whatever turns up) is pleasant; merit is pleasant
at the end of life; the abandonment of all pain is pleasant.

332. Respect for one's mother is pleasant in the world; and respect
for one's father is pleasant; respect for ascetics is pleasant in the
world; and respect for brahrnans is pleasant

333. Virtue lasting until old age is pleasant; established faith is

pleasant; the obtaining of knowledge is pleasant; the11on-doing of
evil deeds is pleasant
 xxrv. CRA YING (T:ll)hii)

334. The craving of a man who acts carelessly grows like a maluva
creeper. rte jumps about hither and thither like a monkey searching
for fruit in the forest

335. Whomsoever this fierce craving, attachment to the world,

overpowers, his sorrows increase like bif:ll)a grass when rained
upon.

336. But whoever overpowers this fierce craving, which is hard to

overcome in the world, from him sorrows 'ran
like a drop of water
from a lotus.

337. This I say to you, sirs, as many as are gathered together here:
dig out the root of craving as one seeking the uslra root digs out the
blr:ll)a grass. May Mara not break you again and again as a stream
breaks a reed.

338. Just as a tree, although cut down, grows again, if the root is
undamaged and firm, in just the same way this suffering returns
again and again, if the latent tendency to aaving is "hot ~moved.

339. Of whom the 36 streams flowing towards pleasant things are

strong, the currents, the thoughts bent on passion, carry that man of
v.Tong views away.

340. The streams flow always, the creeper shooting up stands

(there). And seeing that creeper in existence, cut off its root by
knowledge.

341. To a creature wide-flowing and lovely delights occur. Whoever

are intent on pleasures, seeking happiness, those men indeed
undergo 1'irth and old age. _. _ . .

342. People assailed by craving run around like a hunted hare.

Attached by attachments and fettets they go to misery again and
again for a long time.
50 The Word of 1he Doclrine

343. People assailed by craving run around like a hunted hare.

Therefore a bhikkhu desiring absence of passion for himself should
tlrrust craving away.

344. Come, look at that man who free from desire, (is) intent upon
the forest, free from desire runs only to the forest. Freed, he runs
only to bondage.

345. Wise men say that that fetter is not strong which is made of
iron, wood, or reed. The longing for earrings of precious stones, for
sons and wives, which has impassioned minds,

346: this wise men call a strong fetter, which drags down, is hard to
loosen by those who are slack. Cutting even this, people go forth
without longing, abandoning the happiness of sensual pleasures.

347. Those who are impassioned by passion follow the stream they
have made for themselves, like a spider the web he has made
himsel( Cutting across even that the wise go forth without longing,
abandoning all misery.

}48. Give up in front, give up behind, give up in the middle, having

gone to the far shore of existence. With mind freed in every respect
rou will not return again to birth and old age.

349. Of a creature who is disturbed by thoughts, of fierce passion,

contemplating pleasurable things, the craving increases all the more.
He indeed makes his fetter firm.

350. But he who, delighting in the quiescence of thoughts, cultivates

the unpleasant, always mindful, he indeed will make an end (of
craving), he will break Mara's fetter.
•
35 L Having reached perfection, fearless, without craving, without
blemish, he has cut off the darts of existence. This body is his last

352. Without craving, not taking, skilled in words and their

interpretation, he would know the combination of letters, and which
 XXIV. Craving (Tanha) 51

go before and which after, he indeed, with his last body, having great
knowledge, is called "great man".

353. I have conquered all, I know all, I am not clinging to all

phenomena. Leaving all, freed in the destruction of craving, having
learned (it) myself, to whom should I point (as teacher)?

354. The gift of the doctrine surpasses every gift; the taste of the
doctrine surpasses every taste; delight in the doctrine surpasses every
delight; the destruction of craving overcomes all suffering.

355. Riches kill the fool, but not those who are about to go to the far
shore. Because of craving for riches a· fool kills himself just as (he
kills) others.

356. Fields have weeds as their defect; this people has passion as its
defect Therefore indeed something given to those without passion is
very fruitful.

357. Fields have weeds as their defect; this people has hatred as its
defect Therefore indeed something given to thos_e without hatred is
very fruitful.

358. Fields have weeds as their defect; this people has delusion as its
defect. Therefore indeed something given to those without delusion
is very fruitful

359. Fields have weeds as their defect; this people has desire as its
defect, therefore indeed something given to those without desire is
very fruitful.
 XXV. THE B~ (Bhikkhu)

360. Restraint of the eye is good; restraint of the ear is good; restraint
of the nose is good; restraint of the tongue is good.

361. Restraint of the body is good; restraint ofthe-~oice is good;

restraint of the mind is good; restraint everywhere is good. A
bhikkhu who is restrained everywhere is released from all misery.

362. One restrained in hand, restrained in foot, restrained in voice,

the best of those who are restrained, delighting inwardly,
concentrated, alone and content, him they call a bhikkhu.

363. Whatever bhikkhu is controlled in mouth, speaks gently, is not

puffed up, teaches the meaning and the doctrine, hi• utterance is
S\veet.

364. The bhikkhu whose p_leasure park is the doctrine, who delights
in the doctrine, thinking about the doctrine, remembering the
doctrine, does not fall away from the true doctrine.

365. One should not despise what one receives. One should not
wander about envying others. A bhikkhu envying others does not
attain concentration..

366. If a bhikkhu does not despise what he receives, even though he

receives but little, him indeed the gods praise him as being of pure
livelihood and unwearied.

367. Whoever does not count as his own anything in name-and-form

in any way, and does not grieve because of something which does
not exist, he indeed is called a bhikkhu .
•
368. The bhikkhu who lives in loving kindness, with faith in the
teaching of the Buddha, would attain the peaceful place, the happy
cessation of conditioned things.
 XXV_ The Bhikkhu (Bhikkhu) 53

369. Bail out this boat, o bhikkhu_ When bailed out it will go lightly
for you. Having cut off passion and hatred, then you will go to
nibbana.

370. One should cut off five, one should abandon five, one should
especially develop five. A bhikkhu who has crossed the five
attachments is called "flood crosser''.

371. Meditate, o bhikkhu, and do not be careless. Do not make your

mind wander in the str:and of sensual pleasures. Do not, being
careless, swallow an iron ball. Do not cry out while being burned
"this is sufferingtt.

372. There is no meditation for one who is without wisdom, no

wisdom for one who is not meditating. He, in whom there are
meditation and wisdom. is indeed close to nibbana

373. There is a superlmmanjoy for a bhikkhu who has gone into an

empty house (a secluctro place) with a tranquil heart, rightly seeing
the doctrine.

374. Whenever one comprehends the arising and passing away of

the constituents of peI50nality, one obtains joy and delighl This is
the death-free for those who know.

375. This is the beginning in this matter for a wise bhikkhu here:
control of the senses, contentment, restraint under the rules of
discipline. Being of good livelihood and unwearied, associate with
friends who are noble.

376. One should be in the habit of distributing gifts_ One should be

skilful in conducl Then full ofjoy one will put an end to suffering.

377. As Ille jasmine sheds its withered flowers, so you should shed
yoiii fove and hatred, obliikkhus. -
54 The Word of Ihe Doclrine

378. The bhikkhu who has a calmed body, calmed thought, calmed
speech, and is well concentrated, who has rejected the bait of the
world, is called "calmed".

379. You should urge on the self by. the self; you should examine the
self by the self; guarded by the self, mindful, you will live happily, o
bhikkhu.

380. The self indeed is master of the self; the self indeed is the refuge
of the self; therefore restrain yourself, as a merchant restrains a fine
horse.

381.The bhikkhu who is full of delight, with faith in the teaching of

the Buddha, would attain the peaceful place, the happy cessation of
conditioned things.

382. Whoever indeed as a young bhikkhu applies himself to the

teaching of the Buddha, he illuminates this world like the moon
when freed from a cloud.
 XXVJ. THE BRAHMAN (Briihm;u:ia)

383. Cut across the stream, making an effort; drive away sensual
pleasures, o brahman. Knowing the termination of conditioned
things, you know the uncreated, o brahman.

384. When the brahman has reached the far shore in the t\vo-fold
things, then for him, knowing, all bonds disappear.

385. For,whom there is neither the far shore nor the near shore nor
both, him, free from distress and without connections, I call a
bra!unan.

386. Whoever is meditative, free from defilement, seated, whose

work is done, is without asavas, and has anained to the highest goal,
him !call a brahman.

387. The sun shines by day, the moon is bright by night; the warrior
shines when his armour is fastened on, the brahman shines when
meditating, but the awakened one.shines all day and night by his
radiance.

388. Hiving put aside evils he is a brahman. Because of living in

equanimity he is called an ascetic. Making his impurity go forth
therefore he is called one who has gone forth.

389. A brahnian should not strike a brahman, nor should he release

(angry words) against him. Woe to the killer of a brahrnan, and woe
to him who releases (angry words) against him.

390. This is no advantage for a brahman, when there is restraint of

the mind from pleasant things. The more his mind turns away (from
such things), the more suffering is calmed indeed.
• •

391. By whom no evil is done in body, mind, (and) voice, him.

restrained in these three respects; I call a brahm3.IL
56 The Word of the Doctrine

392. From whom one might learn the doctrine taught by the fully-
awakened one, him one should worship reverently, as a brahman
worships the sacrificial fire.

393_. Not by matted locks, not Ly clan, not by birth, does one become
a brahman. In whom is truth and righteousness, he is pure and he is
a bra!unan_

394. What is the use to you, o fool, of matted locks, what is the use
of a garment of goat skins? There is a thicket inside you, you clean
the outside.

395."A creature wearing clothes from a dust heap, thin, with his
veins showing, meditating alone in the forest, him I call a brahman.

396. But I do not call one born in a (brahman) womb a brahman,

having his origin in a (brahman) mother. He is called "one who says
'bho'", ifhe has possessions. One without possessions and without
attachments, him I call a brahman.
.'
397. Whoever indeed having cut off every fetter does not tremble,
him, gone beyond attachment, without connections, I call a brahman.

398. Whoever having cut the strap and the thong, the fastening with
appurtenances, has thrown up the bar (got rid of obstacles), is
awakened, him I call a brahman.

399. Whoever, (although he) has committed no offence, bears

patiently abuse and flogging and imprisonment, him, with endurance
for his force, and force for his army, I call a brahman.

400. Whoeveris without anger, has taken a vow, is virtuous, is

undefiled, is tamed, with his last body, him I call a brahman.

40 l. Whoever does not cling to sensual pleasures, just as water does

not cling to a lotus leaf, or a mustard seed to the point of an awl, him
I call a brabrnan.
 XXVI. The Brahman (Briihmal}a) 57

402. Whoever in this very place knows the termination of his own
suffering, him, with burden laid aside, without connections, I call a
brahman.

403. Whose knowledge is profound, who possesses wisdom, knows

the right and wrong way, has attained to the highest goal, him I call a
brahman.

404. Whoever does not mingle with householders, or the houselcss,

both, who does not frequent houses, and has few desires, him I call a
brahman.

405. Whoever, having laid aside violence with regard to creatures

moving and still, neither kills nor causes to kill, him I call a brahman.

406. Whoever is not hostile among the hostile, at rest among those
who are violent, not clinging among those who are clinging, him.,
call a brahman.

407. Whose passion and hatred, pride and hypocrisy have been made
to fall, like a mustard seed from the point of an awl, him I call a
brahman.

408. Whoever would utter speech which is not harsh, informative,

pleasant, by which he would offend no one, him I call a brahman.

409. But whoever does not take in the world what is not given, be it
long or short, small or large, pleasant or unpleasant, him I call a
brahman.

4 IO. Whoever has no desires for this world or the next, him, without
desire, without CO!Ulections, I call a brahman.
•
· 41CWlloever has rio desires; is without doubt because of his
knowledge, him, arrived at the firm foundation of the death-free, I
call a brahman.
58 · The Word of the Doctrine

412. But whoever has passed beyond good and evil, both
attachments, him, without grief, without defilement, pure, I call a
brahman.

413. Whoever is spotless and pure like the moon, serene, not defiled,
with joy and existence exhausted, him I call .a brahman.

414. Whoever has gone beyond this difficult patli, journeying-on,

delusion, crossed over, arrived at the far shore, meditative, without
desire, without doubts, has gained nibMna withour grasping, him I
call a brahmin.

415. But whoever, giving up sensual pleasures, would wander about

without a house, him, with sensual pleasures and existence
exhausted, I call a brahman. -

416. But whoever, giving up craving, would wander about without a

house, him, with craving and existence exhausted, I call a brahman.

417. Giving up human connection, he has (also) gone beyond divine

connection. Him, disconnected from all connections, I call a
brahman.

4,18. Whoever giving up pleasure and non-pleasure, has become

cold, without acquisitions (which lead to rebirth), having overcome
all the world, a hero, him I call a brahman.

419. Whoever knows in every. way the passing away and uprising of
living creatu.res, him, unattached, a well-farer, awakened, I call a
brahman.

420. Whose (place of) r~birth gods, gandhabbas and men do not
know, him, with isavas extinguished, an arahant, I call a brahman.

421. For whom there is nothing, before and aft~r and in the middle,
him; having nothing, without grasping, I call a brahman.
 XXVL The Brahman (BrahmaT;a) 59

422. Whoever is a bnll of a man, outstanding, a hero, a great sage, a

conqueror, without d<:sire, cleansed, awakened, him I call a bralunan.

423. Whoever knows his previous abodes, and perceives the

heavens and hells. and has reached the cessation of births, is a sage
perfected in knowledge, him, having perfected all perfections, I call a
brahman.
 NOTES

1-2. I take dhamma in these two vecies in the sense ofihe objects of the
mind, the sixth sense; cf. manovififieyyd d~amm4 manassa O.jJQtha,,,
agacchanti, Vin I 184,24-:zl. Niil)amoli (1994, p. 53, s.v. dhamma) translates
dhamma in this sense as "ideas". I translate as "mental phenomena", instead
of"mental objects", as I translated in EV I. Forothertrnnslations of dhamma,
see the note on 20.
Only the P~li venion has -maya. The other versions have -Java. Brough
(p. 243 ad GDhp 201) suggests that -maya seems almost to imply a Vijniina-
vilda view, while -java reflects the fya~ika nature of the dharmas.
'fere Udiina-v 31.23-24 replaces nal?f by IGl?f; in 42-43 Udiina-v 31.9-10
Jw no parallel; in 157 Udana-v 5.1 reads enGl?f; in 162 Udiina-v 11.10 reads
enGl?f; in 169 Udiina-v 30.5 reads ena'1f; in 230 Udiina-v 22.11 reads ta11f.
Note the consonant group "" in anveti. We should perhaps regard it as an"veti,
or as a Sanskritisation of anneti. See 71 124. For other Sanskritisations of
coruonant groups see l:y in -vakya 320 321; kl in klaa 88; gy in ii.rogya
204; ty in -cetya 188; dr in bhadra 119 120 143A 143B; by in-byii.satta-
281; ly in balya 63 dussilya 162 kalyii~a 375; l'.)' in l'.)'<lsatla-47-48 ryaya
113 374 pathavya 118 vyanti- 350; sm in bhasmii.cduvma 71 };jsmici 74
-asmi(I?') (locative singular) 143A 221 224_ 247 367 395 asmil?f 168
169 242 410 rasmi- 222 asml 353 lqsma 357 380.
Here and in 71 anveti seems to mean "follow". In 124 it seems to mean
..cnta'9. -.
Jn pada d bhii.sati is m.c. to avoid the opening·--·. Cf.13 14 79 98 (?)
140 148 231 245 363 413 415 420. See EV I 61 and EV II IO 44 74.

I. Dhp-a 124,l: tato tividhaduccaritato IGl?f puggalmii duJJJwm anveti.

Dhp-a I 24,5: cakkm/i va vahato padan ti dhuro yuttassa dhurarri vahato boli-
vaddasso pada'1' cakkn11f vrya.
Forvahato (""genitive singular ofvahatu "ox", which is not listed in PED),
see Brough (p. 243 ad GDhp 201).
The word pada occurs in Dhp in several senses: I.a) "foot", as here. Cf. di-
pada in 273 having two feet, biped; Lb) "footplint°"ttack" (of birds in the
Q

sky) in 93; also in 179-80 254-55; 2.a) "word" in dJuunma• 44-45 102,
attha0 I 00, giitha 0 I 0 I, ana11ha 0 I 00-2, nirvtti.. 352; 2.b) "religious
doctrine, saying" 273 (of the four Noble Truths); 3) "place (or state)" in
cunalaptldtf._and_m_accuno padgl?f ;ll, (l!Jl(llaJ/I padmp (twice) .114, san/a'fl
Pacl"'!l 368 - 381. . .
PED states that as the second element of a compoundpada is the eqU.ivalent
ofpatha, but there seems to be no context in Dhp ·where pada unequivocally
means "path". MW does not quote pada in the sense oC"path", nor does
BHSD, so I assume that the word does not occur in this sense in Skt or BHS. ·
 62 The Word ofthe Doctrine

2. Dhp-a 137.J foll.: sarirapafibaddlwttii ... gatagataffhiine anapiiyini chiiya

viya hull'ii na vijalwti.

3-4. Dhp-a 143,20: aJ:kocchiti a.Uosi; avadhi ti paharl.

For the development of aJ:kocchJ <Ski •flkraulqilsee ~ger, § 164.
Dhp-a I 44,1-3: ajinf ti kUfasallii-otaral)ena va vddapa{ivOdena va
kara')uttariyakartlJ)ena vd ajesi; aMsi ti mama sGntaka1p pattiidisu kiiicid eva
avahari.
The case of me in pada b is not clear. The cty seems to be taking it as genitive;
it is perhaps ablative. PDhp 5-6 also has ahiisi me, but there is no equivalent
in Udana-v. .

3. Dhp-a l 44,4-8: ye lal!' akkocchi man ti iidivatthukal!' kodhal!'

saigi/adhur01!1 viya nandhinii piitimacduidini viya ca l:usadihi punappuniIJ!i
vefhentti. upanayhanti tesa1]1 sakiJrt uppannarr: veraJ?1 na sammati ti na
vupasammalL
Th.ere are seven syllables in piida c. We should. have~up=ayihanti in
OvH/KRN to give an eight syllable pada.

4. Dhp-a I 44,8 foll.: ye ftllfJ na upanayhantf ti asati amanasikaravasena va

kammapaccaveklcha7J.avasena va ye ltIJ]J akkosadivatthuka'!l kodluvra taya pi
koci niddoso purimabhave alcku!llw bhavissati, pahafo bhavissatl, kii[a-
sakkhil[I okaretvd jito bhavissati; kassaci te pasayha Jdiici acchinnaT[l
bhavissati, tasmii niddoso hutvil pi akkosJdini pii.pul)Gsi ti eval[J na
upanayhanti, tesu pamii.dena uppann.am pi veral!J iminii. anupanayhancna
nirin.dhano viya jii.tavedo UJXU.ammati. .
To rriatch upanayihanti in 3 we should have read the same here. but ignored
'l the svaral5hakti vowel in upanaylhanti.
Note the sandhi of -01!1 + upa-> -iipa- in pada d.

5. Dhp-a I 51,12-15: esa dhammp sananlano ti, esa averena veriipasamana-

sankhiito porii~ako dhammo sabbesa711 buddhapaccekabuddha-
khi1Jiisavdnaf!J gatamaggo ti.
Note ca= tu "but". Cf. 6 10 12 26 54 56 63 68 69 86 103 106-7 119
120 151172.177 190 207 245 250 254--55 256 259 263 265 268
293 314 319 336 350. See EV TI. p. 74 (ad Thi 55).
•
6. Dhp-a I 65,lS-17: mayal?I yamiimase uparam4_ma nassii.ma satatal[t
-samital,r .maccusantikar[i gacchama-tl lia jananti. yecaliittha-vijiinanti ti ye-
tattha ]JIJ1)4itit l11ay<u/I maccusamipa111 gacchiimii ti vijiinanti.
The cty gives an alternative explanation. Dhp-a I 66,2-3: ettha sanghamajjhe
yamiim;,,e bharpfaniallaI/f vuddhiyil viiyamiiind ti na vijiinanti.J.n EV I (ad
Th 275) I followed the first of these interpretations; whicli seetll$ to be based
upon a word-play (see EV I, p. 174), and translated "we come to an end
 Notes 63

here", because there seemed to be a contrast intended with Th 276, where the
ignorant act as though they are immortal. Here I translate yam- as ..restrain",
because it seems to suit the context better, although PDhp and Udana-v could
be translated 'lS ...exert ourselves". I assume that -Jmase i! intended as an
imperativc(sceGciger, §§ 122,.126).
PDhp 254jaydmatha (<yat ?); Udana-v 14.8 atrodyamiimahe {< ud-yam).
for ca - tu ""but" in pida c. see the note on 5. Ud~-v has hi in pada a. and
tu in pilda c.
Fur the construction of vijiinanti with the quasi-locative tattha, see EV I,
p. 174 (ad Th 275) and GD II, p. 135 (ad Sn 9).
Dhp-a I 66,7-1: J:JJ!aliasaJ!J:Mtli medhaga.
For the alternations thldh and klg in medhaga (< Skt 'methalca), see EV I,
p. 175 (ad Th 275). For the opposite alternation of dhlth see pithiyati 173.

7. Dhp-a I 75,4-1: pariyesanamattii pa{iggahaJJamattii paribhogamattii ti

imis.sii mattQya a}iinanalo bhojiinamhi ca amattaiiiiuT?t; api ca
paccaveilhm}amattii vissajjanamattii ti imi.ssa pi mattiiya iijiinanato
amattaiiiQori_
For ascup.vuta1r1 with the locative indriyesu, see LUders, 1954, § 224. For
sarirvuta with the instrumerttal case, see the note on 225.
For -iiiiu <Ski -jiia, see 8, alcata-iiiiii 97 383, and viiiiiu 65 229, and cf.
-gii < Skt -ga in piira-gii 348 3 84.
In pada a there is resolution of the sixth syllable. In pada c the cadence is
- -- -without a caesura after the fifth syllable; cf. the noies ·on 126 218. This
has occurred because ccimattaEiiu7!l is the opposite of c.a mattaiiiiuJ?1 in&. The
edition quoted by Brough avoids the problem by omitting ca. For opposites
causing metrical problems, cf. 14 1I7 364. We could repair the metre here
by reading c* amattafifiu~. In pada e pasahati is I!LC. to give the - - -- •
(paJhyli) cadcnee.

8. For the proclitic use of na before ppasahati, cf. 128 168 I 72 177 259.
Dhp-a 176,14-IS: OJnD1taiiEutlipa/ipakkhena bhojanmnhi mattaiiiilll!I.
For -iiiiii < Skt -Jiia, see the note on 7.
Dhp-a I 76,ltH9: liraddhaviriyan ti,paggahitaviriyGI/f paripu~~aviriyGJ!'.
Note the Jong -i- in -viriyllI/f m.c. in piida d, and cf. li- in liriya in 208 and
ariya- > liTiYa- in 236. There is resolntion of the first and sixth syllables in
plida a, and of the fQurth syllable in pada e. In pada e pasahatf is m.c. to give
the·---A .•• "'cadence.
-- ---·-----~J_ - -- --- - - - -- - --

9--JO. Tb= is a word-play on lcasava and kiisava: "The one who merits the
hisitva is the onewlio has go! rid of his lcasawi'. For other word-plays see
44--45 71 79 81 92-93 229 230 283 294-95 313 344 347 410.
In the cadence ofpada d we should ignore the svarabhakti vowel in m"hati.
Cf. 230. For another cadence of this type, see kayirati 292.
 64 Tht. JVord ofth.e Doc!rine

9. Dhp-a 182,13-!S: aniklcasilvo Ii riigiidihi kasiivehi saka.\0vo.paridahessa/i

tl nivJ.sanapdrupal)a-attharalJ.avasena paribhuiijissati; paridahissatl ti pl
pd/ho.
Tho v.L is-found in Th 969. Udana-v 29.7 reads par/dhiJsyatl; GDhp 192
parlhasadl; PDhp 94 paridhehlrl. In EV I, p. 259 (ad Th 969) I mention !ho
v.l. paridhassatl quoted by Th-a. and the B• reading parldhlssatl. Some of
!heso variants arc probably !he result of later normalisation to avoid !he nino-
syllable pMa.
Although some oflhesc readings suggest !hat !here is a svarabhakti vowel in
paridahessatl, it is more likely that dah- < dadh- has replaced dhii-, and we
should assume resolution of !he fourth syllable in pacla b.

10. Dhp-a ! 82.19-21: wmta.h:zsiiv' assii ti catiihl maggehl var.takasiivo

chaefiJitalwsirvo pahir.a.h:zsiivo assa; siksii ti catuparisuddhisilesu.
For the sand.hi of -o +ass a > -a.ssa in vantakasiivassa in pada a. cf. 3 02
389, and see Norman, 1993C, and GD II index, s.v. sandhL
For assa < •asyJJ < syiit (• siya), see 124 376.
For ca~ tu "but", see !he note on 5.

11. Dhp-a I 114,10-14: re pana /GJ!I micchiidi!fhigaha~a'!I gahetvii {hilil

kti~avitakkadina'T' vasena micchasankappagocara hutva sflasiira1J1
samQdhisiira111 paiiiidsiiraT'[t vimuJtisJraT{1 vimuttifiGl).adassanasiiTaJ?'J
paramatthasiirG'!I ti nibbiinaii ca niidlrigacchanti.
ln padas bd !here is resolution of !he fuunh syllable.

· 12. !'.>hp-a I 114,14-19: silrCU/f ca ti tam eva silasiiradisiirG'!I siiro niima

?-a1?1 vuttappakaral[t ca as&aJrs asaro ayCI1[1 ti iiatvii; te siiraip ti le pmJefiJQ
eva'l' santmfidassanaJrt gahetva fMta neillammasaiikappiidina'!I vaseJUJ
sammasankappagocnra hutvJ /Ql{f w.ttappakaral"(l stiraJ?1 adhigacchanti.
For ca= tu ..but", see the note on 5.

13-14. ln pMa b we should prob.ably read vut1hi (with !he v~L) m..c. to avoid
!he cpening • • • •. Sec the note on 1-2.

14. There is tesalution of !he fourth syllable in padas bd, because they are the
opposites ofpadas bd in 13. For !he effects upon scansion of such opposites
see the note on 7. '
- -- -- - ---- - - -- \
15. Dbp-a I 118,27 foll.: di:rva J:ammaJiliyhiiiji aiiiino iraiiano kiliffha-
kammaip passltvd socatl 1UinappaJ:iiral:mp vilapanto vihaiiiiali.
We should read kamma kiliEthGT/I as separate words, with !he original neuter
singular form kamma < Skt karma (cf. 96 217). Udana-v. 28.34 karma hi
kli1fizm agrees in reading a noun·wi!h an adjective. Note !he change of
construction to a tatp~a compound kammakilesa'!I in PDhp 3 (to match
 Notes 65

kammav/Juddh/T[1 in PDhp 4). C&P's translation ''the stain of his own action"
and other similar translations imply that the past participle kili!fha is being
taken rui a noun, the equivalent qf (or even a replacement for) kili//hl, which
occurs in Pkt but is not attested for P.aJL
For vlhaiUlad, cf. 62.
The metre is VaitllJya. In pilda b we should read papakmi rui papakari m.c. to
give tho syncopated opening - - - - - - . Cf. 17.

16. Note the change of construction to a tatpuru~a compound l:amma-

vlsuddh/T[1 where we might have expected a noun l:amma with an adjective
vlsuddhalft to match 15. Udana-v 28.35 is consistent in reading karma
v/Juddham, i.e. a noun with an adjective; PDhp 4 is consistent in reading
l:ammaviJuddhim. i.e. a tatplllllµ compound
The metre is Vaitiliya.

17-18. The metre is Vaitaliya. In piida c we should scan -e in me"' short,

ic. mlm.c.

17. Note that idha is contrasted with pecea.

In pilda b we should read papakiiri as papaJ:iiri m.c. to give the syncojiated
opening------.Cf.15.

18. In pida d -gg- in suggatiT[1 is m.c. to match duggatiT[1 in 17. Cf.

suggatbp m.c. in 319. ·.

19--20. Dhp-a I 157,ll-12: tattha sahilaJ?I ti /epi/akilssa Buddhavacanass'

.elQT[1 niimQT[1.
I follow the cty and take sahitl11/f in the sense ofsl11/fhitQT[1.
For samaiiiia (Skt .irama~ya) "the state of being a sama~a (Skt .irama~a)
'ascetic-, cf. 311.
The metre is ~tubh. In pada d in 19 and in pilda fin 20 the short sixth and
seventh syllables are replaced by one long syllable. Cf. 4.0 and 125. See Sadd -
8.2 (p.1151), Schubring, 1910, p. 54, Alsdort; 1962, p. 133, EV I, p. 178 (ad
Th 303). and GD II, p. 153 (ad Sn 61).

19. Dhp-a I 157,14-15: ta711 dhamma~ sutva yaT[1 kiirakena puggalena

lr:attabbalf' lfll/lkaro na hoti.
Note ta1frkaro for takl:aro. For the -T[1k-l-kk- (Le. NC/CC) alternation, cf.
···~u,,,lhiila31-Z65-409;:mucc<fu>aliiiuncey,yai21;hiiialhiiiiia390. -

20. Dhp-a I 158, ll'-17: dhammassa hati amullzammaciiri ti atlhaT[1 aiiiiaya

dhamltlfD!I aiiiiiiya ncivalokuttaradhammassa anuriipartt dhamma1J1 pubba-
bhagapafipadiisankhata711 catuparisuddhisiladhutaitga-asubhal:amma-
l!hiirilidibhedQT[1 CfJrol)ato anudhammaciiri hoti.
66 Th< Word ofrh< Doctrln<

I translate diiamma al .... doctrine .. here. For the translation ..mental

phenomena", sec the note on 1-2. Other translations for dhamma are
generally in line with those which ( adopted in EV I (sec p. 118 ad Th 2):
(a) ·~eachings'" when used in the plural of the Buddha's te3clllng (e.g. 82)
(b) "nature'" or "characteristic" for the sense of (moral) quality. It often occurs
withpiipaka or bisala (e.g. 242), or at the end of compounds (e.g. 45 307)
(c) "phenomena" in tho sense of the constituent pans of the world of
phenomena (e.g. 353)
(d) "righteousness" for the proper conduct of a judge or brahman (e.g. 257
393)
(c) "law" for a less specialised aspect of(d)(e.g. 176)
(f) "rule" (e.g. 5)
(f) "(virtuous) sure" (e.g. 273)
(g):~ing"(c.g. 1093&4)
(h) "practice'" or"way oflife" (e.g. 167 266)
Dhp-a I 158,22-25: anupadiyano idha va hura'll va ti idhplokaparaloka-
pariyiipanniI vii ajjhattika/Xihirii vii khandhiiyatanadhiituyo catWii upiidanehi
anupiidiyanto.
For hural/I sec EV I, p, 121(adTh10).
For the palatalisation. in anupadiyiino (< •an-upiidayiino), cf. iidiyatl 246
409, and sec Norman. 1976A, p. 338.
For the present participle ending -<ina in anupadiyiino, cl abhisambudhiino
46; esiino 131-32; anu!!hahiino 280.
Dhp-a I 158,25 foll.: maha.khil),Osavo maggasaillata.ssa sdmaiiiU:zssa vasena
Ogatassa phalasiimaiiRassa c · eva panca aseilladhamma.J:J:handhassa ca
bhiigavii hoti. •
In pada b we should read Aotf m.c. to give the break - - •• In pada e there is
molutiorrofthe first syllabic.

21. Dhp-a I 228,19-23: amatapada'!' ti ama/O'!J vuccali nibbiina'!J la'!' hi

ajQtatta na jiyyati na miyyati, lasmti amalalJl d vuccati~JXUlarri ti iminQ ti
pada1[1 amalO'!J piip!ll)lVl!i ti artho, amatassa pada1[1 amatapadm;i, amatassa
adhigamavupiiyo vutth01[1 hoti.
For pad01[I ti in the cty, we should read pada1[1ti= padar.ti, i.e. the third plural
of the root pad- (sec C&P. p. 431note4).
I take pada and (amata-)pada to mean ''place". See the note on 1. Cf.
ama/O'!J pada1[1(twice)in114 and sanlO'll padam in 368 381.
I take amata to mean "death-free", i.e. the place whereJh~i> no death. Cf.
374 411 and see Norman. 1994A, pp. 217 foll · -
Dhp-a I 229;1-3: maccuno ti mara~assa. padm[i ti up(i:yo maggo pamatto hi
maeamo
jiiti1[1 niitivartati,jiito jiyyati c. eva mfyillivd tipaiiiiido pad01[I niima
hoti marG(I01[1 upeti. ·
 Notes 67

Dhp-a I 229,12-1.5: ye pamattii yathi:i matii ti ye. pana sattii pamatt3- te

pamtidamara7Je.na matalld. yathG hi fivitindrlyiipacchedena matii diiru-
kk.h.a.rulh.madisd apagataviiiiii:i¢ tath' eva honti.
There are nine syllables in pada a. We could regard this as resolution of the
fi!lh. sixth or seventh syllables. I list it as doubtful. ·

22. Dhp-a I 230,7-12: ariylrna'!1 gocare ratd ti te eva'f1 appamiid~

pamodan!J l"'T' appamiJdaTfl var/efhetvd ariyiJnaTfl buddhapaccekahuddha-
siJvakiJruvp gocarasaTflkhiJte catusatipa{/haniJdibhede sattallTflsahodha-
pallkiyadhamme navavldhe Jokuttaradhamme ca ratiJ nlratll abhiratll hoti ti
attho.
In pJda d we should ignore the svarabhakli vowel in a,JyiJMTfl.

23. Dbp-al 231.1: dhira pa~efitll.

The word dhira has two meanings: "fum" and "wise". I normally translate in
accordance with the cty"s explanation. if there is one. Here I translate "wise".
Cf. 28 122 175 177 181 193 207 234 290 328-29 345-47. For
dhira as '"firm" see 208 261.
Dhp-a I 231,1-S: yogakkhe.mtllfl arruttaran ti ye va cattiJro yo gll maha}l17llITfl
va{/e osidifpoitl tehl khemaTfl nibb~ sahbehl lokiyalokuttaradhammehi
sef!liattii anuttaran ti.
For the tanslation "rest from exertion" for yogaillema, see EV I, p. 128 (ad
Th 32). .

244 For the use of the absolutive nisamma in a synci:ctic compound, cf.
niggaylw-vddi(n) 16; vivicca-sayana 271. For other syntactic compounds,
see yattM-kama-nipati(n) 35-36, and Norman, 199 IB, pp. 3-9.
The mdie is Vaitiliya. In pada a satima!o is m.c. to give the cadence· - • - . In
pada c we should scan saiiiiatassa as .. - - - , i.e. sa[n]fiatassa, m.c. Cf. the
scansioo of samiato in 362. In pada d we should read a[p]pamattassa m.c.

25. Dhp-a I 255,13-16: medhiivi imasmi171 atidul/abhapati{!hitaya ati-

gamb/Ull saTflSdrasagare attano pati{/habhiilGTfl arahattapha/GTfl diJJ<C!1
kayirotlia kare;rya kalu'fl saH:u!Jeyya ti attho.
Dhp-a I 255,17-19:yaTfl ogho ndbhilirati YGTfl catubbidho pi kilesogho
abhikiritJD!i viddhal!Zsef'urp na salloti, na hi sak.ka arahatla'!' oghena
viliribattl..
GDhp 111 feldsyaJara nabhimardati;Udiina-v4.5 tam ogha nabhimardati.
Sincelhc flood is saTflSara,~here dipa mnst mean «island". Cf. 236 238.
AttemJ'ing to explain why the Chinese vcnion of this vcne has «darJcness"
instcidof"flood", Brough said (p. 211) "an a.Jtcmative would be laJ/ll1I ogho
(since fiom early times a superfluous anusvara is often written in such a
position). read as laTfl mogho, and inteipreted as molio, or even appearing in
this fonn thmugh a Prakrit vcnion", i.e. a Prakrit which turned -gh- > -h-. He
68 The Word oftlze Doctrine

continued, "From this, 'darkness' would be an easy step". fn these

circwrutances the !r.mslation of dipa as "lamp" would be understandable.
For tho nasalisation ofvoweb before -m-, see the notes on 54 74 414, and
Norman, 1992C, pp. 331-38..
In pada c we should ignore the svarab~ vowel in ka/rillha.

26. Dbp-a I 257,ll-19:pamMalii anuyuii}antlpavalltnll pamadena lriiia'I'

vftindmentf.
Dhp-a I 257,21: J:ulava,,.s4gata,,. sel/hmp ll/tamaJ/I sa/lara/anadhana'I' viya.
PDhp 17 reads dhana,,. irq/ha'I' va raJ:kharl; GDhp 117 dha~a ie;hi va
ralqadi; Ud!na-v 4.10 dliana,,. Jrq/}Uva raqatt. For the selfha'l'lse{{hl
variation, see Brough (p. 212 and p. xxii). C&P (p. 433 note 17) prefer the
ide:u>f "merchanf".
For ca~ tu "but", see lhenote on 5.

27. Dhp-a I 258,4-'= mdpamadan ti tasmd lllmhe ma pamildam anuyuiijetha

ma pamiidena kiikuri YlliniJmayittha.
The cty is explaining anuyuii}elha, a second plural active optative, by means
of a negative injunctive. For the reverse of this, whereby the cty explains .a
negative injunctive by an optative, see the note on 133.
Dhp-a I 258,S-8: J:fimaratisanthavan ti val/hulriimakilesalriimesu rati-
sankhiita'!l tal)hasanlhava111 pi ma anuyuiijetha ma cintayittha ma
pa[ilabhittha.
PD hp 26 reads pram&k pramoderha na lriimaraiisandhave, and Udana-v
4.12 pramiidam niinuyujyela na J:iimaralis"'!fS/aVam. •
}t would also be possible to explain lamaratisanthava as a dvandva
'compound: "sensc.al pleasures, delight and acquaintmce".
In piida a -m in pamiidam is m.c., to allow resolution of the fourth syllable.

28. Dhp-a I 261,3: dhiro J>01!4ito.

Here dhira = "wise", as opposed. to bale "fools".
I translate pafifia, and its derivatives, as "knowledge.. or ..wisdom",
depending on !he context.

29. Dhp-a I 262.13 foll: abalassa'I' vif ti hl?fhapiida'I' chinnajava'I'

dubbalassal'{I sigluifavo s1ndhavtij&iiyo viya. sumedhaso ti uttamapaiiiio
tathiiriiJ>lll!I puggafaiii iigamena pi adhigamena pi hitvii yiiti _.. va//apadlJJ!l
jahitvii<:ha4efetvilva{takHlissaT<mtoyiitiyevii ti.- - -- ·· · · ·
In sumedhaso we have 3n example of a consonants-stem being changed into
an a-stem. Cf: ayasi124, and the comparable change of an n-stem in verinesu
197. - .

30. In pada d we should ignore the svarabhakti vowel in gar"hiro.

 Notes 69

31-32. In these verses bhayadassivii means "seeing something to be afr.iid

of', i.e. "seeing danger". Cf. bhayadassino 317.

31. Dbp-a I 283,2-4: appamiidaraJo bhikkhu appamiidiidhigatena iiliryagginii

eta1]1 sarriyojanof!1 r:fahanto abhabbuppattikmp karonto gacchatf ti.
In pacb d aggi va should probably be printed as aggiva, since it rcprescnlS the
sandhi of aggi + iva. Contrast aliipun 'eva 149. ·
For aryu'l'thn/a, with -'l'th- <-Ith-, cf. 265 409, and see CPD (s.v.
aryU'l'thiila) and OvH, Oberbl § 113. For the CC/NC alternation see the note
on 19.

32. Dhp-a I 285,ll foll.: abhabbo parihiiniiyii ti so evariipo bhiUhu

samathavipassanadhammehi vii maggaphal.J.i vii parihiiniiya ab/iabbo niipl
pattehi parihiiyati na appattiinl na piipuryiiti.
For abhabba in the sense of"unable, incapable", with the dative parihiiniiya,
see CPD, s.v. abhabba.
The dental n in nibbiina is an eastern form. For other eastern forms, see
iivudha 40; -iini 82 188;pal/ala 91 155; antaliilla 127-28; t>isodhaye
281; ve 275 315; suddha]Tve atandite 315;pa[igha 398;pure 421.

33. Dbp-a I 288,l-S: phandanG'l' ti riipiiaisu iirammaryesu vipphand-

amiina'!l. capo/an ti ekiriyOpathena asa!J!ha!ato g&nadiirako viya ekasmi1!2
iirammal)e asar;fhahanato capalarri. i:ittan ti vifiiW1)lll?J bhiimikavatthu-
iiramma1)akin"yadicittat0ya pan• elm[J cittan ti vut!Of!2.
Note uju here, but ujju- in 108.
As Rau points out (p. 161), uju'!' l:aroti is the equivalent of namayanti 80
145.

34. Dbp-a I 289 ,}-8: okamoi:ata ubbhato ti ol:apury~ryehi civarehi ti e11ha

udal:aTfl, oka'!' ol:a'!' pahiiya aniketasiiri ti ettha iilayo, idha ubhayG'l' pi
labbhati okamol:ato ti hi ettha ol:amoi:ato udaJ:asG'l'khiitii iilayii ti ayam
attho.ubbhatotiuddho.[o. ·
The ciy suggesis that the first oi:a means "Watd'. Sec the note on oka iil EV
II, p. 105 (ad Thi 236), where I noted that there is a v.L ogha- at Sp 1100,14
for oka-puryryehi at Vin I 253,14, which the cty quotes here. C&P accept the
translation "water" (p. 435 note!). For okzr="home" cf. 87 and 9L
·---·PED.says_~at uhbhata is derived<:: Skt uddlz;ta. but I see no reason to doubt
a derivation < Ski Udbh[la. ..
Dbp-a I 289,18-20: MiiradheyyG'l' pahiitave yena kilesava![asOTflkhliJena
· Miii:adheyyon' eva Partphandati li1J!' pahiitabban ti.
PDbp 343 reads miiradheyyG'l' prahiitaye; Udana-v 31.2 miiradheya'!'
prahiitavai. It would appear that the cty, or the tradition which it was
following,. did· not recognise pahiitave as an infinitive of purpose, but
assmned that -cave was fOi-tawe~ Le. ah castem neuter singular of the future
70 The JVord ofthe Doctrine

passive participle. For eastern forms sec \he note on 32. C&P follow the cty
in their transfatioa. and seem to have difficulty in interpreting the infinitive of
purpose. in their notes (p. 435 note 1). For rhe infinitive ending, cf. netave
180.
For the sandhi -m- in oka-m-oJ:ata, cf. 103 247 390.
In p.Jda_ b --0kata is m..c. to give the CJdcncc - - - • .

35-36. Dhp-a I 295,12 foll.: yallhai:iimanipatino ti ya/JM ka11hacid eva

nipatanasilassa etlll'(I hi labhitabba{!hana'!l va alabhitabba{/hana'!l va
yutta{lhiinaf!l .vd ayuttaflhdnar,t va najDnQti. n' eva )iiti1!f ololltf na gottal{1 na
vayaqt:., yattha yattha icdrari ta tr ha latth' eva ni'patati ri ya!Ihal:iimanip&i ti
VUCCfd.
Dh11ra I 300,2-4: yatlhaJ:amanipiitina'!l ti jt'iti-adinl anoloketva
lahlu1abbiilabb!UtabbayutttJ!!hanesu ya11hakatrhaci nipatanasaOl[l.
For the synt3ctic compoundyatrhak&nanipiiti(n), sec the note Qn 24.

37. Dbp-a I 304,15-17: gu1uJ niima catumahabhutaguhii idaii ca hadaya-

riipaici n/ssQya pavattaii tJ g,,Jiiis 0)'0'!1 nQ ma j GIC'!'.
Dhp-a I 305,t-2: moillanli J..!Grabandhana ti sabbe le kilcsa-
bandhanQbhfyena MJralxw:iluJnasa.nkf.ara tebhiimakava!Ja muccissanti ti,
The cty is herefore explaining moillanri as a future passive. Cf. pamollianli
276.
For saiiiiamtssantf, ct saiiiiJmay" 380.

38. Dhp-a I 308..22-23: anava({hitacittassii ti citta'!I nam~ etal[J kassaci

..,nibaddharrr va thiivarCU[I va n • atth i.
Forava{!hita.cf. Th 1140.
Dhp-a I 309,7-11: parittasaddhat15ya vJ upplavana-ullopanasaddhatiiya vii
parip/avapasQdassa lciim<Ivacararilpiivacariidibhediipi paiiiia na paripiirari
kamiivacarDya pi aparipUriyamiinQya kuto ca rUpiivacarQ.riipQvacara-
lakuttarapaiinii paripUrissari.

39. Dhp-a I 309,12-13: anavasssutacittassii ti rJgena atintacittassa.

ananviihatacetaso ti ahatacirto khilajGto ti Dgataffhiine dosena cittassa
pahafabl:avo vutto.
For an-avassuta, cf. an-uss7ita 400.

40. Dhp-a I 317,16-17: anivesana siyii ti analayo bhavwa,---

Dhp-a I-318,13-15: anivesano siya ti samiipattinivesanaf!J katvii tattha na
>ivese;ya ii/ay<IJ/I na kareyyii Ii al/ho. _ _ _
The alternation vly in iivudha, which is the equivalent of Skt iiyudha, is an
eastern feature. For the alternation, see the notes on miiluvli in 162 and
parissaya in 328, and cf. GD II, p. 169 (ad Sn JOO). For eastern fonns, see
the note on 32.
 71

The metre is Tri~ubh abc; Jagati d. In p5.d<:i c the short sixth ;1nd scvcilth
sylbhks are replaced by one long syllable. Sec the note on 19--20. In p5da b
then: is re.solution of the first syllabic.

41 Dhp-a I 321.2-3: nirall~DT'(I va kn/iiJgaraJ?l nirupakiira'!' niratrhal.of!I

4

ka{!luillianefa'!l viya.
It is debatable whether nirattha is< nirartha (POhp 349 niriittha'!') or
nirasta (Udana-v 1.35 so). I follow the cty.

42-43. In pida a we should ignore: rhe svarabhakti vowel in ka_}Jrd. On the

proaunciation of J:ayird. sec C&P (p. 439 note 26).
ForltaJ?I, sec the note on 1-2.

42. Dhp-a I 324,8-10: diso disa1.n Ii coro cora/'fl, disvii ti piifhascso. ya.?J IGr[I
ka]UQ Ii Y"1!I taqi t=a anayavyasana'1' knreyya.
M. C&P say (p. 439 note 26), the cty seems not to understand the
construction of k.ayira with two :?CC'..!s:~:ivcs, and s1:;\1Ecs 1Hs\·,l.
his not t:lsy to translate yan re ..:: k::yir,:. ?_:-i:! ~1. : :: ~) ~.:::_-.~:-::'.:; :<::·.~::·.~. _;;.::,·:
J:.ayira. We could, however, ~st:~nc tha11hc ~::'.ccc,'.~·::: h~d be::: ::ich:~'.~·d
)'C?l!f
in the relative clause: ... That whicb an cncn1y r:1ight Jo ... ". CL y.Jyc::'1 55:
ya.· U,,ani 149;y=n no 200. Udiina-v 31.9 changes the construction aD<l
hnneithcr ya111 nor IC17!f: na dveyi dve;1>;a~ kury0d vairi va vairfr::o l:itam.
C.lP (p. 439 nO!c 26) call this a more modern construction =de by using a
(<htive-)genitive, but it seems to me that the genitives depend upon hitam.
Forro pana, cf. vii pima 271.
Dhp-a I 324,17-18: dasasu akusalakammapathesu micchii{hapitattii micchii-
~ ciltQ'1'.

43_ Dhp-a I 332,.3-4: na fmri ti lal'!l karal)O'!J n' eva mfittl kareyya na pitfi na
oi;k iialaka.
Dhp-a I 332,5-7: seyyaso nal'!l taro knre ti tato karal}-ato se;yaso na'!I
J:areyya karotf ti attho.
vamtaral]I uttarilarcu!J
Udina-v 31.10 has: na Lal!J mQtii pitii viipi kuryiij ... ciltlJI?l yat kuryii.d dhitam
at.anaJi, again introduciiig hitam: "Parents, etc., would not do that benefit
which a rightly disposed mind would do".
We sbould expect 43 to be parallel to 42: "Whatever an enemy might do to
Jllm._a "'1"o'igly dispos'd mind woulddo_\Vorse than that. Whatever his
pz=ts. etc., might do to him. a rightly disposed mind would do better that
thal". To get such a parallelism, we should need to ready"'?' faJ?t in pada a. in
place of na 10f11. If there has" been a change in the reading, the reading na laJ?t
in Udina-v 31.10 clearly shows that the change is old. Rau (p. 161) suggests
n:ading na YQJ/I miiJii
For seyyaso, with the suffix -so added to the comparative seyya (< Ski
ireyru), see Geiger.§ 100.3.
 72 The JVord ofthe Doclrinc

44-45. Dhp-a I jJ4,6-7: ko vije:ssati vijinissati vtj/ini.ssati pafivijjhissati

sacchi.h:uissan· ti pucchati.
In the explanation (3J4.9-I l ). this is expanded to: ko vijexsati vijinissati
vijiinfrsati pa{ivijjhissati sacchiknrissali ti pucchati.
Dhp-a I 334,13-14: pacessari ti vicinissati upaparikkhissati pafivljfhissati
sacchiJ:arissati ti al/ho.
For the suggestion that there is :i \VOrd-play here between vici- in pada a ..to
undastand" and paci- in pada d '1o pluck", which suggests chat the verse was
transmitted, and perhaps composed ~a dialect \vhcre intervocalic -c- and--j-
both became -j- or-y-, see Norman, 19898, pp. 158-59. The similarity of
explanations for vijessati and pacessati St' .... ports the view that a word~play
was intended. For a similar word-play based on the dj al<emation, see the note
on J3.S. For the cl} alternation. see OvH, 1981, p. 821, note I. For word-
plays see the note on 9-10.
Dhp-a I 334,7-9: Yama/okaii cii Ii catubbidharri apayalokaii ca. imarri
sadevabm ti manussalokaii ca devalokena saddhi'!I.
This implies that, despite the position in the verse. imarri sadevakar,i goes
withpaflurvi'T', not Yamalokar;.. Udana-v 18.1-2 has sadeval=n, which can
only agree with Yamalokar,i, and I translate accordingly.
Dhp-a I 334,11-14: dhamrnapadarri sudesitan ti yathiisabhavato kathitattii
sattafimsabadhapaillikadhammasG'!'khiitG'!' dhammapadarri.
Dhp-a I 335,&-9: imarri sukathi1arri suniddi//harri bodhapakkhikadhamma-
pada7!1.
I take pada to mean ''word" in dhammapadar;.. See the note on 1. PDhp has
pluraLforms dJ..ammapade sudciite.
-rpe metre is V aitfiliya. In pada d ppa- is m.c.
44. The n1etrc is Vaitiliya. The opening ko ima'!J gives th.:: scansion - · - ;
Udana-v 18.l has ka imiim, giving lhe correct scansion.

45. Dhp-a I 334,16 foll.: sekho ti adhisilasikkhii adhicillasikkhii adhipaiiiiii-

siklcha J! imii tisso sikkh.a sikk.hal)alo sotiipattimagga/fhQ7fJ iidil!l katvii yQva
arahaitamagga({hG sattavidho sekho imam artabhii.vapa{havil{J arahalla-
maggena tato chandartiga1!1 apakatfcjhanto vijessati vijinissati vijDnissali
pa{iVijjhissati sacchik:uissati..
Dhp-a I 335.10-11: pace.rsati vicinissati upapariillissati pa{ivijjhissati
· sacchi.huissati.
The metre is Vaitfillya Piida c has a redundant long syllable at the beginning
which stlpports the view that sekho is an old mistake for sa hi or sa hu "he
indeed" (Udana-vl8.2 has sa hz). See Norman, 1971B, p. 220 (ad GDhp
302). .

46. Dhp-a I 337,s-<;: ablUsambudhiino bujjhanto jananto ti atth_o.;

 Nota 73

For the ending -/lna in abhisambudhllno, the present participle of the verb
abhisambudhati, see the note on 20.
Dbp-a I 337,6-&: papupphabini ti Miirassa papupphakasa'l'J:hiitllni te-
bhiimikiinl va{{llnl ariyamaggena chindilvii. ·
I do not Jmow of any connection between Milra and flowcn, other than this
metaphorical usage of puppha_ See the note on 47-48. GDhp, PDbp and
Udina-v (N) have p(r)a-, but Udana-v (B) has ltl_p!Lfpalani. On this, see
C&P (p. 442 note 7).
The metre is Tri~tubh. Note the brealc- • - in pada d..

47-48. Dhp-a I 361,9-tO: byllsattamanasa'l' nara'l' ti sampatte vd

asampalle vd laggamdnasl1171.
For the consonant group".>'" in vydsalla-, see the note on 1-2.
In pada ah' eva is presumably <hi+ eva , but it is not C3S"f to see what the
force of eva is here. It is possibly to be taken u an emphatic particle
"galhcringjlower.r", but perhaps we should follow C&:P, and translate "only".
PDhp 128 lw h~ GDhp294hasyewr;Udlna-v18.!4 haseva.
Dhp-a 1362,~: Mmagw;asaJlJ:hiiriinl puppMnl htw:1 pacbranJmrf sampatte
va asampalle va J:iimagw;e byi!.ra/taJfumasaJ/I nararri.
The cty makes it clear that puppha is being talcen Dlelapborically, and perhaps
Mara's papupphakiinl in 46 refer to this association of puppha and
Jcamagw;a. .
In piida a there is resolution of the sixth syllable.

48. Dhp-a I 366,10-12: arilla'l' yeva kiimesii ti va11huJ:amaWesakiimesu

pariyesanena pi pa(i/Gbhena pi paribhogena pl nidhlinena pi atittal[1 o>a
sam.Gnarp.
Dhp-a I 366,18-20: antako kurute vasaJ?I ti maral)QSalikh5to antako
1:andaituuri paridevantcop gahetvii gacdumto attano vascop piipeti ti al/ho.
For an/aka "making an end, causing death. i.e. death", see MW, s.v. Gntaka.

49. Dbp-a I 374,16-17:pupphaii ca vtu;~aii ca gandluzii ca ahe/hento

avinJsento carati. _
For the suggestion that pupp~ wvp;agandhG'l' is a split compound with the
split introduced to give a caesura at the end of the pada, see EV II, p. 64 (ad
Thi 25: Kiisijanapado suilko). For split compounds with "'!'· see the notes on
98 114 1;!3 147 171 346. For a discussion, and rejection, of the
-- suggestion-that-puppha'l'-here is ablative,-see Nonnan,-1971B, p. 219 (ad
GDhp 292), althongh Dhp-a I 374,I ("m the stoiy, not the explanation) reads:
pupphato rasaga~~antena (read rascop ?) bhamarena va.. Cf. PDhp 127
pu;pli; Udana-v· IS.8 pu;piid. Brough (p. 267) seems to prefer the
constm<;tion of iidiiya with two accnsatives.
For possible examples of the ablative in -ctl/I, see the notes on 135 201 259
314.
74 The Word of Ihe Doctrine

In p5.da d muni is me. to give the _cadence · - - - .

50. Dhp-a I 379,7-9: na paresarri vilomQni ti pare.sa'!l vilomQni pharusiini

mammacchedakavacan&i na inanasikiitabbiinl.
In p:ida c C&P translate "but", iinplying that they are reading ca for va. For
the alternation c/v, o( 55 63 83 108 lJS--139 (?) 159 (?) 163 (?) 229 (?)
237 306 396, and see GD II, p. 146 (ad Sn 38). I take va hero as the
equivalent of eva "of oneself alone, only of oneself".
We might have thought that J:aralcatam was m.c. for Ire.tam/am "the various
things done" but. if this was the original meaning of the piida, tradition has
taken it to be J:ara + a!ala, as pada d makes clear.
This verse is not appropriate to the Pupphavagga. as Rau (p. 162) points out.
Cf. 57.

51-52. In the pada d a-kubbato is opposed tosa-kubbato, although the laner

is ungrammatical, and sa- is unne<essary. It is not found in PDbp 126
kurvvato, GDhp 291 kuvadu orUdana-v 18.7 kurvataf). The al!Cr:nation a-Isa-
is probably by analogy with a-gandhaJ:a/sa-gandhaJ:a and a-phala/sa-pha/a.
See EV I, p. 182 (ad Th 324).
In piida d there is resolution of the first syllable.

53. For the meaning "kind" for Skt gu~a at the end of compounds, see MW,
s.v. gu~a. Cf. kiima~ 371.
Here kusa/a is probably the equivalent of puiiiia "merit. meritorious deed".
For the opposition of ~ala to piipa, see 173 183 and EV I, p. 247 (ad Th
872).
In piida b we should ignore the svarabhakti vowel in J:aylrii.

54. Dhp-a I 422,17-18: candanan ti candanagandho. tagarama//ikii vii Ii

imesam pi gandJ:.o eva adhippeto.
As the cty makes clear, we have to understand that candana'!J, etc .• are the
equivalent of car.danagandho. etc. Cf. 56. They may, therefore, bo inter-
preted as abbreviated compounds. See the notes on 75 257, and GD II,
p. 186 (ad Sn 195).
The v.l. lagarmri mal/ilii bas probably arisen from tagaramallil::a by the
nasalisation of -0 >-Gl!f bcfore'm-. See the note on 25. The reading may,
however, be m.c., as a result of the break - - - being changed to give the more
common - - - . ,_
Dhp-a 1422,20 foll.: saran ca gandho ti sappurisiinam pana buddha•
paccekab~ si/agandho pa{iviilam eti. J:bpkiinu;ii ? sabbii disa
sappuriso p<rViiliyasmii sappuriso si/cigandhiha sabbii disii tijjliotthdritila
gacchati tasmii tossa gandho pa!iviitam eli ti vattabbo, tena vutlall' pa{iviitam
etiti.· . - ' .
 Notes 75

The cry recognises thal sata1]1 here is the genitive plural of the prCsent
participle of the verb as-: "of the good ones". Cf. 77 151, and see the
nominative plural santo 83 151 304. Contrast asala'!' "of the bad" 73 77,
and see asanl(o) 304.
For ca• tu ••but", see the .note oil 5.
The metre is T~tubh.

SS. Dbp-a I 423,2-4: etesan Ii imesaf!J candan°adinQl[I. gandhajatana'!'

gandliato sf/avanttfna'lf sappurisJ.na113 silagandho va anuttaro asadiso
appa{ibhiigo.
There is a v.l. ca forva in the cty. For the vie alternation. sec the note on 50.
Forjata in the sense ofjJti "'Ort, kind", see MW, s.v.]Jta.
Here at ha bas the sense of "and", in contrast to "but" in 85 136 387.

56. Dhp-a 1430,IO: appama//o ti parillapon·ma~o.

!n pada b I take ytiya'!' to be <yo+ ayD'!'. with yo agreeing with gandho. B•
readsyvJya111 tagaracandanam, indicating yo. Ud.ina-v 6.18 has yo YD'!'
tagaracandanJt and PDhp 123 yoya'!'. For the sandbi of -o +a-> -li-, cf.
appassultiyD'!' 152. For the antecedent ayD'!' appearing in the relative clause,
see the note on 42.
We must assume that tagaracandanf stands for tagaracandanigandho. Cf.
54.
Dhp-a 1430,10-11: yo ca silava/D'!' ti yo Pana silavattinmri si/agadho.
For ca= tu ..but", see the note on 5. ·

57. For the sandbi -d- in samma-d-aiiiiii-, cf. 72 86 96 166.

Since is no connection with flowers. this verse is not appropriate to the
Puppbavagga, as Rau (p. 162) points out Cf. 50. It was presumably inserted
after 56 because of the mention of sila.

58. Dhp-a I 445.14-15: SaT?Jkilradh3nasmin ti sarpkQra{{hilnasmirri

kacavararO.simhi ti attho.
Note -dhiina for the more usual -kU{a.

59. Dhp-a I 445,18: sa'!'kiirabhiitesiJ ti sa'?'kiiram iva bhiitesu.

It is not easy to see how -bhiitesu can go with the two singular fonns in_,,_
For the SUj&CStion Jhat we should divide -bhiite su, where su is probably <
Skt sma, see Norman, 197IB, p. 218 (ad GDhp 304 -dhama'u)- PDbp 136
has-bhiitesu, Ud.ina-v 18.13 has-bhute '.rminn.
Dhp-a I 445,19:puthuijane ti puthflna111 kilesiina'!'jananato eva111 /addha-
nJme /okiyamahiijizne.
- Note the etymology which the cty gives for puthuj/ana.
 76 Th~ IYord ofthe Doctn'n~

60. Dhp-a II 12,14-lS: dighii ri raui niim' esa tiyiimamattd va }Ogarantassa

pana diglui horl.
The long -1 in ram is not m.c.
For the combination of tlyiima andjtigaratl, see the note on 157.
Dhp-a Il 13.4-6: yojanan ti yojanam pi catugihnitil matt.am eva sa111assa pa;1a
Jilantassa dighalfl hoti diguryCU!' viya khiiyali.
Note santa < Slct iriinta "weary".

61. Dhp-a Il 23,10 foll.: seyyalfl sadisalfl a/tana Ii altano silasamiJdhl-

paiiMguryehl adhiliitara,,, viJ sadisGlfl vii na lahheyya c' eva.
For the accusative singular seyyalfl, showing the dropping of tho final -s of
Jrcyas and transference to the a--Oeclension, see Geiger,§ 100.3.
Dhp-a II 24,10-11: d:acariyCU!' da/hGlfl kayiriJ tlcibhiivam tva thiralfl katviJ
sabba-iriyapathau eko va vihareyya.
Thc:'cty is taking da/ha'!' as an adjective, but it is possibly an adverbial
accusative. CL 112 31I.Udina-v14.15 has ekacaryiilfl drefhaT[I lcuryiin,
which seems to be an adverbial accusative.
Dhp-a II 24,11-16: n' at/hi bii/e sahiiyatii ciifasilam ... cha abhiiiiiii ayGT[I
sahiiyatiiguryo, biiJarri nlssiiya n' alt hi.
We can translate biile in piida d as a locative "in a foor, but it is more likely an
instrumental plural with -e < Slct -ais. Cf_ 330 and the note on dhiro 207.
For the instrumental plural in-e, see Geiger,§ 79.6. Udina-v 14.15 has niisti
biile sahiiyatii, but 14 .16 has bii/asahiiyalii.
In pfu1a c we should ignore the svarabhalcti vowels in -carya,,i and kay1rii.

62.. Dhp-a II 28,J: haiiiiatiyihaiiiiati dukkhiyati.

For vihaiiiiati, cf_ 15.
,C&P translate hi as "even~ as though it were pi, but Udana-v 1.20 reads
·iitmafva Jry Gtmano ntlsti kasya putrab kuto dhanaT{f, confirming that it is an
emphatic particle

63. Dhp-a Il 30,6: maiiiiati jiiniiri.

Dhp-a II 30,7-8: tena so ti tena kiiraQena so pugga/o paQr/lto vii hot/
pmy/itasadiso vii holi.
In piida b vapi may be < va (< iva) + api, or va (< eva) + apl, or va + api.
C&P translate "even like", presumably taking va as< iva. The cty clearly
talces it as va, and so has to.add an alternative. I assume that va is< eva, and I
take api as an emphatic particle: "learned indeed". PDhp 184 has pa~gito ciipi
tattha so; Udana-v 25.22 has iii biila~ sa pa~fjital}. For the dv alternation see
the note on 50.
For the consonant group -/y- in balyGT[I in padaa, see the note on 1-2.
PDhp 184 has biilamilni, which matches piida c better.
 Notes 77

For -miinln inpa~efitamanln, see EV I, p. 256 (ad Th 953). Rau (p. 163)
suggests reading pa~efilaTflmlinf, presumably m.c. to produce the - -.- •
(pat/¢) cadence.
For ca• tu "but", see the note on 5.
In pAda a maflfla!f is m.c.

64-65. Dhp-a II 31,17 foll;: dabbf suparaSGTfl yathJ, yathJ hi dabbi yiiva
pariillayii nlinappaJ:Jriiya supavikatJyiJ SOTflparivat/amanli pi ida171 /o~/ka'fl
ida171 alo~ikaTfl tittaka171 kJiarika171 ka[ukaTfl ambi/QJ71 anambila171 kas<ivan ti
suparasGJ71 najiinlili, evGTfl biilo ... dhammOTfl na}iinlitl.
PDhp 191-92 and Udana-v 25.13-14 have -rasiin (i.e. plural "flavours") in
piida d, perhaps because of the list of flavours in the cty tradition. For -a171 as
an accusative plural ending, cf. d7 167 320 373 399 412 417 423, and
see Liiders, 1954, §§ 188-219.
In piida b we should ignore the svarabhakti vowel in pa/rupiisati.

65. Here api has the concessive sense. For the sense of totality see the note
on JOO-I.
For-iiiiii < Skt -jiia, see the note oa 7.

66. Dhp-a IT 36,IJ• and Dhp-a !IT 271,7' (ad 207) print amitten • eva, giving
the impression that we are dealing wih the eraphatie particle eva (and C&P
take it that way here, but as "as with i foe" in 207), but the cty explains
(Dhp-a II 37,1): amitten' evii ti amittabhiitena viya verinii viya hutvii (cf.
Dhp-a III 271,22 ad 201:yathii amittena), showing that amilleneva is<
amittena + iva. PDhp 174 amitlel}a-r-iva; Udina-v 9.13 amilrair iva. See
also 207, and ct: ka{/hakasseva 164 rajatasseva 239, ·and contrast aliipun'
eva 149.
Dbp-a JI 36,18: dummedha Ii duppaiiiiii.
PDhp 174 dummedhii; Udana-v 9.13 dw;prajiia. For the cty explanation in
oae tradition agreeing with the canonical reading in another, cf. 83 380, and
see Norman, 1980, pp. 71-73.
In p1ida d -pphafmrz is m.e. This is in contrast to -pphalOTfl in mahappha!OJ71 <
mahat-pha!OJ71. See the note on 312.

67. Dhp-a II 40,7-8: /aJ7I Xa/OJ71 na siidhu na szouiaro,,, na laddhakaTfl.

Dhp-a II 40,9-10: vipiil:aqipa{isevati Ii anubhoti.

-·· 68. Dhp-a IIt4i,14-i5:-pitlVegena jJatiio somana.isavegena ea sumi:mo hutvii.

For ca = tu "but", see the note on 5.

69. Dhp-a II 50,13-16: madhuvii ti hiilassa hl plipGTfl akusalakammGTfl

lcarontassa ftl/!Z kammarrz madhtl1!1 viya madhurarastD!J vQ.ia i!!hC11!% /canla'!f
maniiPGJ71 viya ujxzghiiti iii n(llll so ma.dhuvii ma.iiiiati
 78 The Word ofthe Doctrine

l;Jdana-v 28.18 (D) reads madhuvad manyate bii/o as a conjecture. (N) is

missing at this point. lt would be possible for Skt madhuvad to develop >
madhuvarri > madhuvii in Pali. Alternatively, we might suppose that madhuva
developed< madhii va (- iva). GDhp 283 reads mahoru manna/I sadhu, and
Brough (pp. 262-{;3 ad GDhp 283), guessed ;hat the Udana-v tt3ding (not
available in Skt) was probably madhurGJ/I. C&P (p. 447 note 21), following
Brough, suggest that the GD hp reading raises the possibility that madhuvii is
a change from an original madhurarrz. It is, however, by np means certain that
the GDhp piida is the parallel of the Dhp piida, since the other three piidas of
the verse differ from the Dbp padas.
Dhp-a ii 50,21-22: a!ha so biilo duJdrlwJri nigacchati vir.dati pa[ilabhati.
For ca= tu '"but·-. see the note on 5.
In pada a mannari is m.c., and in pad.a c paccpti is m.c. to give the w - - ..

(pathyii) cadence. There are ten syllables in 'pada cl M it stands it is a

posterior Vaitaliya piida, and I list this verse as mixed Anu~\Ubh
(Sloka)!VaiLiiliya. We could produce a Slolca piida by omitting either atha or
biilo. Udana-v 28.18 does not include bii/o. fausb0ll sugge5ted the omission
of biilo. Cf. 119 120.

70. Dhp-a ii 63,6: sa'!'khiitadhamma vuccanli iiiitadhamma tulitadhammii.

lesu he{[himako{iyii sotiipanno· sGl/lkhiitadhammo uparimako/iyiJ lrhii]iisavo
imesa'I' sa'l'khiitadhammiina'I' so balo kala'I' niigghatl so/asin ti,
puggaliidhi!(hiinii l1J"l'1' pan' ettha aJtho.
See GD II, p. 367 (ad Sn 1038), and cf. Pj II 587,12-15:sG'l'!hiiladhamnul ti
anicdidivasena parivimD'1'Siladhammii. PDhp 385 has sii1hat~iii;ii'I';
389 siildchiita-. Mvu III 435,20-: sviilrhyiita-dhanniii;fiiz:. Udana~v 24.20E (B)
has sviikhyiita-dhannasya as.a conjecture; (N) reads sviillryiita-dhannmasya.
, Brough (index p. 310) says that sal{lkhata is a mistake for sarpkhiita. See
· C&P (p. 447 note 23), who say that the Pali form saillata is based on
sankhOta, without explaining the short vowel. Utt 9.44: na so .sakJJJya-
dhammassa kalaT{J agghai solasirri, with a v.l. suyaJckbayassa dluunmassa,
with suya- presumably < suva- < sva-, which agrees with the Udina-v
reading. The cty clearly takes it as a possessive compound "ofthose who have
realised the doctrine". If we take it as a descriptive compound. then dhamma
must be plural "of the well realised mental phenomena-. This is to assume a
derivation < siurikhytlta; saTpkhata must. however, be < sarpsJ:rta, in the
sense of "well formed, perfected, refined". In the accompanying verses in
GDhp and PDhp there is an inanimate object;e.g.prasiida. dtta,anukampiJ,
which favours a descriptive compound here. The fact; however. that Mvu III
435,16* foll. has sampannasiliinii'I' and dhyiinaprasiidanii'I' as parallels,
supports the interpretation as a possessive compound. .
The original version doubtless had sakhaia in a script which did not write
anusviira•. long vowels or double consonants, giving the possibility of
interpreting saldz- as SiIJ??lrh- or salch- or salckh-, and-ala as-ala or-dta.
 Notes 79

Note the sandhi of na+ agghati > nligghati in pada d, violating the two morae
rule (see Geiger, §§ 5-7). This is probably the effect of Sanskitisation upon
the Pali scribal lllldition. For other violations of the two morae rule see
bhasmiicchanno 71; miipp- 121-22; niissa 124 389; niibba~arrz 124;
dussilyarrz 162; n.l~no 165; iirogyarrz 204; pftvi!. 205; p/yiippiyam 21!;
viikkara~a- 262; valyam 320-21; (yathii1tana 323;) ntiiiiiesarrz 365.
In place of so in piida c, PDhp has tarrz and Udana-v tad "that is not worth
... ",but Utt 9.44 has so.

71. Dhp-a II 67,19 foll.: sajjulchiran ti IQJ/1 kha~arrz yeva dhenuya thanehi
n.iillanlal!f abbhUl}lsakhira'!I mucc.ati ti pan·namati; ida111 vuttcnri hoti: yathii
ida1'{1 sajjuV:iral!f fa1?t klra')an ii.eva na muccati na parlnamati na pakatil[I
jahan, yasmbir pa1111 bhiijane duyhitvi! gahita711 yi1>u tattha takhldi-ambilarrz
na pakkhipanti,yihu dadhibhiijaniidikarrz ambilabhiijanarrz na piipu~iiti tiiva
pakati711 avijahitvii pacchii pajahati, evarrz piipakammam pi kayiramiinam eva
na paccatl, yadi vq..,c<.>Ya na koci piiparrz kamma'!l J:iitwri visah<.>Ya. yiiva
kusaliibhinibbatta Band/iii dharantl tava na'!I te rakkhantl, taarrz bhedli
apdye nibbattal:khaadhesu vipaccati vipaccamiinaii ca efahan taJp balam
anvell•.
The cty seems to be1aking sajju and khirG'T' as a compound "s3ID<>-<lay milk",
which it explains zs abbhu~ha-l:hira711 "still warm milk". PDhp 107,
however, has saff"'!'chfram and Udana-v 9. l 7 sadya~ l:;iram.
Although all Illl!Ilmaipts and editions kllown to me read muccati, PDhp l 07
mucchati and Udaa-v 9.17 mW-chati and the glosspari»amati make it certain
that the reading should be mucchati, as Liiders (1940, p. 184) suggested. In a
context where milk is said to coagulate or curdle, mucchati makes better
sense. It might be llacught that muccaJI was due to a dialect which de-aspirated
aspirated sounds. hat it probably represents a word-play intended orally upon
mucc- and muccJ..._ When the redactor wished to· write the text down and bad
to choose betwea mucc- and mucch-, be probably chose the mucc- form
because the verse is about kamma and he assumed that the usage of the verb
mucc- with .tam-was the more important of the possible intepretations of
the verse. See Noonan, 1989C, pp. 379-80. For other word-plays, see the
note on 9-10.
Dbp-a II 68,12-1-t;oam evarrz papakammam pi yena hztturi hoti la'!f /Xi/am
dutiye vii tatiye Yiidtabhiive niroylidisu nibbattarrz t/ahanlaI[t anugacchati.
Rau (p. 163) suggats reading <fahan IQ1/l, and compares efaharrz aggf va
gacchitti 31. Thea; however, aggf is masculine and ¢aha'!' is therefore
appropriatc.tt.J~9.17 reaas dahan-tadM/am anvili, which would seem
. to support Rllu's mggestion, but we really need a neuter participle to agree
-· with kamma. Aldlaogh in theory Ski dahat could develop > Piili daha711,
Geiger lists no D<IRr forms with this ending, and Bernhard suggests that the
Udana-v form is a pseudo-Sanskritisation of dahantalp. PDhp 107 agrees
with Dhp in readiogdahantarrz.
 80 The Word ofthe Doctrine

For the consonant group -nv- in anved in plda c, sec the note On 1-2. For the
consonant group -sm- in bhasmdcchanno in piida d, see the note on 1-2. For
the Jong vowel before -cch-, violating the two morac rule, see the note on 70.
In originpiivaka was probably an epithet "cleaning, purifjing" often applied
to fire 8lld then used as a synonym for fire. Jn 140 we find bothpdvoka and
aggi.

72. Dhp-a l1 73,8: yiivad evii ti avadhipmicchedanatthe nipiito.

For yiivad eva in the sense of"merely,just simply", see BHSD (s.v.yiivad-
eva) and PED (s.v. yiiva). For -d as a historic form in a fossilised sandhi
position, cf. tad 326; etad 390. lt could, however, be analysed as a sandhi
-d-, i.e. yiiva-d-eva. For sandhi -d-, see the note on 57.
Dhp-a l1 73,"-12: iiattan ti jiinanasabliiivo, yam pi sipparp jiiniiri yamhi va
.
issariye yase sampatn"yan ca thito jaJ10Ja iiGyati pQkafo paiiitlto hoti lass·
-
eta1J1 nama.
.
lt seems therefore to be the fame or reputation which one acquires as the result
of knowledge of some sort, i.e. "reputation for skill", as BHSD (s.v.) defines
jiiGtra.
Dhp-a 11 73.14-19: suJ:kalflsan ti kusalako!{hiisalfl, biilassa hi sippolfl vii
issariyal{I va uppajjamO.naT{J kusalako{fhdsal(I ghii/entam eva uppajjati,
muddhan Ii paiiiiiiy~ ellllfl niimaT[J. vipii.tayan ti viddhcup.sayamanal]f tassa
hilaf!t suJ:ka'!'SGT[f hanantaT[I paiiflfisal[lk.Juilalfl muddhal(l vipOtental[I
viddhalflsentam eva Jumti ti.

73. Dhp-a ll 76,19 foll: asatan ti biilo bhikkhu avij/amiinarp sambhiivoJU11!1

iccheyya assaddho samiino saddho ti malfl jano jfiniltiJ ti icchatl ti ...
asanlasambhiivantllfl icchat/.
' The cty iecms to be tiling aSa/alfl as aa:usati;,e singular, i.e. as the equivalent
of asantarp, "non-existent (honour)". If this is correct, then we must supp=
that asanta'!' bas become asiilalfl m.c. to allow resolution of the first syllable
in pada a. I have translated it as a genitive plural (as in 77) "honour of(~
from) bad mentt. PDhp 178 reads asat!Vp ~ genitive plural U<ilna-v 13.3
reads asanto /abham icchanti, making asanto a non:Uo.a.tive plural as the
subject of the sentence.
Dhp-a II 77,16-20: pujii paraJ:ulesu cii tin ' eva miitiipi/Unntllfl na iiiitakiinalfl
paresa.J{1 yeva lrulesu: aho vat' ime mayham eva dade:y)'W'!I na aEiiesan ti
evalfl catiiM paccayehi piijmrz icchati.
The cty's explanation includes pii}arp icchati (Dhp-a II 77,19-20). PDhp 178
bas piijiilfl, and Udiina-v 13.3 piijiilfl; Rall (p. 164) suggests readingpil}"'zl
I for piijii, but there seems to be no objection to taking piija as plural,
I' presumably on the basis of one piijii in each family.
j '- In pada c we should ignore the svarabhakti vowel in issa,,Jyarp.
Iv'-J- ~,-
 Nor'es ~ 81

74. Dhp-a II 78,2-3: eva'!' gihi ca pabbajitii ca ubho pi mam eva nissiiya
kalD.l!J parini{{hitarr maniiantii. ti Saf!lknppo uppajjati.
The absence of any case ending on kata calls for comment, and fuiu (p, 164)
sug6ests reading katarri maiiiiantu. The form has clearly caused problems for
recensionists. In place of mam' eva kata maiiiiantu "'let them think that this
has been done by me". UdJna-v 13.5 has miim eva nityaT{ljiinlyur ..let them
IG1ow-". If the Udana-v tradition thoughc that the pada CO!ltained the verbjiiii-,
then we could assume that the original form of the Pali·\vas mam' eva katam
aiiiiantu (the present indicative aiiiiiiti occurs at Vism 200,29). By nasalisation
(see the note on 25) this became katarµm afiiiantu. which must have been
interpreted as kata'!l maiiiiantu at some time before the commentary .was
composed, since it includes the reading maiiiiantu in the explanation. On~ the
existence of the word maiiiiantu was confirmed by the cry, any cha.Ilges
introduced into the pada revolved around the existence of that word. The
introduction of nasalisation had an effect upon the metre. Since the opening
- - - - is not the usual one for the cadence - , - - .. in a prior piida in the Sloka
metre, a short syllable was produced by dropping the nasal of l:atlll/l. This
returned the verse to its original form, but beCause the commentary included
the word maiiiiantu, the verse was interpreted as containing kata maiiiiantu.
The parallel version at PDhp 179 reads mameva k.atamannentu gr'hf
pravrajitii ca ye, for which an explanation similar to that deduced for the Pali
version can be sunnised.
Dhp-a II 78,S-6: mam' eva ativasii assii ti gihi ca pabbajitii ca sabbe pi mam'
eva vase vottantu.
Dhp-a II 78,1:5: navavidhamiino ca vatfcj.hati. For mQna see the note on 221.
For the consonant group -sm- in kismici, see the note on 1-2.

75. Dhp-a II 102,6-8: aniiii hi liibhUpanisii afinii nibbiinag&nini ti

liibhUpanisii nam' esii anna eva aiiiiii nibbanagGmini pa{ipada.
I understand upanisii with nibbiinagiimini, although the cty's suggestion of
pafipada would also be a possibility. This would be an example of an
abbreviated compound, for nibbiinagiiminf-upanisii. For abbreviated
compounds see the note on 54. For thC Illeaning_"means, way" for Upanisii,
see CPD (s.v. upanisii), and BHSD (s.v. upan4ad).
Note that in piida f -hr- does not make position in anuhriihaye. CT. Bralununii
105; brohaya 285; brlihma~o 294-95 383-423. This suggests that these
vmes were composed in a dialect where initial br- had become b-, and the
spellingswipi hr- arethe result of later Sanskritisation.

76. Dhp-a II 108,8-13: ya pana tathiiriipalfl vajjal{I dinoli vajjibruriipalfl

tajjen_to ptDJ_ii_mento ·daTJ<fakamnimp karonto Vihiii-a niharantO sikkhapeti
aytlJ!l -niggayhaviidi namii seyyathiipi Sammiisaritbuddho. vuttmp h • etal!I:
niggayha niggayhGha'!l·Ananda vakkhami pavayha pavayh'iz s&o so /hassati
ti.
 82 The H'ord of the Doclrine

For Ylll?f _in the sense of si quem, see 108 and cf. yo = si quis 102 103 106
107 11~15, and see GD II. p. 169 (ad Sn 96).
For the absolutive niggayha- in a syntactic compound, see the note on 24.

77. Dhp-a II 110,4--9: safa'!J hi so pijo hori ti evwiipo puggalo buddhadinG.l[I

sappurisdnal]1 piyo hoti, ye pana adif{hadhammii vilil)')aparalokii
iimisacaklchukii jlvikatthiiya pabbajilii /esOTJ1 asala'!J so ovadako anusG.sako
na tvcuri amhiikal'fl upajjhiiyo no iicariyo kasmii amhe ovadasi ti eva'!I
mukhasattihi vijjhanttinwri appiyo hoti Ii.
For satal[f., see the note on 54; for asatQI{l, see the note on 73.

78. For a discussion of kalyiil}a miua, see Collins, 1987, and EV II, p. 100
(ad Thi 213).

79. Dhp--a II 126,15~16: dhamn:apT11· ti dhammapiiyako dhammDIIJ pivanti ti

attho.
Udana-v 30.13 reads dhannapritih sukha'!' .iete;.PDhp 348 readl dhamma-
pritirasa'!' piitra; GDhp 224 has dhamapridi suhu .Sayadi. There is a word-
play on piti < Skt piri "drinking" and piti < Skt priti ·~oy''. Pada a says that he
sleeps with a clear mind. perhaps unfuddled by drinking the dhamma as
opposed to the intoxication he would have experienced if he had drunk strong
drink. Pada d. however, states that the wise man delights in the dhamTTJa.
which suggests that piti is also to be taken as "joy''. The cty presumably did
not see the possibility of the word-play, since it explains dhammapiti only as
drinking the dhamma. The redactor of the Udana-v had to choose between piti
and priti, and perhaps because of the idea of "delight" in the last pada. or
perhaps because he was follo,ving a different comrnentarial tradition, he
'decided to read priti. GDhp and PDhp similarly have forms from priti. See
Brough (p. 244 ad GDhp 224). I follow the cty in my translation. For other
word-plays, see the note on ~10.
Dhp-a 11 127 ,5-7: sadii ramati ti evariipo dhammapiti vippas!JJ1nena ceiasa ti
viharanto paJJ.¢iccena samannQgato sadii ram.a.ti abhiramaii ti.
In pad.ab vipasannena in OvH/KRN is a misprint for vippasannena. In
pada c we should ignore the svarabhakti vowel in ar"ya-. For -pp- m..c. in
pada c to avoid the opening .. - ~ .... see the note on 1-2.

80. Dhp-a II 147,11-12: usulcii.riipi tOpetvO tejana111 namayanti u.iwf2 karonti.

As !um (p. 164) points out, namayanti is the equivalent ofujwri karoriin 33.
Cf. 145 (where subbatii replaces panefitfi), and Th 19 877. See EV I. p. 125
{ad Th 19) on the possibility of reading namenri in pada a and namayanti in
pilda d. The translation "bend the bow" in EV I is incorrect. It should read
"bend the arrow (straight)".
The metre is Vaiifiliya.
 Notes 83

Bl. Dhp-a II 148,20 foll.: ekagha'JO asusiro selo puratthimGdibhedena v6tena

na samirafi na iiljati na calati evaJ?1 af!hasu pi Iok.adhammesu ajjhottaranJesu
paryfitii. na samiiijanti pa(ighavasena vQ anunayavasena va na calanti na
kampanti.
For the word-play on samirati and samiiljati. made originally in a dialect
where 1be two forms Coincided, presumably through the alternation
-lry-1-ijj-l-inj-, see Brough (p. 250 ad GDbp 239)- For other word-plays see
the note on 9-10. PDhp 93 reads samirati and samiranti; Udana. v 29.49 (B)
prakampyare and kampyante, (N) (missing) and pramathyanli; GDhp 239
sabhijadi and sammijadi.

82_ Dhp--a II 153,4: eva'!l dhammii.ni ti desan§dlumJniini.

Note that desaniU!hamma'!' sutva occurs in the explanation (Dbp-a II 153,6).
On the basis of dhamu 1u~itvana at GDbp 225, Brough suggested (p. 245)
that, as dhamma is not neuter, the reading dhamm&U sutvOna is probably an
old mistake for dhammt5 sunitvii.na. This suggestion.., however, does not solve
the problem, because dhamma would still be neuta, and we should need to
change it to dhammaf'(I. It is, however, more likely that dham"1iini is an
eastern masculine ending. cf. pabbaliini 188. For eastern forms see the note
on 32_ For the metathesis of syllables, see the note on 106 and EV II, P- 119
(ad Thi 271). PDhp 275 bas dhal!'m<i~i folldna; Udana-v 17.l l has evam
irutva hi saddh.armcvµ.
Note the parallel between vipasanno (which is a misprint for vippasanno in
OvHIKRN) and vippasidanti.
In pada a we should ignore the svarabhakti vowel in r"hado.

83. Dhp-a TI 156,IS: sappurisii. cajanti ti arahatlamaggaiiii.;;ena

apak.arf4Juintii chandarii.gaJ{J vijahanti.
Udiina-v 30.52 reads vrajanti, GDhp 226 vivedi, and PDbp 80 bhavan/i_ See
Brough (p_ 245)- OvHJKRN adopted the reading vajanti, despite the cty and
the v.l. cnjanti. For the dv alternation see the note on 50_
Dbp-a ll 156,15 fo!L: kiimakii_mii ti kiime l:iimayanI<i kiimahetu kiimakZira~ii
na lapayanti santo ti buddhadayo santo kiimahetu n • eva attanii lapayanti na
pararri lapiipenti.
It would appear that the cty is taking kiimakiima as an ablative of cause: '"from
love of passion". PDhp 80 bas kiimakiimii; GDbp 26 bas kamakama. These
forms could be taken as a possessive compound "having love of passion", but
Udana-v 3fl-52 bas an ablative: kiimahetor_ For the Pili cty having the reading
of a canonical text in another tradition.., see the note on 66_
C&P (p-452 note 16) point out that the cty takes lapayanti as both simplex
and causative. If it is for /Qpayati, then there is shortening of~- m.c. Note,·
boweva,sankhillarri a/Iha'!' lapayissdmi SI 31,3•, and GD II, P- 344 (ad Sn
929), quoting BHSD (s.v. lapayati "boast")- It is more likely to be a tenth
class verb than a causative.
 84 The Word ofthe Doctrine

Dhp--a-11 15 7, IO- I2: sukhena phu!{hd atltava dukhena ti desaniimattam ela'!I.

a{{hahi pana lokadhammehl phu//hi1 tu{/hibhiivamankubhiivavasena vii
var;l)ahhal}ana-aval}!JObhaT)anavasena vd uccdvacalJi QkJ.rarri parycjitii. na
dassayanti.
For phuffha used in connection with an external object impingeing on an
individual, see the note onphuseyyu 133,
The metre is Tri;tubb. In pada c duJ:hena is m.c. Pada d has twelve syllables.
We could correct the metre by reading n· uccOvacarri or noccdvacaJ?1.

&4. Dhp-a II 159,9-10: na pultaf!J icche ti puttcuri va dJo.anaJ?l va ra((hOJ!l va

papakammena na iccheyya.
The metre is Triw1bb ab; Jagatl ed. Pida c has fourteen syllable3. We could
read n · iccheyy' (or icche with the v.l) for na iccheyya. PDhp 326 has necche;
GDbp 324 na ichi. SST reads na-y-icche, where -y- may be for yo or may be
a glide consonanL It is probable that iccheyya was introduced into the pada in
place of icche because of the gloss iccheyya. .-
In pada a we should read -hetil (- *heto < Skt heto~. i.e. ablative) m.c., to
give the break-. - - ; cf. EV II, p. Jn (ad.Thi508).

85. Dhp-a II_ 160,20 foll.: athiiya7[1 itari1 pa}i1 ti ya panliya7[1 avasesii pajii
sakkiiyatiram eva anudhdvati.
With piira-giimino, cf. p"ara-g'J 348 384 para-gavesino 355 and para-goto
414.
For the implication of inferiority in itard paj~ ci 104 222.
Here atha ~ ..but". Cf. 13~ 387, and contrast 55.

86. Dhp-a II 161,J-2: sammadakkhate ti samma ailliite sukathite dhamme

'ti.
For sand.bi -d- in samma-d-akkhiite, see the note on 57.
For ca= tu "but", see the note on 5.
Dhp-a II 161,5-10: piiram essantf ti te evarUpG janii. nibbclnapiiral[I
gamissanti maccudheyyan ti kilesamiirasaJFJkl2iitassa maccussa niviisana-
ghiinabhiita'!I tebhiimakavaffa1!1 suduttaran ti yejanii dhammiinuvattino re
etairr suduttaral!J duraiikkama,ft maccu.dheyyaJ?.'I tari."tvii atikkamitvii nibbiina-
piira'!' gamissanlf ti artho.
Presumably those who reac.h the far shore have crossed beyond the realm of
death, as the cty stales; but the text does ·not say that: It makes DO sense to take
maccudheyya1[t in apposition to pfiral[I. and so pad;_<fd must be taken as a
separate sentence.

87. Dhp-a II 162,6-9: ka~haT{I dhamman ti kiiyailuccaritiidibheda'!'

akusaladhammGJ?I JaMtvii sukkaT{I bhiivethii ti pa~¢ito· bhikkhu abhi-
nikkhamal)ato pa{!hiiya yiiva arahi:J.ttamagga .kayasucarit&libhedmri Sulck.mµ.
dhamm"'!' bhiiveyya: ·· ··
 Nota 85

Jn this verse about abJndoning the kru:ham dhammallf, literally .. the blftck
dhamma", and cultivating the sukka'll (dham'""'f'), ·~he white (or bright)
<lhamma", C&P (p. J 70) take these forms as singular, and translate "Having
forsaken a shadowy dhamma, the wise one would cultivate the bright";
Kalupahana, 1986, p. 121 translates "A wise IIl3l1 should develop a bright
charJctcr abandoning the shady one.. ; Radhakrishnan. 1950. p. 87 translates
"Le< the wise man leave the \vay of darkness and follow the way of light'".
The conunentary on the verse where it occun at S V 24.21 makes it clear that
it is refening to the very common categorisation ordhammas (in the plural) as
akusala and k.usala (kal}ha~ ti, a!usalapd}uvrunaf!t. suAJ:an ti, kusala-
dhammam. Spk III 132,20-21). The paijlllel vcciom of the verse have plural
forms: PDhp 263 kihne dhamme viprah'iiyaSukre bluiveJha pa~tjitii; Udina-v
16.14 krenDf!I dharm3f!1 viprahaya Jukld,,. bAlivayata bhik,avah. For
dhammaf!1 as a possible accusative plural. sec the note on 64-65.
Dhp-a II 162,!0-13: okii anoka'll ligammii ti ohv/I vuccati iilayo, anol:am
vuccati anQ/ayo, Qfayato niklchamitvJ andlay~CU{J nibbanClJ!I pa{icca
iirabbha tom panhayamiino bhiiveyyli ti attho. .
For aka "home'', cf. 34 and 91.

88. Dhp-a I1 162,17-18: cittaklesehl ti paiicahl nivar~ehl al/iina'!' vodapeyya

parisodheyyii ti attho.
For the consonant group kl- in -klesehlinpada d, Stt the note on 1-2.
In pada c we should ignore the svarabbiliivowel inparlyodapeyya.

89. Dhp-a II 163,3-5: jutimanto te iinub/riivavanlo arahattamaggan3~ajutiy3

khandhiidibhede dhamme jotetvii !hitii Ii attho.
Dhp-a II 163,9-JO: anupiid3no viya padipo apG(ll!altil:abhiiva'!I gatii ti attho.
CPD gives "not to be designate~ having no name" for a-ppaiifiattika, i.e.
having no paniiatti (< SkJ.prajnapll).
For the iisavas, sec C&P (p. 454 note 7), and EVI,p. 133 (ad Tb 47).
In piida ejutimanto is m.c. to give thc---"{path).i)cadence.

90. Dbp-a II 166,7-9: va!(a,,; pana khepetvii Jhitokhi~3savo gataddhl niima

hoti, tassa gataddhino visokassa ti va!{amiilakassa sokassa vigatatta
visokassa.
Dhp-a II 166,IJ-14:pari[aho na vijjati ti Juvidho pari,lnho kiiyiko cetasika ca.

9 l: bhp:'\ IT 170ifolL: yathii goearosampa,u.apal/a/e saku~ii attano

gocara'!' gahetva gamanakii/e mama uda~ mama padumal{J mama
uppalaJ!l mama km}T}ikii ti tasmiT?J fhane kismid iilaytl1[1 akatva anapekkha va
l0f!1- padesal!l pahQya uppatitvcl awe ki[amiinli gacchanti. evam eva
khfl)iisavG --· aniilaya anapekkha va gacchanti.
. Dhp-a II l 70,14-15: okam okan ti iill1J'G7!1 sabbiilay< pariccajanti ii al/ho.
86 The JVord ofrh~ Docrn·ne

For okam.okal!I. cf_ okdmokata 34. PDhp 2J l has oknrri olw'!I jaharriti but
Udina-v 17.1 re.ads okam ogha'!ljahar.Je le, which would mean "they leave
their home. the water'* or "their watery home .. , which is appropriate: both for
gec:sc, and also for the mindful, if we take ogha in the sense of the flood of
saJ!Uiira. Sec GD IJ, p. !38 (ad Sn 21-22) and p. 265 (ad Sn 638). Herc
C&P (p. 453 note 5) quote the Umna-v reading, acd suggest it as an
altanativc translation (p. 174).
For pal/ala, c[ 155. Forthcr// alternation, see Liidcrs (1954, § 61), and cf.
antalilJ:ha 127-28; pa/igha 398; palipatha 414. For rll as an eastern
fearure. see the note on 32.
ln pada"a satimanlo is m.c.to give the · · · • (pathyii) C<ldcnce.

92'--93. Dop-a n 173,IS foll.: gali /esfl7/1 durannayii ti yathii niimii ii.l:iisena
gar~ saku!)&Ul'!l padanikkhepassa adassanena gari durannayii. na Saliii
jii.nilw!f evam eva'!l yesal[I du.vidho sannicayo n' auhi. ... gamanassa
apaiiiiiiyanalo gati durannaya na sakka paiiiiapelun n·.
Thcte seems to be a word-play upon the possible translations of duranni:Iya as
"hanl to follow" and "hard to find". As it is difficult to follow the track of
birds in the sky. because they leave no foot-prillts, so it is difficult to find the
place of rebirth of such persons_. For word-plays, see the note on 9-10.

92. Dhp-a II 171,2 l foll.: pariiiiiiitabhojanii ti lihi pariHiiiihi pariiiiiiita-

bhojanii: yiigu-iidinaf!I hi yiigubhiivQdijiinana'!l iiiitapariiiiiii, ahare
pafillilasaiiiiQvasena pana bhojanassa parijiinana'!I tiral)apariiiiiii,
kahaliilkarahare chandariiga-apakm}.efhanaiiiil}-Df!l pahiinapariiiiili, imiihi
lfhi pariniiahi ___ pariiiiiO.labhojanii.
I take pariiiii&a in the same sense as pa:riNia-ya ...(knowing and) renouncing'"'.
SeeGD11, p. 187 (ad Sn 202).
Here gan· has two senses: (way of) going. i.e. track or path, and also place of
rebi~ as in sugati and duggati. See the note on 420. For gati in the sense of
refuge, see the note on 380.
In pad.ad we should read yesa[l!1] to give the cadence - - - - _

93. Dhp-a II 175,18-23: padan lassa durannayan ti yathii likase

gacchantiinal!f saku'}i5.narr imasmi'!J Jhiine pii.dehi akh:imitvO. gatii idaJ?1
{luilla'?1 urena paharitv_ii gatii idal[J sisen.a i.dmri pakkhehi ti na sakka iiii!UJ?1,
evam .eva1!1 cvariipassa bhikkhussa nirayapadena va gato_ t{ra_c_c_hiina-
yonipadena vli ti--adini5. nayena padairi paiiiiiipe_fW!J nOmli na sakkii ti.
Although pada here is the parallel of gati in 92, I take it to mean "foot-print,
trace, track" not "path" for the birds, and "place (of existence)" for the
khil)Osava persori.. See the note ..01'. 1.

94_ Dhp-a £1 1-77,15::-.17::.indriyO.ni ;~,;wtb~-dantabh®af!1 nibbisevana-

bhiivam 2atiini. tassa navavidhami5.nDJ?1 pahiiya !hitartii. pahiil.amiinassa.
 Notes 87

For mOna. Cf. 221.

Pali pihayari is< Skt sprhayati with dissimilation of aspirates. Skt sprh- takes
the genitive, da!ive or accusative case. Cf. pihayanti IBI;pihet' 209;
pihayam 365.
The mctre is Tristubh abc; Jagati cl

95. Dhp-a JI 182,3-4:.tiidino ti lasso pana eviirii.passa sugatiduggatisu

sar:saranavasen.a SQl/1SiirQ ndma na hoti ti.
Here and in 94 and 961 take tlidino in pilda din the general sense of"such a
one". In p3da _b, however, I take tiidi in the more specific sense of ..of such a
kind = Buddha-like or holy, venerable". Se EV I, p. 131 (ad Th 41) and
BHSD, s.v. ti'iyin
Dhp-a 11 181,7-12: bhiillaveyatM nlima pa/havi}'a!?' sucini gandhami'iliidini
pi nikkliipanti asucini multakarfs&lini pi niWipanti tathd nagaradvdre
nikhGra'!I indakhilaJ{I dJrakadayo omuttenti pi iihadayanti pi, apare pana IW!J
gandhama/adrhi sakkaronti, tattha pa/haviya va indak.hilp.ssa vQ n' eva
anurodho uppajjati na virodho.
For indakhi/a as a simile forunshakability, see GD IL p. 192 (ad Sn-229).
Dhp-a llI 181,14--15: vati'i""'I' nmdarataya subbato.
For subbata, see tl)e note on 145, and cf. abbato 264.
Dhp-a II 181,15--18: so ime hi ma1f! catiihl paccayehi sakJ:aronti, ime pana na
sakknrontf ti sakkaraii ca asakkarafi ca karonte.su n' eva anurujjhati no
virujjhati, atha kho pafhavisamo ca indaJ:hiliipamo eva~ ca hoti.
The metre is Vaitallya. In pad.a a we should scan the final -o in samo as short
m.c .• i.e. -samO. In pad.ab we should scan -i- and the final -o in -khiliipamo as
short m.c., i.e. -khI!iipamO.

96. For the form of kamma, sec the note on 15.

Dhp-a II 185.24 foll.: santan ti tassa khil)Qsavasama(lassa ahhijjhiidinaJ{I
abhiivena mana'!1 santam eva hoti upasanla'!f. nibbula'!l, tathii musii-
vadadina'!1 abhavena vaca ca pGl)O.tipiitadinfll!J abhiivena kiiyakammaii ca
santam eva hoti.
Dhp-a II 186,3-4: sammadaiiiiirvimuttassii ti nayena hetuna jiinitvQ paiica-
vimuttihi vimuttassa.
For the sand.bi -d- in samma-d-aii.iiii, see the note on 57.

97. For the paradoxical alternative translations given for assaddha, akatafiiiii,
sandhicche¥, hatCivakasa and vantO.Sa in this verse, see Rau (p. 165) and
Norman. 1979A. We might have expected a second verse, with a variant for
uttamaporisa at the end. giving the opposite .!lleaning, but for the suggestion
that uttamaporisa is also a paradox, see Hara; 1992
There is an ambiguity about akata. It can mean "not made", i.e. "'uncreated"; Jr
can also mean "having nothing which has been made Or created". Either
meaning is appropriate as an epithet of nibbiina. but .in view of the other
 88 The Word ofthe DoctrlM

cpithell beginning with a- applied to nibMno 1 it is probable that the second is

usually intended. See 383 and Norman, 1994A, p. 221. · · · ·
for the ending -iiiiil < Skt-jiio, cf. 383 and see the note on 7.

98. Dhp-a II 195,20-22: ya11ho orahonto yihorontl tam bhiimirr

rdmal)e;ytzkan ti so bhWnipppadeso ramal).iyo evd ti a/tho.
It is not clear how bhiimi111 can be neuter. It is possible that it is accusative,
after vlharantl, in some unusual way. PDhp 245 reads /a1fl bhomilrr
rama~iyakarr (presumably understanding bhauma [nc:ufer] "that is earthly
delightfulness"); Udina-v 29.18 reads te dcliJ. rama~iya!Wi) (making it
nominative). For the explanation as··a split compound, Le. bhiimi-
rtlma~eyyaka'l' "delightfulness of earth", see EV I, p. 262 (ad Th 991),
although there is no metrical reason for the insertion of -rr here. Note that Ch
readJ bhilml-. For split compounds, see the note on 49. It is possible that
bhilmi7!1 is a ""correction" for bhoma111 (< bhauma) via• bhiimal{l.
for ninna see Norman. 1979D, pp. 48-49.
In pada c we should either ignore the sva:rabbakti vowel in ar"hanto, or see
the second and third syllables as short (see the note on 1-2) with resolution
of the sixth syllable.

99. In pada a there is resolution of the first syllable. In pada b ramati is m.c.

I 00--2. I take -pada- to mean "word''. in pada b. See the note on 1.

I 00--1. I take api in the sense of giving totality to a numeral or quasi-

numeral, rather than in tho concessive sense "even". See MW (!.v. ap1) antl
GD 11, p. 241 (ad Sn p. 87,4). Cf. sobbam pl 108; ell0'7' pi 196; ubho pi
'306. .

100. Dhp-a II 208,20 foll.: anotthapadasal[lhita akiisava~~anapobboto­

va1)1Jar0vanava1)1')an3dini pakii.sakehi aniyyJnadipakehi anatthakehi padehi
SC!J!1hitJ ydva bahuJ:ii Jiond tiiva pQpiJ:a evQ ti attho.
I take -poda to mean "word" in atthopoda. See the note on 1.

101. Dhp-a 11 217,1-3: ekal[I gathapodm,i S<,Y.YO ti, appamMo amatapodilm

... pe ... yatha mata (= 21) Ii, evariipd ehl gatha pi seyyo ti ottho.
The cty seems to be tiling gathopada to mean "verse". I take -pada to mean
"word". See the note on 1.

102-3 •. For yo= si quis, see the note on 76.

102. I take -pada to mean "word" in dham;,,apado. See the note on I. I

assume that the syntax is lax, and'anotthopadasiv/lhit.i agrees with gathii in
gathiisatOl[I. PDhp 377 reac!S gathafotOl[I .- aitatthapadiisiihiJaip. On tlie
 Notes 89

basis ofUc!Ana-v 24.2 we can assurnc that Ud.lna-v 24.l hadg~tha.fatmn ...
anarthapadasa,,,hitam. The parallel in GDhp 309 is inconclusive.

103. Dhp-a II 226, 17 foll.: el:Liii ca jeyya-m-attlinan ti yo ·pana

rallil(luJnad/vd{/hlJnau ajjhattikakammaf(hO.na'll sammasanto attar.o
lobhddikilaajayena attilna.l!f fineyya save sarlgflmajullamo d so sa1ttgdma.;
jitJIUI1!2 uttamo pavaro sarrigiimasisayodho ti.
For the sandhi consonant -m- in jeyya-m-attiinarri, se·e rp,e note on 34.
Note the sandhi of -i + u- > -u- in sangJmaji (< Skt -fit)+ ullamo >
:raltglimajll!tamo. It would be possible to take this as a compound, or as two
sepanue words: "he is the best conqueror'' or "he is the best of conquerors".
For ca - tu "but", suggested by pana in the cty, see the note on 5. ·

104. Padas ab go with 103, and padas cd with 105.

Dhp-a 11228,17 foll.:jitan ti lingavipallaso, arrano kilesajayaia attlijito senho
d attho. ya cJya,,, itarii pajii ti yii paniiyaf[1 avasesli pafiJ jiitena va
dhana/uuw:rena va sa,,,glim< baliibhibhavena jitii bhaveyya IQT[1 jinantena
)'Q1!1jitmp na ta,,, seyyo ti attho.
Qearly the cty was p=led by the fact that jilaTf' seems to be in agreement
with attif, but is neuter. Rather than following the cty in assuming a change of
gender, I take jlta,,, as the past participle used as a noun "conquering" or
..conquest", in apposition to a11a, i.e. "the self as a conquest" is better than
other people as a conquest, i.e. the conquest of the self is better than the
conquest of other people. For a possible example of tr change of gender, see
the note on 201. For a past participle used as a noun, cf.jita 105 179;
apafila 105; hula 106-1;pajjalita 146; mamiiyita 361.
Dhp-a Ill 34,3-4 (ad 125): posassii ti idam pi apareniikiinna saltiidhi-
ltlcanam eva.
For posa, see 125. For the relation between poso and Skt jJ1Ulq(1, see Bailey
(1960, p. 84) and GD 11. p. 171 (ad Sn 110).
lnyii ~we presumably haveyii ce aya,,,; cf.yaii cein 106 foll.
For the implication of inferiority m. itarii pajfi, see the note on 85.

105. Dhp-a II 229,11-14: yathii dhaniidihi pariijito paillantaro hutvii

itarena jitmp puna jinanto apajilaJ[Z kareyya eYGJ'/l apajita.rp lfitW{J n • eva
~llattho.
Forfitm!o and apajita'?' as past participles used as nouns, see the note on 104.
For the labialisation· of -a-> -u- in Brahmunii, cf. 230 and sotthim 219, and
. - seeNonnan,•1916B." - . - ...
In pada c we should ignore the svarabhakti vowel in kaylrJ.

106-7. For huta,,, as a past participle used as i noun "sacrificing", see the
note on 104.
For yo= si quis, see the note on _76_
 90 The Word ofthe Doctrine

· For yaii ce - "than if', cf. 229 308 and yd ciiy1117' 104.
For ca .. tu "but". see rhe note on 5.

I 06. It i> not easy to analyse sat017Uanta17f, since -sam1117' "year" appears to
be singular. Ud.ina-v 24.21 has the compound samMatam. Mvu Ill 434,19•
has fotalfl SoflITfl. The Udina-v reading suggests that the original form in Fali
was samasataf!1, which developed to samaqtsatal"[l by the common
alternation of a long vowel and a short nasalised vowel before a consonant
(VcNNC). This compound was probably assumed to co~ist of two separate
words sama171 sal(IJ71, both apparently accusative singular forms, which could
therefore be recited/written in the reverse order. since the scansion was the
same .either way. See Norman. 1992D, p. 216.
In Mvu Ul 434,19• jayeta occurs for yajeta. For the metathesis of syllables,
see .the note on 82

108, Dhp-a II 234,9-ll:yi!lhan ti yehhuyy<namaitga/a);jriyqdivasesu dinna-

diina1'[l. hutan ti_ abhisal'{Jkharitvii k.alaJ?1 pahutadanaii c'.eva k.ammaii ca
phalaii ca sadda}dtva katadiinan ca.
For the cognate accusative yi{/haJ71 with yaje!ha, cf. acari carikmp 326.
The cty seems to draw a distinction berween yi!lha and iwla.
For YGJ?I = si quern, see the note on 76.
Note the v.L va ... va for ca ... ca; for the vie alternatio~ see the note on 50.
If we read va ... va, we must suppose it is m.c. for vd ·-- va.
Dhp-a II 234,!8-19:yOJ71 pha/aJ7J tato catubhagam pi sabbalfl 11111' diinm;i na
agghati, tasmd ujjugatesu ahhiviidanam eva S<.Y.)'O d.
I take pl after sabbai;i as implying totality. See the note on api in 10(}...I.
We might regard -ff- in ujju-gatesu a. m.c, but it occurs elsewhere in contexts
:where lengthening m.c. is not required. It is possible that this is a rare
development of r > ur, i.e. rju > •urju > ujju. See Geiger, § 12, note I.
Contrast uju in 33.
The metre is Tri~tubh. In pada d there is resolution of the first syllable.

109. For -sili(n) at the end of compounds in the sense of "practising', see
MW. s.v. Slla.
Dhp-a II 239,5--9: vaddhapaciiyino d gihissa tadahu pabbajite dahara-
siima~erepi pabbajitassa va pana pabbajjiiya va upasampadiiya va bu<fdha-
tare gu't)avuddhe apaciiycimanassa abhivadanena va niccaJ?1 piijentassa ti
attho.
As Fausb0ll pointed out (see Muller, 1881, p. 33 note), the same verse, with
slight variations, occurs in Manu 11.21. Winternitt (1933, p. 84 note 2)
suggested that this probable quotation from a brahmanical text was given
simply to connect the following verses (ll(}...15) with iL It is noteworthy that
it is not found in this introductory position in PDhp, GDhp or Udana-v, and
the verse is in fact found only in GDhp 172, in the Suha-varga.
 Notes 91

110--15. For yo - si quis, sec the note on 76.

112. Dhp-a II 260,8-10: vlriyam tirabhato da/hon ti duvidhm,. jhana'T'

nibbaltanasamatthal'fl viri)KUTI iirabhantassa.
I take dafha'T' as an accusative used adverbially. Cf. 61.
In pada d we Shotild ignore the svarabhak1i vowel ia vir'YGJfJ. C-Ootrast -viriyo
in pWab. .

113, In p!das bd note the conson.ant group -vy- in -vyayaf!I instead of

-bbayarp. For the consonant group, see the note on 1-2.

114. In padas bd amala'T' pada'T' is probably a split compound, with -'11-

\ m.c. Sec the note on 49, and cf. amata-pada in 21. for ama!a sec the note
on 21.
I take~ to mean "place". See the note on 1.

116. Dbj>-a III 4,11-12: ahhiltharethd ti turitaturitmn sighasighOJ;< hzn.r.yti ti.

I take~· to be a locative of the field of activity: "One should lusten in
respect ofrhc good".
In pida c dandha'TI is presumably an adverbial accusative. I translate it
"slowly" to give the contrast with abhiuharetha. For dandha, =Turner
(CDIAL 6169 •Jandha- "stupid"), and cf. BHSD (s.v. dhandha "slow,
weal; dutr'). Note dandheti "to be slow'" (Th 291 ), and cf. BHS dJuuuihiiyati
"is slow, is dull". ·.
For -tlh- instead of -1/- in ahhittharati (< abhi-tvarati "to be in haste"), see
CPD (s.v.) and Norman, 1989C, p. 383. Rau (p. 165) suggests that
abhittharetha is the equivalent of ahhistareta.
In pada d rama!f is m.c. to give the cadence - - - - .

I 17. In piidas abc we should ignore the svarabhak'.i vowel in kay1r-.

I I 8. In padas ab we should ignore the svarabhakli vowel in kay1r-. In pad.a c

the svarabhakti vowel is required m.c. For the effect of a negative OD the
scansio~ see the note on 7.

119-20. For ca= tu "but", sec the note on 5.

In piida a passati is m.c., and in piida c paccati is m.c., to give the - - - •
·- (pathya) capencc. - .

119. For the consonant group -dr- in bhadra'T' in piida a, see the note OD 1-
2.
There are ten syllables in pada d. ill it stands it is a posterior Vaifiliya pada,
r
and list this verse as nrixed Anugubh (Sloka)!Vaitiil!ya. We C()ufd produce a
 92 The Word of the Doctrine

· Sloka p3.da by omitting either a1ha or p6po. Uctana-v docs not have piipo, but
PDhp does. Fausboll, followed by Rau (p. 165). suggested the omision of
piipo. Cf. 69 120.

120. For the consonant group dr in bhadrcvri in pl!®' abed, sec the noto on
1-2.
There are ten syllables in pilda d. As it stands it is a posterior Vaitaliya p:lda.
and I list this ve= as mixed Anu;\Ubh (SlokayYaitfillya. We could produce a
Sloka piida by omitting either a1ha or bhadro. Uctana-v does not have bhadro,
but PDhp does. Fausooll, followed by Rau (p. 165), suggested the omision of
bhadro. Cf. 69 119.

121-22. Dhp-a Ill 16,tt-t4: miippamaiiiiethii ti na avajiineyya piipassii Ii

p0pa1?1. na man lam ii.gamissali ti appamattaJ:arri me pii.parri kata'!J, kadi'i
ela:FfJ vipaccissarf ti evaJ?t pJJXll!l nJvajtineyyii. tt aJtho.
The prefix ava- in the v.l. mivamaiinetha must have been ~en over from the
cty. Udiina-v 17.5--0 (B) niilpa,,. manyeta piipasyalpu~yasya; (N) niilpa-
mannyeta; PDhp 193--94 niipp0'1' piipassa!pu'1'iiassa manyeyii. GDhp 209-
10 na apu mane 'a pmusa/puiiasa is ambiguous.
It sccJDS clear that the Udana-v and PDhp traditions assumed that the
equivalent of mappamaiiiiati which they inherited stood for ma appa,,.
maiiiiati, rather than mii apamaiiiiati. (N) must go back to a compound form
similar to the Pali, whereas (B) and PDhp have resolved the compound. Rau
(p. I 65) suggests reading miippa'1' maiiiietha. Cf. appabodhati in l43A.
For the violation of the two morae rule in miipp-, see the note on 70.
Dhp-a III 16,17-18: evcvri biilapuggalo thoJ:athoA:am pi piipa'1' iicinanto
~ laronto v~4efhento pOpassa piirati yeva n· attho.

122. Dhp-a III 20,t7-t8: dhiro pa~<jiwpuriso.

The meaning~.. is assured by the opposition to biilo in 121.

123. Dhp-a III 23,2t foll.: bhayan Ii bhiiyitabba'1' corehi pariyu!lhitatta

sappafibhayan ti auho, idllI{I vurta'!l hoti: yath.J mahadhano viil}Jjo appa-
sattho sappafibhaya1'[1 magga1!1, yathti ca jivitukamo ha/iihalarµ visa'!1
parivajjeti, eva.J!J pal)c}ito bhikkhu appamanakani pi papani parivajjeyyii ti.
I take bhaya,,. magg"'1' to be a split compound "a road of fear/danger", with
-nt- m.c., to give the - - - ··(pachyii) cadence. See the note on 49.
For mahaddhana. showing a development from_mahat- rathe;r iban_ ~gh_lJ:­
(i.e. < •mahad-dhana), cf. mahapphala 312 356-59; mahagghasa 325,
and see GD II, p. 186 (ad Sn 191).
124. Dhp-a Ill 28,t4-15: nassa ti na bhaveyya, hareyyii ti atiharitu'1'
s~exva-
For the sandhi 9f niissa in pada a and nabba[Ull!J in pada c, which violates the
two morae rule, see the note on 70.
 Notes 93

Dhp-a Ill 28,19-20: abba~atp pani'f' visa,,. viya nassa cfttatp jxlpa'f'
anugacduuf n·.
The cty seems to be ta.Ung anvt'.li in the sense of "follow". For this Dl?llin&
see the note on 1-2. For anvetlm the sense of"enter", see CPD (1.v. 1UrV<t1)
and GD IL p. 298 (ad Sn 770).
For the consonant group -nv- in anveti in p3.da c, see the note on 1-2.
For assa < 'asyat < syat (• siytl) see the note on 10. Although the cty
correctly explai/1$ this (see abov<), it also seems to be taking it aa asya "of
him".

125. Dbp-a Ill 34,5: pacwf ti p<>!ieti.

The cty cxplairu pacc-<li by the miassimilated form pa[i-eti, which Fausboll
ecceptM, to give a nOimA! elevcn-syllable Trl>tubh pada. This is unnecessary,
since this is the variety of the Tfii!ubh pada where the short sixth and seventh
syllables are replaced by a long .syllable. See the note on 19-20.
For posa aee the note on 104.
Although suJ:Jiumo includes a narabhakti vowel, it differs from the other
words including svarabhakti vowels, in that ignoring the vowef does not
return the con>0nant group to its unresolved Skt form I therefore prefer to
classify pi!da d as showing resohn>n of the fmt syllable.
The lmt edition of OvH/KRN (1994) wrongly stated that this verse wa. all
Tr41ubh. Pi!da a is Jagali.

126- Dbp-a ill 37, 18: gabbhan tHdha manussagabbh o va adhippeto.

There are nine syllables in pi!da ._but assuming resolution of the fillh syllable
would leave the unacceptable adence - - - •;see the note on 7. We could
perhaps scan eJ:l, and assume reodution of the fourth syllable. In pi!da d there
is resolution of the fmt syllable.

127-28. For the eastern C/-1-r- mantaliille, see the note on 91. For eastern
forms see the note on 32.
Although, as a general rule, an obsolutive is to be taken with the subject of a
sentence~ here we must assmne that pa . .·issa is to be taken with an
unexpressed instrumental goinj: with vijjati: "that place is not found (by
anyone) entering an opening"_
In pilda c vijjati is m_c.

127. Dhpia III 44,10-11: na so ~a/aggamatto pi ol:iiso atthi yattha {hito

papakammato muccftw/t salrho;qyii ti attho. .
We should read mucceyya formWiceyya in pada d, since the passive sense is
required, as the cty makes clear_ fur the confusion between muiic~ and mucc-,
see PED (s.v. muiicafl). For the NC/CC alternation see the note on 19.
The metre is T~tubb._
 94 The }Vord ofthe Doctrine

128. The metre is Tri~!ubh. For the procliric use of nu before ppasahetha in
p5da d m.c. to give the break - - - , sec the note on 8.

129-30. Dhp-a III 49,J-J: sabbe tasan!i ri sabbc pi sattJ attani da1J.¢e
nipalante tassa dar;ujassa tasanti. maccuno ti maral)assa pi bhayanti yeva.
In Skt tras- is constructed with the ablative, ge:iitive, or rarely the
instrumental.
Dhp-a III 49.12-14: na haneyya na gh5Iaye ti yathii aha'!l eva'!l aiiiie pi satta
tin ' eva parC!J!l paharq-ya na paharCpeyyii ti allho.
We could also transla!e p3:da c as "comparing others with oneself'.

13 0. Dhp-a ·III 50.14-16: sabbesaf!l jivilaf!I piyan Ii khil}Gsave {hapetv<i

sesasattQnaI[1jivita0 piyal'(J madhuraT[l. khi!J.ii.savo panajivite va mar/Jl}e wi
up~ako va hoti.

131-32. For the present participle ending -Gna in esiino, seethe note on 20.

131. There is resolution of the founh syllable in piida d. The scansion is due
to the fact that the pada is the negative of !32d. Cf. the note on 7.

133. Dhp-a lll 57,1&--20: kaiici ti kaiici ekapugga/am pi pharusaJ[I ma avoca.

vuttii ti tayfi pare dussi/J ti vutta tam pi ta th' eva pafivadeyywr1. siirambha-
k.athii ti esa kiiral}urtarii yugaggiihakarha niima duillii, pa{idal)fjii ti kaya-
dal)rjadihi parOJ?i pahart1J1tassa tiidisii pafidaJJ.efO. ca lava m.anhaJ:e pateyywp..
I follow the cty in taking kaiici as referring to the person spoken to (a. in
408), and pharusam as an adjective describing 'the thing said. The v.l. kiiid
~ would agree with pharusW[J. ·
The cty explains the negative injunctive mii voca by ma avoca. For other
negative injunctiv~s see mii samOgailchi 210 (explained by an optative); na
randhayw,, 248 (explained by an imperative); mapiidi (?) 272; mii upaccagii
315; ma bhaiiji 337; mii piimado, mii gili, mii kandi 311. For an optative
being explained by a negative injunctive, see the note on 27.
For phusati in the sense of an external object, sensation or discipline
impingeing upon an individual, cf. phwrha 83 and see Norman, ! 962, p. 326.
In p3.das bd. the-omision of -1!1 in -e;.yu is m.c., to give the cadence w - w •.

134. Dhp-a III 58,3-5: sace neresi ti sace attiina1r1 niccala'!t kiitu'!t
sakkhissasi ka'!'J.SO upahato yathii ti mukhava{!iyaT[t chinditvQ talamaltGT!J
katva fhapitakaf!tSathiifaf?1 viya. tOdisa'!l hi hatthapiidehi va dar;.!faJ:ena va
pahatam pi sadda'!I na karoti. -
The sen.Se of sace neresi (= na + frest) a11anal]J is not so much "if you do not
make yourself utter a sound" as "i(you stop yourself making a sound". Le ...if
you keep silent".
 Notes 9)

For esa with the sa::ond pasoo verb si, cf. GD II, p. 163 (ad Sn p. 15,ll ). CL
rn with second person vabsio 236 238 379_
!)lip-a III 5 8,9-12: siiramhho l! n.a vijjati ti ei'l:l''!J santt: ca pana rvarr1 dussilo,
tun1.he du.ssila ti ewi.mlidiko llLJarakarar;avDciilakJh.aM sOrambho nn vijjat1
na bhavissafi yel-0 ti atiho.

13 5 _ Ilhp-a Ill 59,20 foL: pacefi Ii chdn gapalaiw kcdiiranlaram

puvisantiyo glim darµ/,ow. wirervQ ten· elU potf.ento wlabhntin.odakn'!1
gocara171 neti. O}'W'{I pdcadi tijivjtindn"yaf!1 chindanti fupenri.
There was probably a word-pby on pac- and paj- in an c:uiicr version of this
vcrsc, presumably in a dialo:t where inlavocahc --c- and -j- lxnh bcc.Jrnc -j- or
-y-_ Sec Lildcrs (1954, § 140)_ 10e cty docs oo< undcnwid thi word-play
For pacenti (< pra + a;) "!omve" in pada b, sec Ovll, 1981, pp. 82 I -22_ Jn
p'id;i d we have the ausativcof pac-, "to bring IO an md" (sec MW, s_ v_ pac).
For another worrliJlay basOtl upon the same dj alu:rrution. sec the note on
44-45. See also the note 009-IO.
Mchcndale (1955-56B, P- 161) suggests th:lt gocaram ' gocariil, i_e_ tS an
ablative "from the pasture". For the ablative in-lll!f sec the: note on 49.
For paninaf!! m.c. in pada dscc Gcigcr, § 95.L

136. Fora/ha ~"but",seclS_

13 7 _ There arc nine syllabics in pada. e. We could correct the mrue by

reading dasa[n]Mf!!, and ;muming rcsollllion of the first syllable, or by
eliding the final syllable d"'""1![am],Pada c Jil<nlly means "to one st;Jte of
the ten", which asesetollliol38-39.

138-39. Udana-v 28.26--27 has va or vapi consistently, except for 27d

abhytikhyiintlT(JC1J dnru!Jtm,, and 28b Sarirasya ca bhedanaf!l, where vQ
v,:ould not be possible in the--¥" cadence. In 138d va is m.c. for vO, and it is
probable that vain 139abalis also for va, although only in pada bis it m_c_
ln 138a ca is probably for• (despite Udarra-v ca), whch is m.c. for wl_ For
the dv alternation= the-on 50.

13 8. Dhp-a Ill 70,18: bhetl.tan ti hatlhacchediidikam sarirabhedanam.

13 9. In pada a -ss- in UJX1SS11Cgam is m.c. to give the - - - • (path ya) cadence_

t
140_ ln pada d we could IDCI so upapajjati with Dhp-a and H, and assume
resolution of the first sylblilc. The parallel vcrse at It 26,10· has saggam,
where the metre allo'W'S so ~jja ti.
For piivaka used alone as UJ110nym for aggi. see 71.
In pada b aggi is m_c. to noid the opening· - - • _
96 The tVord oft he Doctrine

UJina-v JJ. has the compound rajoma!a "dus"t and dirt" in place of rajo va
ja/Ia'!1. which avoids the redundant fifth syllabic in p5.<la c. For rcdundJ.nt
syllables, sec the notes on 177 329-30.
Dhp-a III 78,6- 7: affhavatthukilya kari.khiiya aviliQ~alhiivena aviti!)l}O-
kankha'!l.
The metre is Tri$\ubh.. In pJda a we should ignore the svanbhakti VO\vcl in
-car"ya. In p5d.'.l c va i~ m.c. for va.

142. Dbp-a Ill 83,17: alal!lknto ti vatrhiibharal}apafimal}<;fito.

Here ...adorned" means adorned in fine clothes, as opposed to wearing ascetic
garments.
Dbp-a III 83,22 foll.: so cvarUpo biihitap. _Jatlii brQhmal)O ti pi samitapJpaaii
samw;o ri pi bhinnakilesattil bhi1Jc.hU ti pi vattabbo yeva ti.
The cty gives derivations for brahma!)a (< btlhita), samal}a (< samana) and
bhilliu ( < bhinna). Sec the note on 267. We should. however, note that the
verse supports only derivations from soma. brahma and bhiita.
PED says that sama is< iama, but PDhp 196 has sama here, GDhp 80 has
dhamu and Ud.Jna-v 33.2 has dharma1!J. See also the note on samacariyii
388. Contrast the note on 265.
There is an error in the footnote in OvH/KRN. This ve;;:se is the equivalent of
PDhp 196. not I 6.
The metre is Tri~p.ibh. Note the opening - - - - in pada b.

143A-44. The division of these verses in OvH/KRN does not agree with
some other editions.

l43A... This verse is found at SI 7,22•-23•_

Dhp-a ITI 86,I0-11: attano uppannarp ninda0 apaharanto bujjhari ti
appabodhan·.
Spk I 37,14-t5: yo nindan, appabodhati ti yo garaham apaharanto bujjhati.
The cty's explanation of appabodhatj as apabodhari < apa-harali bodhari
seems rather forced, and it might be bener to follow Rau's suggestion (p. 167)
oftalcing it as appa(m) bodhati "know ninda to be a little thing". CPD (s.v.
apa-bodhan) th.inks that the cry·s mention of apaharati supports a belief that
we· are dealing with apa..bodhati, which it translates as ·'to ward off, to
prevent. keep clear of'. Perhaps in support of appa- is appamaiiiietha in
121-22. Udiina-v 19.5 has sCirYapiipaT(ljahiity '-iO-
Dhp-a· III 86,8-9: koci /okasmin ii evarUpo dullabho kocid eva loka.smi'!I
vijjati.
The cry is taking koci as ... Is any man found in the world"~ and this is
supported by Spk I 37,13.(ad S l 7,22'): kcci evariipo toke vijjatf ti pucchati).
Nevertheless, I take koci with loka.rmi in the sense of kvaci "anywhere in the
world". Cf. koci toke 179, and see SI 59,tS (na kho diini d1Nassa koci diinam
diyati; Spk I 114,25: koci ti kat1hac1); Ja IV 92,7& (kocid EYG suva.ry~aA:ayuril
 Notes 97

nJgarGjOI!f bharanti pit;cj.ena; 92,17·: kntlhacid eva gJme vJ nigame va); VI

89,2r (n' ev' amhO.kn'!l bhayaf!1 k.oci vane viifesu vijjati; 89,.2•': imasmi'!l
vane kanhaci ekapadese pi ... ; cf. Sadd 305,Jo (l:oci ti kvaa)); Bv-a 267,22
(na kvaci parihayCimi ti ... na kaci parihayiimi ti pi pli/ho,so ycv' attho).
For the consonant groups -sm- in -asmi in pJ.da b, and -dr- in hhadro iii pada
d, see the note on 1-2.
In p3da a hiri- is ITLC. In pJda b the Joss of -'!1 in Iokasmi is m...c. Cf. appasmi
224; vanasmi 334. For kasJm m.c. in pJda d, cf. ajj_atan.Gm 1.27 and see EV
Il, p. 145 (ad Thi 392) and GD II, p. 285 (ad Sn 712).

143B. This verse and 144 occur together in PDhp 329, giving a mixed
Tri;\Ubh!Jagati verse, with six padas, but they are separa\ed in Udina-v 19.1
~L .
Dhp-a lII 86,16-17: l:asaya nivi{tho aham pi niima l:asiiya pahato ti apara-
bhiige iitapp"'?' karoti.
If nivifflui is the correct reading, then we could assume a derivation from
nivfy/a "entered (the influence of) the whip", as we say "the horn: is under the
whip". The readings abhispr;/a (Udina-v 19.1; paralleled by abhitaefita in
192) andJ!U!!ha (PDhp 329), however, suggest that-nivi!lha is a dialect form
for •nipi!!ha < nispr;/a "touched by the whip" (with the change of -p- > -v-,
and the same loss of -h- as in pu/{ha), or posssibly < nipif(a "ground down,
tormented by", which was misunderstood. by the other traditions.
It is pomole that bhaviitha is a subjunctive with the long -li- in -&ha, but it is
more likcly to be an imperative in -atha with lengthening m.c. See Geiger,
§ 123 and Cailla~ 1970, p. 26.
For the oonsonant group -dr- in bhadro in pada a, see the note on 1-2.
·The merre is Tri;Jubb.

144. Dhp-a III 86,21: kiiral)iikQralJajiinanalakkha'l)ena dhamma-

vinicchayena.
Dhp-a 1II 87,I-2: icfmri an£Jppakmp va11adukkh0'!' pajahissatha..
The merre is Jagati
In pada a we should read viriyena m.c. Cf. viriya in 7--8 112 In pada d we
should read pahassathii m.c.

145. Dhp-a III 99,16-17: subbatii ti suvadii sukhena ovaditahbii

anusiisilabbii Ii attlw.
Although suhbata must be< su-vrata '"having good vowsM - "virtuous", it is
explained es su-vada here.-It is not clear why this should be so, when the cty
on 95 gives the correct explanation_ The equation subhata = suvada suggests
a transmission throllgb a dialect where the form was su-vrata > su-wada,
with + > -d-, probably written as suvada.
Cf. 80, where pa~efitii occurs in replace of subbatii.
As Rau (p. 167) points out, na_mayanti is the equivalent of ujJO!t. karoti in_33.
 98 The Word oflhe Doctrine

The metre is Vaitillya_

146. I translate hiiso as "laughter", but Udana-v L4 has harSaf:i, PDbp ha.So
and GDbp har;o, confirming that the tradition recognised that hii.sa CJ~ stJ.nd
for har!a "'rejoicing" (via• hassa), as well as hiiw.
Dhp-a Ill 103,5-7: imasmi1!J /okasanniv5.se riigadihi ekiidasahi aggihi r.iccaf!i
pajjalite sari.
Although I translate pa;jalite sari as though it were a locative absolute,
understanding "(the \vorld)", I presume thatpajjalita is a past participle used
as a noun "burning" (see MW. s.v.pra-..Jjval): "what joy is there in burning
(which is) constantly existing·?" For a past participle used as a noun, see the
note on 104.
Dhp-a III 103,8-1 J: affhavatthukena hi avijjandhakiirena onaddhii. lumhe
ta.rs· andhakiirassa vidhamanatthiiya ki1?1karal)ii iiii'}apadipcr;i na
gavessathii na karotha.
It is not clear why gavessatha has -ss-. since it is not required by the metre. It
was perhaps thought to be a future fonn < •gave ti. The inclusioD. of
ki'!'kiiranjj shows that the cty undmtood piida d to be a question.
In place of onaddhii Udana-v 1.4 bas andhakar&!t pravfy/Q~ stha prad:pa'!l
na gcrvqa1ha. GDhp 143 has palqili, PDhp 233 prakkhillii, and Mvu Ill
376,13• praksiplii.

147. Dhp-a III 109,6-8: cittakatan ti !atacittarri varrhabhara1Ja-

mOliilattakamhi vicittan ti anho. bimban ti dighadiyuttaffhQnesu dighi5dihi
arlgapaccaiJgehi sal){hitcuri attabhifvwlt.
Since Sk! ciJ:rclriya m=.s ''painting", I translaie ciltakata as ''painted" .
.... Dhp-a ill 109,12-13: bahusafrikappan ti mahlija:ne:na bahudha SOJ?lkoppitW[J.
· Ps l!J 302,&-!2 (ad M II 64,26"): bahu.rankappan Ii paresa'!' uppanna-
patthanasaitknppehi bahusaitkappa'!l. illhinClJ!I hi kiiye purisiina0 saf.T..appiI
uppajjanti tesa'!I kaye itthinClJ!I susiine chacj.fj.italalebarabhU.taT(l pi c · elaf!l
ktikakuladayo pa11hayanli yevii Ii, bahu-sailkappo niima holi.
Dhp-a lll 109,13-14 "'Th-a III 35,14-16 (ad Th 769): yassa n · allhi dhuvam
Jhiti Ii yassa dhuvabhGvo vQ {hiribJiavo vQ n' atthi, eknntena bhedanavikiTC!}O-
viddha'!fSanadhammam eva.
This seems to be taking dhuvlllfl and [hiti as two separate nouns.
Ps III 302, 13-15 (ad M II 611,26•): yassa n '01/IU dhuvW!l fhili Ii yassa kiiyassa
mayii man-ciphel)a-udakabubbufQdina'!t viya ekartJsen' eva {hiti nG.ma n ·
atthi, bhijjanadhammatii va niyalii.
Here dhuva'!I is being taken as an adverbial accusati~e. the equivalent of Skt
dhruvam "cenainly'. See MW, s.v. dhruva.
Udana-v 27.20 has dhruva-slhitih, and it would seem that dhuvam lhili is a
·split compound, or rather a comp9und with -'!'fh- for-/[h-. See EV i. p. 236
_(ad Th 769), and cf. a~um1hU/a 31 and see the note on49.
 Notes 99

148. Dhp-a III 110,18-20: ida'!f lava san-rasarµkhiilaf!I rUpaf!! maha//aka-

bhiiw:na panjil;r:i.a'!l laii ca kho sabbarogO.naf!J nivOsana[fhiir:u{!hena roga-
nh!eflra'!f.
Note: nirjef.a and nirjq/ha - Skt niefa, with the VCC/\1C alternation.
GObp 142 readspravhaguf}o with the same ending as Dhp. The ending of the
v.l. pabhailgura'!I agrees· with PDhp 259 prabhaf!!pira'!I and UdJ.n.a-v l .34
prabh.ailguraf!I.
Dhp-a III l l l.J-4: so eso pUtiko samiino tad' eva deho bhijjari na cirass' eva
bhiffissari ti veditabbo.
The cty talces bhijjali as the equivalent of the future. PD hp has bhijjihi1i;
GDbp bhetsidi and Ud3.na-v bhetsyate. For the suggestion that the correct
reading is bheccha1i, see Brough (p. 217 ad GDbp 142).
In p>da c we could avoid the opening···• by reading bhijjati or bhijjate m_c.
See the note on 1-2.

149. Dhp-a Ill 112,14-16: apa//honi Ii cha4efi1ani, sarade Ii saradai:Zi/e

viita!cpah.atiini tattha tattha vippaki!J.r:.ia-aJQpiini viya kiipotakiini ti k.apotala-
vaJ?'!iini.
For imiini in the relative clause, see the note on 42
In pida b note a!Jpii.ni + iva > aliipiineva. This may be an ex.ample of the
change f> e (Le. a!GpUneva < aliipUnfva), or of eva in the sense of Iva (see
Noonan, !979E, p. 392). See Brough (p. 226 ad GDhp 154), who quotes
Faosboll: "eva pro iva". Contrast aggi va 31, which represents the sandbi of
aggi + iva. and see the note on 66, where eva is the sand.hi of -a + iva, rather
than an example of eva.
Foc-p- ~ -b- in aliipunl < Skt a/abu, cf. pabbaja 345, and see Geiger, § 39.6
and Brough (p. 226 ad GDbp 154).
Foctlie inclusion of the antecedent imiini'in the relative clause, see the note on
42_

150. Dhp-a III 118,12 foll.: yath' eva hi pubbar:.il}apara!J.1J.3dina1!J

odaAanatthGya kaffhiini ussapetva valtihi bandhitvQ mattikiiya limpitvii
nagarasa'!lkhiila'!J bahiddhii geha'!I karonti, evam ida'!l ajjhattikam pi fir:.ii
afµ,isatiini ussiipetvii nahiiruvinaddha'!I ma'!Isalohitalepana'!I taca-
paefcchanna~jiral}alakkha1J.liyajariiya, maral}QlalJ:hOJ).O..Ssa maccuno, miino
maH:ho ca ohito ti tassa iirogasampadiidfni pa(icca maiifianalakkhal)assa
rnfiztassa sukatakiirqf}avinOsanalakkhal}assa maillassa ca odahanatthaya
nagaTO'!' latm,i.

151. Dhp-a III 123,2-5: sataii cii ti buddhiidinmri pana sanffilUlJ!I navavidha-
lohtttariidhammo ca kiiici upaghiita1!1- na karoti ti na jaraf!J upeti nama.
]XIPtdayanti 'ti eva'!1 santo buddhiidayo sabbhi parµjitehi saddhil!l kathenti ti
attlto.
 100 The 1Vord of the Doctrine

In pada d sabbhi is the dJtive of giving, not Lhe instrumental. See Brough
(p. 227 ad GDhp l 60). L'.dJna-v 1.28 has a locative: santo hi ta'!' satsu
nivedayanti.
For ca= tu ..but", see the note on 5.
The metre is Tri$\llbh.

152. Dbp-a III 126, S foll: appassu1iiyan ti ... appass-.J!O ayal[l.

For the sand.hi -o +a-> -0-, see the note on 56.

153. Dbp-a III 128,&--7: gahakarakn'!l gavesanto ti aJ-..a~ imassa attabhiiva-

gehassa kara~ tal)h5varjefhakirri gavesanto.
For "house"~ ..body'', see GD 11, p. 133 (ad Sn l&-19).
Dhp-a III 128,11-12 "Th-a I 182,19 (ad Th 78): anibbisa'!I la'!I iiiinam
avindanto alabhanto yeva sandhiivissa'!1 SOf!lSOril!l apariipara1]1 anuvicarirri.
If anibbisGf!J is to be derived< Skt nirviS- with a negative prefix.. it should be
noted that MW (s.v. nirvi.S-) gives no meaning which would-support the cty's
explanation by vindanro and labhanro. Udana-v 31.6 has sQ'!Jdhiivitva punah
puna};. If punah puna~ is a genuine equivalent to anibbisa1!1~ it implies a
meaning such as "not ceasing, not puning an end to"". I would suggest that
-bb- is m.c., to give the - - - ·cadence, and anibbisa'!' stands for nivisal[I <
niviS-. in which case we might compare Skt anivi.Sam3r.a ..not retiring to rest,
.restlessn (see MW, s.v.). We could therefore translate ..not resting, without
respite". For the translation ''without expiation" which I previously suggested
for anibbisarri, on the as_surnption that it is the equivalent of Sk:t anirveiam,
see EV I, p. 141. Cf. nibbisa < Skt nirveia Tb 606 1003.
It seems that all editions read sandhiivissOJ!l, despite the explanation as a past
--. tense. For -ss- , which is not required by the metre. see EV I, p. 141.
In pada C there is resolution of tb.e first syllable.

154. Dhp-a III 128,21 folL: gahakU[a'!l visa'!lkhitan ti imassa tayii !-.:Jtassa
attabhii.vagehassa avijjiisa0khiitar!l kal)ryikama!J¢a/af!l pi mayii
viddhal[lSila'!J, visGFflkhG.ragata'!l cirran ti idii.ni mama cifla7r1 visaJT>k.~iirGJ!f
nibbilna1!1 ii.rammGl)Gkaral}avasena gatGr?J anupavi[!ha7r1, lal)hiina0 khayam
ajjhagii. ti tal}hiina0 khayasOJ?!khiira'!' arahallClr!1 adhigato 'smi ti.
For the future in -h- L., kahasi, see the note on 236.
For visa'!'J:.hital!J and vjsa0khfira, see BHSD, s.vv. visa1!Jskrra and
visGJ?lSkiira.
The cty, it seems, takes pada f to be separate from p3da e, and undersL3Ilds
ajjhagii to be first person singular. "The mind has gone to non-sankhara-ness
(= nibbiina); I have attained Ibe destruction oflhe cravings".
There are nine syllables in pada c. We could normalise the metrC by reading r~
for ere with th_e_v.I.

~55-156. We should ignore the svarabhakti vowel in -car1yal[l l·n pada a.

 Notes IOI

155. For pal/ala, cf. 91. For eastern /Ir see the note on 32.
In Sia krunca and baunca means "curlew" and the latter also (lex.) "osprey".
Jt would appear that the \\'Ord is onomatopoeic, and Skt kraunca an~ Pili
koiica can therefore be used of any animal which makes the no_ise i:uiica., cf.
kofica-nllda used of elephants (see PED, s.v. kcnca'). Although kofica may
sometimes be the crane, in _this fish-eating context it is more likely to be the
heron, as Wright (quoting Thieme, 1975) states ( 1996, P: 60).

156. Dhp-a Ill 132,24-25: ctipiitii:f.ina va ti capato atii:f.ina clipti vinimmunll

ti attha.
For the suggested development of arikhfr;a < •ariskirl)a (- •atistin;.a), see
Neiman. 19798, pp. 325-26. For capiito in the cty, see the note on 320. ·
Dhf>-a Ill 133.4-7: purGnQn/ anutthunan ri iri amhehi khiidita'!' iti pitan ti
pubhe katdni ildditapitanaccagitavii.ditGdfni anutthunantii socantd
anusoconM soui.
The singularpiesent participle anutthunam with a plural verb seems strange.
We might rcprd anutthunarrz as a l)tlmul type absolutive (see the notes On
201 325, and GD II, p. 299 [ad Sn 773)), or a present participle
singular taken over from a differen~ context.. where the singular was more
appropriate.

157. Dbp-a ill 138,t-3: yaman ti Sattha attano dhammissaratiiya desanii·

kusalatliya idha ti]Jl)t1'!1 vayii1lDJ!1 aiiiuilararr: va;-·mri y~n ti katvii dassesi.
PDhp 312 tli1Jl!D1t1. aiiatarai!f ytimiinal!l; Udana-v 5.15 frayiinfil{i anyataraf!J
yii~.
With the vctb pa!ijaggeyya, one would expectyiima to have its usual meaning
of"watch of the night", but the cty explains that it refers to the three stages of
life.
For the combination of (ti- )yiima and pa[ijaggati, see the note on 60.
For the prolcptic use of the cognate accusative suralckhitQI;l with railleyya,
cf. 168 169231-33 296-301 and see EV I, p. 210 (ad Th 560). For the
use of the cognate nominative, sec the note on 159.
For 1UJ1TI., sec the note on 1-2.

158. Dhp-a Ill 142,l foll.: patiriipe nivesaye ti anucchavike ~e

patiffh1ipeyya •.• attGnam eva pa{hamatarGIJ1 tasmi'!f gul)e pati{fhfipeyya,
eva'!1 patiffhiipetva _atha a.iiiia'!t tehi gul)ehi anusiisitulcamo so attanam eva
pa{h.amatara;' t'::mi'!1 gu]Je pati!{hGpeyya, -eva1!J·patif!hiipetvii atha aiiHmµ
tehi guIJehi mnaaseyya.
Here mi is the emphatic particle: "first one should establish oneselfin ... ".
Dhp-a ill 142,6-10: attiina'!' hi tallha anivesetvii kevalalfJ param eva
anusiisamano parato nindaT?t laf?hitvii kilissati n5ma, tattha attJ.n.mri
nivesetva anuslisamfino parato plllGJrtSa1!1 labhati, tasmii na kilissa'ti niima,
evam /caronto JXDJ4ito na kilisseyya.
102 Th• Word of th• Doctrin•

PED translates kiliseyya as ..do wrong". I presume that it is rather "become

impure, become stained (as a result of acting wrongly)".

159. In pada a I translate a as "if', but it is possible that we should read ve

force, since there is no reference to ce. in the cty. Cf. Udina-v 23.8 iitmlinaJTI
hi and PD hp 318 Jttand y<(where y• is identified as an emphatic particle by
Cone, referring to Nonman, 1967, pp. 161-{)3; see also the note on 165). No
v.L ca is given for the text, but ca is found as a v.L in the explanation in the
cty. The reading dttand in PDhp seems to make better sense: "One should so
act oneself as one advises others". See the note on 162. For the va/ca
alternation, see the note on 50.
Dh~a Ill 144,5-12: yathd hi bhiillu pa/hamaylimdalSU cankamitabban ti
vatvii para'!! ovadali sayalfl can.kamadini adhi_t{hahanto attiinaii ce tatliii
kayiid yath' aiiM111 anusasati. ev!lJ11 santt. sudanto vata dametha, yena
gia;.ena paraf!t anusDsati tena at/and sudanto Fiu!Va damryya. attii hi kira
duddamo ti ayaT[J hi atta niima duddamo, tasmd yatha so rudanto hoti tatha
dametabbo ti.
This seems to be a use of the cognate nominative su.dcmto with dametabbo:
"(the self) is to be tamed so that it is well tamed". Cf. so hi niina bhante
suvimutto vimuccissati, S Ill 121,.26--27. For the comparable use of the cognate
accusatlve, see the note on 157.
1assume that an active verb is required, i.e. dametha. Dhi:ra has dameyya in
the verse and the explanation; PDhp has dameyd; Udana-v changes the
construction and has no verb in the piida_ The reading -mm- in J (misprinted
as dhammetha in OvHIKRN) is either m.c. to give the patlrya cadence, or is a
fourth class verb. I think that piida c is a continuation of the sentiment in the
fest line: "Only when you are tamed yourself should you tune others", and an
lnterpretation of dammetha as a passive form would be inappropriate.
In piida a we should ignore the svarabhalcti vowel in kaJriL

160. Dhp~a Ill 149,l foll.: atta va attano pat(tJhG. ... attanii. eva hi sudantena
nibbisevanena arahattaphalasGJ!lkhalGJ!I dullabhwri n&Ju11!1 labhari.
C&P read va for hi in p3.da c. and attana va for attana eva hi in the cry. I
assume that hi in padas abc is the emphatic particle. PDhp 321 has hi in all
three padas; Udana-v 23.11 reads rv ___ nu ... hi.
PDhp 321 has sucil)l)ena fat sudantena in pad.a c.

161. Dhp-a III 151,2 foll.: .vam eva'!I _atf{ln_J_kata'!I attani jdtam
attasambhaval{t p[ipairr.
The parallelism between attaja and attasambhava makes it certain that both
are adjectives.
Dhp-a lll 152,2: abhimatihati kantatf vidhQI[lSeti.
PDhp 307 reads anuma'!ldhati dummedha'!I; Udana-v '28.12 reads
abhimadhnJti tam pdpa'!I- Although the verb is quoted in the form
 Notes 103

abhiMaUlr.ari at Sadd 365,15, it is perhaps better to read -manthati. See Geiger,

§ 53-1.
Note the v .I. asma- for amha- < aJman.
We 1hould ignore the svarabhakti vowel in. vajirCIJ71 in plda d~

162. Dbp-a III 153,!0 foll.: mQ/uvii siilam iv' otatan ti yassapuggalassa ta1"(1
tar:ihfisankhJta'!l dussfiya'!1 yathii niima m<i/uv<i s<ilam otarantf
sabbatrhaJ:am eva pariyonandhati eva'!' otatal'(J pariyonandhitvii [hi/OJ!I.
MW lists (lex.) mlih.;vli "a l::ind of sweet potato'", but this cannot be intended
here. for mllluva see GD II, p. 201 (ad Sn 272) and Liiders, 1954, § 9 I. MW
lists mLilu (f.) .. creeper", and there was presumably a fonn •mQ/uk.5 based
upon thU. which developed> •ma!uya. Pali inherited an eastern form '"';th
glide-v-. from 'mll!uyiVrnliluvll Udi!na-v and PDhp back-fonned mlifutii (cf.
PDhp 137 and Udana-v 3.4 ~ 334); i.e. k > ylv >I. For the glide consonant
altanation v/y, see the note on 40.
The ctysten.! to be talcing otata as active, and if that were possible, then the
pi!da pi=t5 no difficulties. if we assume that otalGJil agrees with dussilyGJ!1.
MW quoe< "spreading over, extending to" for tata (but only on the authority
of Wilson), and gives no information about it being able to tal::e a direct object
in the accusative case. PDhp 306b has the same construction: ma1utii slilam
ivo/a/J, with otatii agreeing with miilutil "'like the creeper spreading over the
sat tree"; Udi!na-v l l.10 (B) has slilav(hrt miilutii yathil ("like the creeper
pOSS<SSing the sli/ tree"?) and (N) siillV?' va mli/utatatam "like a sli/ tree
ovc:rsptad (covered) by a creeper". This is the most grarrunatical version but
may for that reason be a normalisation since, to get a parallel, we should
exp<d padas ab to mean "whose bad conduct has spread over (him) as the
mJluvO aeeper spreads over a sal tree". A reference to the same simile is
found in Ja V 452,26•-21•: rattacittam (so read) atfve(hayand nGJil I siila (me.
in the Rathoddhata metre) mli/uva/ata (-va- m.c.) <va> kiinane.
GDhp330 has malu·a va vi/ad.a val}-i, but this seems to be nearer to Sn 272 =
SI 207.Js•: maluva va vitatii vane, which would probably be a better simile
for the dussilya spreading out, while 162b would go better as a comparison
to pudmvisatta J:amesu.
Dhp-a ID 153,17-18: yatha naJ!I anatthakiimo diso icchati tatha att5na'?1
karoti lliima.
There is a v.l. attana for alliinarrr. Cf. the note on 159.
For the consonant group -ly- in -dussilytv?I. see the note on 1-2 and forthe
long vowot before a double consonant violating the two morae rule, see the
notcoa70.
for._ see the note on 1-2.
Theicm: nine syllables in pada b. Fausooll reads miilvii, as though -u- were a
svarabhakti vowel which can be disregarded. Cf. 334, where the metre
requires the -u-. In pada d icchati is m.c.
 104 The JVord of the Doctrine

163. We should expect a contrast between padas ab (sukarani) and pM.u cd

(-dukJ:aro111). and we might suppose that in pJ.da c ve"" ce <ca "but''. &:e the
note once in 229. If this is so, then the change is a very old one. PDhp 167
has ve ... ve; Udana.v 28.16 has vai ... vai; GDhp 264 has du (Q tu, w'iich
·would support the idea of ca) ... gu (~ kho < kha/u). Forthe dv variation see
the note on 50.

164. Dhp-a Ill 156,17-21: tassa laf!I pafikkosanaf!! 9J ca pJpi/W di[(hi

ve/usmllhiilassa kanJ:akassa phaliini viya hori, tasmii yathfi ka/thalo phaliini
ga1Jhanto attagh.aiiiiaya phallati attano ghiitaaham eva phaleti, e:YCIJ!l so pi
attagiu'itaya plwl/ati.
The cty theri quotes SI 154,J•-4•:
' plwlOlfl ve kada/bp hanti phalllTf! ve/u pha!Of!I na/Of!!
.. sakkaro kfipurisQJ'[f hanti gabbho assararfrr yathd.
PDhp 315 reads ka~(a.i:asseva; GDhp 258 kadakaseva; Udana-v ka~/aka­
vo~ur. These readings suggest that ka{{haka is a mistake fOr ka~(aka. For a
discussion of this, and of the idea that the act of bearing fruit kills the banana.
bamboo, reed (and mule), see Brough (p. 255 ad GDhp 258). For the death of
the ka~faka after fruiting. see Rau (p. 169).
For the sandhi of-assa + i'va > -asseva, see the note on 66.
For -ghaiiiiaya (~ dative of purpose), cf. tanaya 288, and see Norman,
1989A. pp. 221-22.
In piida b we should ignore the svarabhakti vowel in ar'yanOf!I.

165. Dhp-a III 158,3-4:yenapana attana akataf!I paparr sa sugatin c· eva

agatiil cagacchanto attani1 va visujjhati.
' The fact tl"'1 va in pada a is ·the emphatic particle is confirmed by PDhp 308
and Udana-v 28.11 which both have hi. The fact that vain pada dis the
emphatic particle is confirmed by Udana-v atmanaiva and PDhp attana ye
vifujjhati. For ye a5 an emphatic panicle, see the note on 159.
Dhp-a Jil 158,.i-;;; kusalakammasankhata suddhi akusalakmnmasanldiiila ca
asuddhi paccattarrz karakasattiina'!l attani yeva vipaccati.
PDhp 308 has praccattam and Udana-v 28.12 pratyatmaf!!. See MW (s.v.
pratyiitmam), "indecl., ·for every soul, in every soul, singly'".
Dhp-a_ ill 158~6-8: aniio puggalo aiiiia'!l puggala7!1 na visodhaye tin' eva
visodheti na ki/issan· ti vutklJ!I ho ti.
For visodhaye as an optative, see 289. For visodhaye as a present participle,
see the note on 281.
For the violation of the two morae rule in niiiino, see the note on 70.

166. For sandhi -J. in atta·d-atthGf!! (twice); and sa-d-attha, see the note on
57.
There is doubt about the derivation of pasuta, and PED (s. v.pasuta) suggests
that it is < Skt prasita with -i- changed > -u-. The other versions do not have
 Notes 105

a direct equivalent here: PDhp 325 reads sadJ.uhaparamo siytl; Ud3n-a-v

23. l 0 svak.Jrthaparamo bhavet; GDhp 265 svakatha-paramu. As parallels to
181, however, we find PDhp 244 ye jhtinaprasutii dhfrii and Udana-v .21.9
ye dhyJ.naprasrtti dhird. These readings support the belief that pasuta is <
Skt pras[fa "intent upon, devoted to", with -r- > -u-.
Rau (p. 169) suggests th.at·there may be a word-play on sadattha-prasita <
sadarthaprasita and sv?i.rthaprasira: "one's ov.11 aim"= 'lhe good aim".

167. Dhp-a III 163,.5-8: hina'!f dhamman ti paiicakamagul).adhammaf!l, so

hi hlno d,iammo anramaso o(!hago~Qdihi pi pa/isevitabbo, hinesu ca
nirayQdisu /hanesu nibbatriipeti ti hino n&na, /aJ?l na sev0ya.
Udana-v 4.8 reads hind'!f dharmt11'[1 na seveta, which suggests that hinarri
dhamma,,,. may be an accusative plural form. For -aTfl as an accusative plural
ending, see the note on 64-65. PDhp 31 has hinam dhammaf!!, and l
translate as a singular_
Dhj>-a lII 163,10-12: loka-vaddhano ti yo hi evDf!! karoti (i.e. does the things
in padaS abc), so loJ:avaddhano fiama h.oti, tasmQ t:VGf!J akara~ena na siyQ
lokavaddhano.
The other versions have fonru comparable to lokavaefefhano: GDhp 121 has
lokavaefha~o; PDhp 31 iokavaddhano; Udiina-v 4.8 lokavardhana!J. !Urn
(p. 169) compares Skt bhiimivardhana, but I assume that /okavaefefhana does
not have the same meaning as bhumivaefefhana (see EV II, p. 140 [ad Thi
380]), Brough (p. 213) comments on the cty's "frank confession of
ignorance", and suggests that the compound is likely "to be a bahuvrihi, in
which case it would mean something like "possessing increase (~ properity ?)
in the world". I translate "worldly"-

168. Dhp-a III 165,1-2: utti{{he ti utti//hitvii paresalJl gharadv5re fhatvii

gahetabbapi~efe na ppamajjeyya.
This seems to be taking utti!{he as the locative of a noun "standing up {for
alms)". It may be the alternative form of ucchir,ha (see CPD, s.v. u//i(!ha), but
in view of the parallel versions (PDhp 27 u[{heyii; Udana-v 4.35 ulfiHhen) l
assume that it is an optative. For the verb,· cf. uf!hdnakii/amhi anu!fhahanto
280.
Dhp-a III 165,6-7: dhamman ti anesanGJ']2 pahGya sapadana'{f caranto tam
eva blrWJUjcariyadlumzmairi sucarilaf!I care.
For the proclitic use of na before pp- in pada a. see the note on 8_
FOr.thC Pi"oleptiCUSe of the Cognate accUSative suCaritalJI with care, see the
note on 157.
For the consonant group -sm- in asmi1!1 in pad.ad, see the note.on 1-2.

169. For the proleptic use of the cognate accusatives sucaritarrr and
duccaritalJJ with care, see the note on 157. For the use of the prefix du- cf.
duggahita-giihi DI 342,!5.
 106 The Word of the Doctri=

For the consonant group -sm- in asmirr in piida d, sec ~e note on 1-2.
For narr. See the note on 1-2.

170. Dhp-a III 166.10-12: maricilcan ti miiya'I' mdyd hi di!rato geha-

sa'l)/hiiniidivasena upaflhitdpl upagacchanta'!'afT2 upagacchantiina!J1
agayhiipak.a n·rrakn tLJcchakJ vd.

17 I. For the comparison with kings' painted chariots, see 15 I. For the split
compound with -rrr for cittariijarathii.pamaT(f, sec the note: on 49.
For the phrase etha pruso/ha, cf. eva passatha 344_

172. For c.a =-Ju "but", see the note on· 5.-
For the suggestion that so ~ sma, see EV 1. p. 215 (ad Th 87 !· Rau (p. 169)
sugg,esu reading ca instead of so.
·For the proclitic use of na before ppamajjati in pada b, see the note on 8.

173. Here kvsala is probably the equivalent of puiiii.a "ment, meritorious

deedtt. For the opposition of kusala to piipa, seethe note on 53.
In pada b pilhiyati (Udiina-v 16.9 pilhiyate) i! the passive < Skt (a)pi-dJ,a,
with th < dh. Note the v.l. pidhiyati. For the dhlth alternation see the note on
6.

174. Dhp-a Ill 175.&-7: tanuk" ettha ti tanuko ettha na bahuko anicciidi-
vasena vipassati.
For the sandhi of -o + e- > -e-, cf. 304. Udiina-v 27.5 reads tanuko "tra
vipa.iyaJ:al;.
Dhp-a III 175,11-13: appo kocid eva satto ~aggOya gacchatt sugati1?1 va
)ljb!Janar!f"Vii piipuT)iiti.
For saggiiya as the dative of goal of motion, cf. visuddhiyii 277-79; bhaviiya
vibhavdya 2&2; nirayiiy · 31 L

175. Dhp-a l1I 177,15-19: ime ~ii iidiccapaihe iihlsena gacchanti, yeslJr!1
pana iddhipiidii subhGvita le pi iikiise yanti iddhiyO, dhirii pi pa1)¢ita
saviihan(lJ?1. MiiraJ?1 jetvii imamhii vaffalokii niyanti nissaranti nibbii.naJ?i
piipu')Qllli ti.
The cty assumes that the verse applies to three class= birds, those with iddhi
and the wise. It is, however, not cJear whether pada b applies to the geese or
to the few of 174. It is possible that l 75ab is merely repeating l 74cd_ It
should be noted that if this is so, it is a. device of .the Pali redactor. The
equivalents of 174 and 175 are not together in the Udana-v, and there is no
equivalent of 174 _in PDhp. Plida b could, however, refer to the geese, in
which case there is a contrast between padas ab and cd, although it seems
strange to say that binls fly by means of iddhi. Plida b could also be taken as
referring to th.e wise in piida c. I translate the Piili as it stands, which leaves the
 Notes 107

matter uncertain, but I assuine that the subject of yanti in pAda b is not_ the
geese.
For Mara's wlhana, see Sn 436-39.
The cty takes nfyanti as< nir-ya, not as the passive of ni ·~o lead". PDhp 232
niyya'fltl dhrrd loi:amhi; Udina-v 17.2 niryanti dhirJ lokiin.
Here the meaning Wise" f<;>r dhira is confirmed by the gloss pal)cjitG.

176. Dhp-a Ill 182,26: el:a'fl dhamman ti saccam.

Dhp-a 11I 183,1-2: ·,,jtf~naparalokassJ ti vissa{1haparalokassa.
The meaning required for vithp)a seerns to be ..given up, rejected, abandoned",
as PED suggests, but MW d~s not quote this for Sl:J. vitirl)a. Note issar~na
vitir:zr:io at Ja TV 447,9• (issarena ti ranna vili'YJ-O sf ti pa.riccallo si. 447.JO").
For the mcarung "passed beyond", see 141. '
!n piida d the retention of -ti- in akiiriya"' is DLC.

177. Dhp-a IU 189,2: dhiro ti pa~efito.

The meaning ''wise" is confirmed by the contrast with biiUJ. in piida b.
For ca - tu "but", sec the note on 5.
For the proclitic use of na before ppasarrisanti in pada b, see the note on 8. It
is not necessary m.c.. since the break - - - is common...
The metre is TriW>bh. In pada a we should ignore the svarabbakti vowel in
lauJa,.Jyii. For the redundant fifth syllable in pada a, see the note on 141.

178. For the consonant group -vy- in pathavyii., see the note on 1-2.
I take saggassa to mean ·~o heaven", as a genitive of goal of motion. PDhp
338 has saggiinii'fl which seems to be a reflection of the 26 sorts of heaven
mentioned in Dhp-a Ill 191,26 foll.: saggassa gamanena va ti chabbfsati-
vidhassa va saggassa adhigamanena.

179-80. Dhp-a III 197,19 foll.: anantagocaran ti anantlirammal)assa

sabbafiiiutafi01)assa vasena apariyantagocara111- lena padena ti yassa hi
rligapadii.disu ekapada'!f pi atrhi !aJ?1 tumhe tena padena nessatha,
buddhassa pana ekapadam pin' atthi, !aJ?I apadmri buddhaJ?1 tumhe kena
pa.dena nessatha.
I take aJX1ilm!t to mean "leaving no track, no footprinr".
In pada d PDhp 276 has nehisi; Udana-v 29.52 has nqyati; Mvu lI1 91,20•
has nqyatha.
--- -- ·-· -- ---·-- -
179. Dhp-a III 197,14-17: yassa jita'!' niivajlyatf ti yassa sammii-
- - -

sambuddhassa tena tena maggena jita7!f rJjii.dikilesajiitaT{l puna

asamudacara7.1ato niivaftyati dujjitaJ?t nama-na hoti.
For jita'!' "conquest'" in plida a, see the note on 104 .
Dhp-a m 197.11-19: noyGfi ti na uyyiiti yassa]ilmtt JilesG]ii.talf? riigadisu koci
ekakileso pi Joke pacchatovatti n.Gma no hoti n5nu.lxuu/hatf ti.
 108 The Word of the Doctn·ne

In pada b jil<I7]1 might be the equivalent of vijirmrt."tenitory" ;but even so the

meaning of the pada is not obvious. Although assa can be translated as "of
him. his (victory)", this intrudes into the construction ofyassa ... tam, and we
might rather expect tassa to parallel yassa. For the suggestion that ass a is for
yassa, see Nonrum, 1994C, pp. 215--16. The cry is clearly reading noydtl, not
no ytiti. I take kod Ma locative adverb, because of its proximity to lake, i.e.
"anywhere in the world". For kocl- kvaci, see the note on l43A.
PD hp 276 ha! nJppa.jjiyati in p3:da a;jilalfl assa na upe,ti anrako in p3.da b.
Udiina-v 29.52 1w nopcyTyate in p3da ~jitam anveti na kaJ?1 cid eva Joke in
p3.da b. Mvu Ill 91.1?- has n&ha;1vaJi in pad.a a;jFtarri a.sya r.ajiniiti antako
in pada b.
The metre is VaiLlliya acd; Aupacchandasaka b. Note the syncopated opening
· · - • - of pada b.

1so"'.Dhp-a m 198,2-7: tal}hd ntlm. esii SGf!!Sibbitapariyonandhanarrhena

}ii/am assd atthf ti pi jiilakiiri.M ti pi jil/Upilmii·n·ja!inf, riipQGlfu iiramma!Jesu
visattatiiya vUattiJ:J. visaitama.n.aJGya visiihii.ratiiya visapupphatiiya visa-
pha/atiiya visaparibhogatiiya pi visattikii.
For visattikii as an epithet of ta~ha. cf. 335 and see EV I; p. 189 (ad Th 400)
and GD II, p. 2lO(adSn 333).
ln place of netave in pi!da b, PDhp 277 has netaye, Uctana-v 29.53 iokaniiyini,
Mvu ill 92,1 • netrika. For the infinitive endinirctave, cf. pahdtave 34.
The metre is Vaitfiliya. Note the syncopated opening· - - • ofpiida a.

I 81. Dhp-a III 227,13: neillammilpasame ratii ti el/ha pabbajjii-

nekkhamman ti na gahetabbcu,, b/esiipasamananibbiinarati'!I pana sandhey'
eta'!' Vu!tCl!J..
-::PDhp 244 reads nekkha1!'mopaSame; Ud3na.;v 21.9 reads naif-
kramyopasame. The cty. however, seems to make it clear that nekkhamma
here is not< Skt. nai~k:ramya. For the fact that BHS always has naifkramya
as the equivalent of nekkhamma, even when it ·is clear that the correct
equivalent is nafykamya, see Sasaki, 1986, p. 3. Cf. nekkhammasukha 272.
For pasuta see the note on 166.
I translate dhfrQ as ·~e... There is no guidance from thecty.
For pihayanti ~the note on 94.
There is resolution of the sixth syllable in pada c. In pada d satimata0 is m.c.
to give the cadence~ - - - . "

182. I assume that manussa- = •manussa- (which is not listed in PED)<

Skt manu;ya "manhood". Rau (p. 169} suggests that the compound is the
equvalent of manufyaUanma"Jpratiliibhal;i. PDhp 334 bas sraddhapafildbho.
ln piida b the omission of ·1!1 in macciina is m.c. Cf. 183 185 207 224.
In piida a there is resolution of the sixth syllable. In piida d note the cadence
· - - - . It looks as though this was in origin a prior piida, now being used as a
 Notes 109

posteriorpada. Cf. 194a. We could read Buddhiinam upp.Jdo m.c. PDhp 334
has kiccho buddhiina upptido kicch!i dhammassa deJanii, i.e. upptido i~ in a
prior pacta_

183. Dhp-a III 237,17-18: saciltapariyodapanan tipaiicahi nivaraT}ehi aflano

citrassa vodapanaTJ1.
PDhp 357 reads -pariyodamana'!'; Udilna-v 28.1 -paryavadanam (probably
for -diinam); Mvu II! 420,tJ• -pary{Jd{Jpanam. It is probable that the dam-
and dap- forms are linked through an intermedia~e dav- form. For the m/v
alternation, cf_ -gavesino/• -gamesino 35 .5; bhamassufbhavassu 3 71.
For Jain parallels to this vcne, see Watanabe, 1995.
In pli.da c kusala is probably the equivalent of puiiiia ·merit, meritorious
deed". For the opposition of kwala to papa, see the note on 53.
ln pada a there is resolution of the sixth syllable, and in pada b resolution of
the first syllable. In pada c we should ignore the svarabhakti vowel in
-par!yodapana'!'. In pada d the omission of-'!' in Buddhiina is rn.c. See the
note on 182.

184--85. These two verses occur together at D II 49,26" foll.

184. Dhp-a Ill 237,19-21: l:hanri Ii yil esii rilil:J:Msa'!'l:hiltil l:hanlf niima
ida'!t bnasmiJ?1 siisanc paramarp uttamal!l tapo.
Dhp-a Ill 237,24 foll: na hi pabbajito ti pa~i-ildihi para'!' upahananro
vihe[hento pariipaghliti pahbajito niima na hoti, samal}O ti vuttanayen · eva
part1J!l vihe{hayanto samal}O pi na hoti yeva.
The metre is Aupaccbandasalca In pada a l:hann-is m.c.

185. Dhp-a Ill 238,1-6: anupaviido ti anupavildanaii c eva

anupaviidiipanan ca, anupaghiito ti anupahananaii c • eva anupaghiitanaii
ca, piitimokkhe ti je{!hakasfle sarrzvaro ti pidahanarri, mattannutd ti
mattaiiiiubhiivo pamiil}ajiinanarr:, pantan ti vivitta'!l. adhicitte ti
a{[hasanuipatti.sGl{lkltiiie adhikacitte, iiyogo ti payogakara1)l1J!1.
In pada a there is resolution of the first and sixth syllables. The v.L
anii.paviido anii.paghiito is unacceptable because it makes resolution
impossible.
ln pada f tbe omission of -1!2 in Buddhana is IILC. See the note on 182.

186. Dhp-a III 240,16--19: kahapa~avassena ti ya'!' so apphofetva sarra-

ratanatasscup vassfipesi tlll!" idha kahiipa'f)avassan ti vulla'!'. tenOpi vatthu-
-kamaki/esakamesu titti niima n 'atthi.- ---
The ideaofapi"even" seems required, although it is not in the text.
Dhp-a m240,21-22: iti viiiiiiiyd Ii evam ere kiimejiiJiitvii.
In pada c dukhil is rn.c. to give the • · - • (parhyli) cadence.
110 The Word ofthe Doctrine

188 .. Dhp-a Ill 246,4-6: bahun ti bahil, pabbatani pi (attha 1a11ha' lsigili-
Vepulla Yebbh6r0dike pabba!e.
Since pabbata is masculine (note pabbate in the explanation), we must
assume µiac -Jni is the eastern masculino accusative. plural ending; see the nO(e
on 82_ PDhp parvvate; Ud..ana-v parvata,,,.1 ca. For eastern fonns see the
note on 32.
PDhp 216 reads va11hiini ruk.lrhacittii~/ (showing the assimilation of ty > 11);
Udiina-v 2 7.31 reads lirdnui0 vrk,acaitym,..I ca.
For the consonant group -ty- in -cetyfinl in pada c. U:e the note on 1-2.

190. It is probable that yo• si quis in pada a. See the note on 76.
For ca = tu "but", see the note on 5.
In pada c we-should ignore the svarahbakt:i vowel in ar'ya-.

191. All the nouns in this verse arc accusatives after pas.sari in 190.
We should ignore the svarabhakti vowel in ar'yaii in pada c..-

193. Dhp-a III 248,21 foll.: yattha so Jiiyatf dhiro utramapaiiilo sammG-
sambuddho tam ku/QJ/1 suJ:ha0 edhan· sukhappa11am eva hoti.
For the phrase sukham edhati see EV I, p. 169 (ad Tb 236).
Tue gloss u//amapaiiiio implies the meaning "wise" for dhiro, but PDhp 79:
GDhp 173 and Udiina-v 30.27 have forms from viro "hero" in place of dhiro.
The dhlv alternation arises from the.similarity of the two characters in early
Bcibml scripl Cf. 418 and vdraye!dhiiraye418, and see GD II, p. 148 (ad
Sn 44).
Dhp'-a JJI 248,16-17: dullahho Ii purisd}aiiiio hi du/labho na hatthi-
Gjti.niyO.dayo viya sulabho. .
-'For pw-isQ}aiiiio, with iijaiiM < *iijiinya, cf. iijiiniya (< Skt iijiineya) 322.
In pada c jiiyati is m.c. to give the - - - • (pathy5) cadence.

194. We could translate saddhamma as "the doctrine of the good(~

Buddha(s)r,
In pada a we should read buddhJ.nam m.c. to give the - - - '"(pa!hyQ) cadence,
which would be more usuaJ with the opening - - - - .

195. Dhp-a lll 252,4--5: papaiicasamatikkante ti samatiklumte ta~hiidifthi­

miinapapaiice.
For papaiica, cf. 254 and for the translation "the diversified world", see EV
I, p. 203 (ad Tb 519).
For pariddava, see GD II, p. 370 (ad Sn 1052).
In pada b va is- m.C. for vii.

196. Dhp-a JIJ 252,IO foll.: n(,- sakhi puiiiia S"'!'khiitun ti puiiiia0 ga~etum
na sakka. kathan ti ce im' ettmn api kenaci ti imlllJ1 ettakmri imaJ!l etraka-n -,,.
 Notes 111

a!lho, kenacf ti apisaddo idha sambandhitabbo, kenaci puggalena ma.nena

vJ.
The reading puiiiia (v.I. puiiiiaf!i) in Dhp-a, in both verse and lemma, is
presumably to get the cadence - - - • (pathyJ) instead of-, - - ·,with the
opaling - - - - .
In pada c puiina'!l is nominative, arid saf!1khiituf!1 is to be taken in a passive
sense with the instrumental kenaci: "this merit could not be measured by
anyooe ...
I assume that api in pad.ad emphasises the totality of ettam. See lhe note on
100-1. The cty takes it with kenaci: Nirada's edition (see the v-11. in
OvH/KRN) reads iii" ... by anyone (thinking) it is so great", but tlill reading
would seem to be ruled out by the cty's reference to api.
For<tta (-ti/aka), see CPD, s.v. '

197-200. It would be possible to t2.kejiv6ma as an imperative: "I...d us live".

197-99. It would be possible to take vihariima as an imperative: "Let us

dwell".

197. Note that verinesu in padas be shows the developmC!ll of a consonant n-

stan to an a-stem, doubtless on the analogy of vuinarra 42. For the
comparable change of ans-stem, see the note on 29.

200. Dhp-a Ill 258,19-23: yesan no ti yesGl[l amhakarii palibujjhana11hena

ragiidisu kincanesu ekam pi kiiicanaT[t n, atthi, pitibhallha ti yathii
.lbltassarii de:vii pitibhakkha hutwi pitisukhen' eva vitin&nenJi ewlJ!1 mayam
pi bl:avissiimii ti attho.
Foe the inclusion of the antecedent no in the relative clause ( .. for whom,
(rmncly) for us''), see the note on 42.
Focthe Abhassara gods who feed on joy, see DPPN, s.v. Abhassara.

201. Dhp-a Ill 259;11: jayan ti paraJ?! jinanto veraT{1 po{ilabhati.

Spl: I 154,21-22 (ad S I 83,31 '):jinanto vera'/1 pasavati, veripuggale labhati.
Both cties are taking jayal!I- as a present participle, and I follow this
inlapretation because I think that each verb in the verse should have a person
as subject: the winner. the loser and the one at peace. There are, however,
difficulties in this interpretation. PDhp 81 reads jayam veralll prasavati;
GDhp 18i jaya vera prasahadi; Udiina-v 30. I jayiid vairQJ/1 prasavate. It is
possible thatjayal!' is the result of a change of the gender of the noo:njaya to
neuter, and it can therefore be the subject of the verb. It is clear that that was
the view of the redactor of the Avadana-fataka (quoted by Brough. p. 238),
who changed the verse to jayo vairaT{1 prasavati, The same change of gender
seems to Occur at S I 163,2s•-29•:jaya1!? ve maiiiiati balo vacaya pharosa'!I
b/:aJ;a'!'f jayan c · ev' assa ta'!l hoti yii titikkha vijiinato. This is explained at
 I 12 The iVord of the Docln.ne

5pk I 230,4-5:jayan c' e'.I. aJsii ri ass' eva tarr: j.:yaT[l hoii. so Jayo hotf ti
attho. Here, however, the possibility of contami'12.tion by ja;·aTI} in piida a
cannot be ruled out For the suggestion of a change vf gender, see the note on
104. For the suggestion that)aycur may be a IJ.arr.:.d type absolutive, see EV
II, p. 66 (ad Thi 26), and cf. 156. Udiina-v jcryad s·Jggests the possibility of
takingjaya171 as an ablative in -a'!l (see the note on 49 J.nd Lilde:s, 1954,
§ 190).
For the suggestion thatpasavari, pasahati and pras ....-.:ate are related and show
a glide consonant alternatio:::i -v-1-h- < -y- in • pas:;yati '"'seti, see Norman,
l971B, pp. 217-18.

202. Dhp-a lil 261,11-16: l:ali rl dosena samo aparadho pi n' atlhi,
khandhikfud ti kluuulh~Jti samiinii yathii parihar(vamtlnii khandhii duk..khJ
evam ~ duJJ:.hmri 1Uima n' atth.i, santiparan : -:.ibbiinato uttan·'!I aiiiiarrr
sukham pi n' attAl. aHiiaJrt hi sukha'!I sukham c•a, nibbiina'!i paramaJ?1
sukhan Ii attho.
Following Collim (1982, p. 3 l 9), I adopt "constitcent of personality" as the
translation for l:JumdJ-.a.
For lali in the sense of"evil", see GD II, p. 263 (ad Sn 659). For kali in the
sense of"Josing throw at dice", see 252.
The reiding samiind in the cty, if correct, as opposed to samo in the
explanation of dosasamo, supports the reading khandhiidisii instead of
l:handhasamii. It is not clear how l:hand!uidisa should be analysed. Andersen
(1904-5, p. 82, s.v. lhandha) divides it as khandhii-disa, and (p. 120, s.v.
'disa) derives the second element of the compound from drfo. He does not
explain the final -<'I of kJuuuihii. All the fonns he quotes, except sadisa, have a
long vowel, e.g. etddisa, riidisa, etc., but they are all based upon
' demon=rive or personal pronouns, and I know of no other example of -(0)-
disa being added to a noun. It is possible that the original form of the p3.da
had khandha-sadisa, which would have given a nine-syllable pada,
normalised by assuming resolution of the fourth syllable. If such a form were
"corrected" by omitting a syllable and lengthening to -adisa, on the analogy of
tiidisa, etc, then the present form of the piida would have resulted.
Andersen analyses para in santipara'!I (p. 162, s.v. para2) as meaning
"higher, superior" and translates the compound as "higher than rest". This is
followed by PED (s.v. para), and it must be the explanation, but (despite
PED) I know of no other example of para occurring at the end of a
compound of this nature.

203. Dhp-a lil 263,18* and 203,22 reads paramii rogii as two words. This
would give the meaning '"'greed is the greatest diseases"~ which seems
unsatislilctory. It is better to take the two words together as a compound (cf.
the compounds in 204), literally "diseases have greed as the greatest".
I adopt "conditioned things" as the translation for saT('khara.
 Notes 113

ln p3d.a c the: absolutivc riatva must be taken with an unexpressed ptonoun

"[for one] knowing this".
In pada b duJ:M is m.c. to give the cadence - - - - . It is possible that paramZJ
a.rid dukh.J ti.ivc been attracted into the same number as smrJ:juJrd.

204. Dhira lII 267,6--tO: vissasaparam4 Ii mZlta \<I lwtu pitii ltl >"""
Jaddhif!J vis::JQJo n' aff hi Jo aii!Wtako va, yena pana saddhirrt vissilso atthi Jo
asambandbo pi paramtJ uttamo iiJti, ten.a vullal!" vis~aparamo iititi ti.
For the L'irtt compounds with parama, sec lhe note on 203.
For Loe cooscnant group -gy- in drogya111 in pa.:La a, sec the note on I-2. For
the violatioaof the two morae rule, see the llOle oo 70.

205. Dhp-a Ill 269,J: rligadarath&!a111 abhdvcna niddaro.

I translate Jara as "fear" or "distress". Its derivatioo is not clear. It may be
connected with daratha but. if so, the reason for -dd- here is not clear. It is
not requiml m.c., although it is in vita-ddaro at 385. UcUna-v 28-5 has
nirjvaro hat; but (vita-)raga al 385. Then: an: no olhcr parallel versions IO
this verse, laI1at385 PDhp 40 lw vita-jjara and GDhp 35 viklula-dvaro.
The latter soggests the possibility of altanati"' root fonns dvarldval parallel
IO jvarljval. See Brough 's llOle (p. 186 ad GDhp 35). and cf. Turner, CDIAL
6653 foll,h forms from •dval "bum".
Dhp-a ill 269,4-5: rasa111 pivan Ii. navavidhalokultarodhammavasena
uppannap6Tzs0'1' pivanto.
For dhammzpiti sec 79, where there is probably a worc!'.play intended on the
two rneaniags ofpiti.
For lhe viobtion of the two morae rule in pitl<i, sec the note on 70.

206. In p>da b we must understand ariyiina111 with sa'!lniviiso. PDhp 69 has

inslad of sadii. "living with good men".
satii.rri
In pada a we should ignore lhe svarabhakti vowel in a,;yiina111.

207. DhjH ill 271,21 foll.: sabbadii Ii yathii asihatthena vii amitlena
Qsivisadiii va saddhirri ekato vaso niima niccaduilho, lath. eva baleJu" .
saddhin tillltlw.
For the S3ldli of -ena + iva in amitteneva, sec lhe note on 66.
Dhp-a III 172.2-<: dhU-o ca suJ:hasa111vaso Ii ettha suJ:ho sa111viiso etenii ti
suJ:has~ par_ufitena sadtJJWri ekaff}iane viiso suJ:ho ti attho.
The gloss palJtfilena confirms "wise" as-the meaning of dhiro, as does the
contrast wit1:i biilehi.
It is possilllc that dhiro stands for dhU-e in lhe sense ofSkt dhirai/i. as Lesny
suggestcd(l924"27, pp: 2J5C.36). CT. biile 61. Udana-v 30.26 readS dhirois
tiJ ruJduuiBiiviiso "there is happy dwelling with wise men"; PDhp 70 dhirii ·-
sUkhasa'!"'iisii 'Wise men have a happy dwelling". Far the uncertaill readirig
114 The IVord of1he Doctrine

of GDhp 176, sec Brough (p. 236). Rau (p. 170) suggests that we read sukho
ca dhirasa7!1v3so "dwelling \Vi th ·.i.,ise 1nen is h:i.pp'f'.
For ca - tu "but", see the note on 5.
In pa:°da b the omission of -'!f lo c.ddhfina is m.c. Sec the note on 182

208. Dbp-a III 272,7-8: dhiti.sampannaf!l dhiraii ca.

Here the gloss suggcsu "firm, ste:;dy'' for dhira. Cf. 261.
Dhp-a III 272,9-11: araha!tapCpanasarrikhiit:Jya dh uravahanasilat3ya
dhorayhasilaT{l. ·
For dhoreyya, sec Normrrn, l 994C, pp. 22:>-27.
The metre is Tri~~bb ac: Jagatl bd. Note in pJ.da b iiriya1J1, giving a Jagatl
pada. The oth~r editions have ariyam, giving a Tri$'1lbh pad.a if we ignore the
svarabhakti vowel. Cf. L1e neec! to read iiriya- in 236. CPD (s.v. ariya)
acce_pts the proscxlic value - - - in these two contexts.

209. Dhp-a lI1 275,7 foll.: ayoge ti ayunjitabbe ayonisomanasiwe

yoga.rm.in ti tabbiparite ca yonisomnnasikiire. -
Dhp-a ill 275,10 foll.:yunjanto ... ayojen/o_
This confinns that yunja'!I and ayojayalfl are nominative singular present
participles, one of the simplex verb and the otber of the causative. with no
obvious difference in meaning between them.
With attiinuyoginam ci. Udana-v 5.9 arthayogine; PDhp 173
atthlinuyoginli'!'; GDhp 266 arthanupaii~o. Note the alternation ttltthinh. It
would seem that a contrast is implied between one who has abandoned the
goal and one who applies himself to it, and it is therefore probable thal the
correct reading is attlulnuyogina'!' "applM himself to the goal, or hi> own
interestsH. and the text has been affected by attiinam in pad.a a Hone having
--application of himself', i.e. "applying himself'. Alternatively we might see
aaa < iipta "'what: has been obtain~d" or< iirta "what has been taken up".
For pihet' (cty: pihen), see the note on 94.

210. Dhp-a Ill 275,18-20: mli piyehi ti piyehi sallehi sa'!'khiirehi vii
kudiiCanaJ!I ekakkhal)OI11 pi na scmiigacche:yya tarha appiyehi.
For the explanation of the negative injunctive mii samiigaiichi by the optative
na samiigaccheyya, see the note on 133.
Pada c has nine syllables. We could read piylin[a'!' ]. but I prefer to read
piylinam (i.e.-'!'> -m) and a'ssume resolution of the third syllable.

21 L Dhp-a Ill 275,24 foll.: piyiipliyo ·ri piyehi aplijo · viyogo, papako ti
llima/r:o.
For the violation of the two morae rule iii piyii.ppi)'W!I, see the note on 70~ The
insertion of -ii- before -p;r is undoubtedly a scnoal device to make it clear that
the compound is< priyQ +- apriya, as opposed to priya +-priya.
In p3.da a we should ignore the svarabhakti vowel in kirJrJtha.
 Notes 115

212-16. In pada a and pada b jayati is m.c. to give the cadence· - • •. PDhp
77 and Udana-v 5.1 read jiiyate

217. Dhp-a Ill 288,6-8: a ff a no knmma kubbJnan ti allano k.ammaf!1 .ntima

tisso siillia tii. pUrayamJnan ti allho.
For the form of kamma, sCe the note on 15.

218. Dhp-a Ill 289,16--18: chandajato ti kauukamyatavasena }iitacchando

ussJhappatto. analihtite ti nibban~. laJ'!l hi asu.tena kalalfl vti ni/Qdisu
evariiparp va ti avattabbatiiya anaillJtarµ ruima.
Udana-v 2.9a reads chandajato hy avasrdvi. Sec C&P (p. 471 note 11).
Dbp-a III 289,21 foll.: appafibaddhacltto ti anl!gtlmimaggavasena kilmesu ca
appa{ibaddhacilto. uddha'!'SOIO le evarupo bhiillu Avihesu nibbattitvd taro
paf{hiiya po{isandhivasena Akani[thG'!' gacchanto uddJiairisoto ti vuccati.
For uddharruoto, cf. Pug 17,19-25: idh' ekacco puggalo paiicannG'!' oram-
bhl!giyil1U1J11 saiiifojanilna'!' pariiJ:haya opapiitih! hoti tattha parinibbayl
anilvattidhammo tasmil lokii: so Avihd cuto AlapPG'!' gacchati, Atappil curo
Sudassa'I' gacchaJI, Sudassil cuto Sudassi'!' gacchati, Sud=iylI cuto
Akanif{ha'1' gacchaJI, Akani{fhe ariyamaggG'I' ·sanjaneti uparif{hamil•G'I'
saiino}anllna'1f pahilnilya: "YGJ!I vuccati puggalo uddhairisoto Akanif{hagilml.
Pada c has eleven syllables. If we omit ca and read kilmes' and assume
resolution of the fourth syllable we have a Slob pada. Jfwe read cil m.c. we
have a Tri~!Ubh plida; see EV U, p: 58.(ad Thi 12). In OvH/KRN (1994) it
was listed as Siok.a, mainly because offausbell's suggested emendations. I
now list it as a TriHubh pada. Udana-v 2.9 reads l:iimqu tv aprati-
/xuldhacitta, which does not scan as a T~!Ubh pada and, even if we assume
resolution of the fifth syllable, there still remains a nine syllable p1ida with the
unacceptable cadence- - - ",for which see the note on 7.

219. For the labialisation of -a-> -o- after -v- in sotthi < Skt .rvasri, see the
note on 105.

221. Dhp-a Ill 298,1-2: navavidha'!' pi mO"G'!' jaheyya.

C&P translate miina as "measuremenf', in both the verse and the explanation.
In the notes, however, they write (p. 449 note 39) of "the nine-fold self-
est:imation" and compare the cty ad 94, where they translate miina as "self-
estimation". Elsewhere they translate it as "pride" (74 150) or "conceit"
(407). rle translation "jneasumnent" implies a derivation < Skt ma.a < ma-
"to measure". I thin!:: that in all these contexts
it is more likely to be < Skt
m.5na <man- ""to think".
For the consonant group -sm- in -riipasmim in piida c, sec the note on 1-2.
For duHha anupatanti, cf. duldchi!nupatito 302.
The metre is T ~\uhh.
116 Thi! Word oflht! Doctrine

222. Dbp-a III 301,10-11: yo puggalo uppannaf(' kodharri dharoye

niggal)hiturrt sakkotl.
The cty has the v.t. vdraye in both the verse and the explanation. For the vldh
alternation see the note on 193.
Dhp-a III 301.12-14: itaro jano ti itaro pana ra_;a-upara;adina'll ratha-
sGrathijano rasmfggaJso nifma hoti, na uttamasGrarhi.
For the implication of inferiority in ilaro jano, see th~ note on 85.
For the consonant group -sm-in rasmi- in pada d, see the.note on 1-2.
There are nine syllables in p3da d; Fausboll rt'1ds 'taro.

223. Dhp-a Ill 313,13 foll: allodhenii ti kodhano hi puggalo akkodhena

hutvdjinitaboo, asddhun ti abliaddako bhaddakena hutvii jinitabbo, kadariyo
1haddhamaccharf atlimo santakassa cagadttena jinitabbo, a/ikavadi sacca-
vacanenajinitabbo.
The cty interprets kodha u referring to a person "an angry one", and similarly
takes asadhu and kadarlya aJ referring to persons, and also their opposites.
The construction then has to be expanded with hurvii. It seems impossible to
take kodha aJ referring to a person; asiidliu is ambiguous, as is kadariya. We
may contrast Sn 362 (where kadariya occurs with kodha) and Dh> 1122
(where it occurs 83 a synonym of macchariya), with I 77 (where it i• an
adjective). •
After three abstract nouns it seems strange to find a person mentioned in pada
cL We should expcct"onc >hould conquer falsehood by truth'". Udana-v 20.19
has satyena tv lUl(taJ!r jayd; GDhp 280 haJ sacer1a ali 'a ji~a.
In pada c we should ignore the svarabhakti vowel in kadariyar,, 0

22_4. Dbp-a Ill 316,17-18: saccairi bha~e ti saCCQJ"/I dipeyya vohareyya sacce
PatitJhaheyyii ti attho.
For the suggestion that sacca is< Skt •satya "pleasant'', rather than satya
"true", since "pleasant" makes a more appropriate contrast to "angry", see
Brough's reference to the Tibetan version of Udana-v (p. 184 ad GDbp 22).
Cf. 408, where Udha-v 33.17 (N) suggests the reading siityiirri. For siita,
see the note on 341.
Dhp-a III 317,1-S: yiicito ti yiicanaka nfima silavanto pabbajitQ te hi kiRcOpi
dethii ti aydcitva va gharadvare tiffhanti atthato pana yiicanti yeva ntima,
evaJ?I silavantehi ydcito appGJmil!t pi deyyadhamme vijjam.Gne ap~mattakam
pi dadetha.
Udana-v 20.16 read> ¢adyiid alpiid..api svayam; PDhp 292 deyii appa pi
yiicito; GDhp 281 daya apadu yayiJa. These three "texts seem to have an
ablative or quasi-ablative in place ofappasmi, which is presumably a loca_tiYe
absolute: "there being (only) a little".Jile presence ofpi seems essential~
see the v .11. in OvH/KRN - and we cou.la presume on ihe basis_ of 1Jdana,v
and PDbp that an earlier version cif the pada bad dajjii appii pi yiicito, in
which appii pi w3.s somehow changed t6 appasmi. Cb reads appam pi,
 Notes 117

possibly because of appamattakam pl dadetha in the cry. Rau (p: 171)

suggest> reading dajja 'ppasmil ['ppamhJJ pi yacito.
For tbe consonant group -sm- in appasml in pada b, see tbe note on 1-2. The
omission of -111 is m.c. See the note on 14 3A.
Dhp-a Ill 317,6: devaiokmri gacclieyya.
1 asswne that "would go io the presence of the gods" means "would be reborn
in the devaloka".
Jn p3.da d the omission of -'T" in deviina is m.c. See the note on 182.

225. Dhp-a III 321.~: accutan ti sassalaf!! fhJnan ti aJ:uppa!!hiiTlm'!t dhuva-

![hJnarri yatthJ ti yasmil'fl gantvti na socanti na viJiafiiianti ta'!l {hJna'!!
gacc!umti ri attho.
For '°'l'vuta with the instrumental case, cf. 231-34 281 391, and see
Liiders, 1954, § 224. Cf. sa'l'vora with the same case in 36()--01. Contr.ist
ascf!Milmp with the locative case in 7.
For the third person plural indicative middle ending -ar< (--ante) in socore,
cf. upapajjau 301; lajjau 316.

226. Dhp-a III 324,13-JS: ahorattanusiilli=• ti diva ca rattin ca tlsso

siillti siJ:JJiamall&uu!'. nibbilnarri adhimuttti.nan ti nibbti.Mjjhti.sayiiJuori.
For adhlmuJta "intent upon", cf. 344.

227-30. The cty tells the story (see Burlingame, Vol.1II, pp. 113--15) of the
lay follower Atula who, wishing to hear the dhamma; went to Revata who
remained silent, to Sariputta who spoke at length. and to Ananda who spoke
briefly. Displeased. he went to the Buddha. who uttered these verses.

227. Dhp-a [II 328,12-14: n' etal!l ajjatan.Jm iva ti idarri nindana'!1
pasW!f.sanal]'J va ajjata:nmrz va adhunO uppanna'!J viya na hoti.
The cry is taking iva in the sense of .. Jike, as it were", but in the context it is
more likely that iva is the emphatic particle, used like eva, in the sense of
"only". See CPD. s.v. iva.
Dhp-a III 328,18 folL: mitabhtil}inam pi ki1J1 esa suva"f}r;.ahiranna'!1 viya
anilno vacanmri maiifuzmano e/rmri va dve va vatvG tul)hf ahosf ti nin.danti.
For mita "measured", i.e. "moderate. little", see MW (s.v. 2JrJta).
The traditions differ in their parallels to anindito. PDhp 283 has anindito (past
participle); GDhp 237 anini'a (future passive participle); Udana-v 29.45
qnin_ditat_(pas!. pani_ciple). The past participle/future passive participle
variation depends upon a tly alternation. Cf. 303 and see Norman, l 989C,
pp. 383-89.
For ajjatanti.m m.c. in pada-b, see the note on 143A. In pada e there is
resolution of the first syllable. Rau (p. 171) suggests reading mitahhiiJ:rim pi
nindanti.
 118 The JVord ofth~ Doctrine

228. For ekantaf!Z < Skt eluinram "solely. only. exclusively", see MW, s.v.
ekiinta_
In pJ.da a there is resolution of the fourth syllable.

229-30. Dhp-a III 329,12-14: tar,i suva1J1Jadosavirafii:am ghaf(an~majjana­

kkhanuv?JjambonadaniW.W,. viya /w ninditwn arahati ti attha.
For the suggestion that a word-play is intended on acchiddavuai "faultless
conduct" and linbrokon circle", see Brough (p. 251 ad_GDhp 241-42). For
word-plays, see the note on 9--10.

229. Dhp-a III 329,5 fo!L:yaii ca viiiiiil ti ... J"Ullpana pamjitii divase divase
anuvicca nindt.iliiral).G'!l yQ JXUGI[ISiikiiral)OlJ1 vJ jiinifVQ p,.. ~£ll!1.Santi.
I assume that yaH ca here is the sam~ usage cu in 106. See the note on 106-
7 .•
For-iiiiii < Skt -jna, see the note on 7.
In pad.a ace may stand for ve. For the dv variation see the _note on 50. Here,
however, it seems to have the meaning of ca=- tu ·~ut" (see the note on 5).
Udana-v 27.47 has tu; PDhp 286 has ca; GDhp 241 has nu which is probably
a mistake for du. C&P read ca in text and cty, without comment. Cf_ the
suggestion that.,,- ce - ca "but" in 163.
For anuvicca, cf. Sn 530 and see Norman, 1987, pp. 35-37. PDhp 286 has
anuvicca suvt! suve; GDhp 241 a!Jcu 'ija Suhaiuhu; Udana-v 29.4 7 has
anuyujya §ubhiifuhha'!'-

230. In the cadence of pada b we should ignore the svarabhlcti vowel in

DI"hati. For the cadence, seethe note on 9-10.
For n07!J. see the note on 1-2.
~ For the labialisation of --a- >-u- after-m-in Brahmunii. see the note on 105.

231-34. For SGJ?n'Ula with the instrumental case,. see the note on 225.

231-33. For the proleptic use of the cognate accusative sucarila'!l with care
in pad.ad. see the note on 157. The usage here is slightly different, in that
there is no noun in agreement with sucarita.
For rakkhati "beware of. guard against", cf. manopadosam raklcheyya Sn
702.

~31. In pad.a a -pp- is m.c. to avoid the opening· - - • . See the note on 1-2.
In pada d there is resolution of the fourth syllable.

232. In pada d there is resolution of the fourth syllable.

233. !n pada d we could assume resolution either of the first or of the fourth
syllable. It is probably oftbe fourth, to match 23ld and 232d.
 Notes 119

234. Dhp-a III JJ l ,J: dhiril ti ye pa~(iitl!.

This confirms the meaning ''wise".

235. Dhp-a Ill 335,J-5; Yamapurlsl! ti Yamadiill! vuccanli, Ida'!' pana

maral)aar cva scndhaya vultalf1, mi1ra!J01f'I te paccupa.ffhitan ti al/ho.
for Yarm's men, see DPPN, s.v. Yama.
Dhp-a Ill 335,25-26: uyyogamuk.he ti parihanimukhe avaefefhimukhe ca !hito
·s1 ti allho.
PED docs not list mukha in the sense of ..begi.nning". See M'W, S.v. mukha.
Dhp-a Ill 335,2 foll.: pathryyan ti gamikassa ta~efu/adipillheyya'!I viya
paralo~ gacchantassa lava kusalapQtheyyampl n 'atthi ti anho.
I take C1J •.. ca in p3.da c and d to mean that the two actions occur simul-
taneously: "You are both setting out and hive no provisions", Cf. 237 302
and see EV II, p. 170 (ad Thi 479).
Tue metic is Vaitallya.

236. Dlip-a Ill 336,2-<: sa karahf Ii so /VGl'fl samudde nilvaya bhinnilya

dipasmpliatapalif!~ vrya attano kusalapatf!fliarii l:arohi.
Udina-1116.3 dvipam supports the translation "island" for dfpa here, rather
than "lamp", which is C&P's translation, although they report (p. 474 note 1).
that theay interprets it as "island". Cf. 25.
Dhp-a Ill 336,9 foll.: dibba'!I arryabhiimin d ••• pai.cavidham SuddhlivCisa-
bhiimirppilpu'!)issasi ti aJtho.
For the iiture in -h- with the ending -isl in e/ris/, cf. 369 and upehisi 23 8
348; vi!>iihisi 379; and with the ending-id-inkiihill 350. For the ending
-osi, cf.liihasi 154.
For so Rh a second person singular imperative. sec the note on 134.
The m<m is Vaitiiliya. In pada a there is the syncopated opening - - - - . In
piida d we should read iiriya- m.c. Cf. iiriymrl in 208.

237. Dip-a III 337,9-11: upanitavayo ti upd ti nipQJamaltOJ?t, nitavayo

galavGJIO atikkantavayo tvafi ca si idiini tayo vaye atikkamitvii maral)amukhe
[hito ti «tho.
I take or ... ca in pad.a c_ and d to mean that the two actions occur simul-
taneously: ·There is no stopping place and you have no provisions". Cf. 235.
In piidaathere is a v.L ca forva. Forthec/v altanation see the note on 50.
Them-is Vaitaliya In pada b we should scan sa['?']payiito, i.e. siipayiito
m.c. Pafitc is an posterior pada instead_ of the cxpcaed prior. Ch omits pi ca,
htU the pii<la scans if we read viiso and te. --

238. For so with a second person singular imperative, see the note on 134.
For dipain the sense of"island", see the note an 25.
For the inure in -h- and the ending -lsi in upelrisi, see the note on 236.
The m<:lte is V;Ut.affya. In pada a there is the syncopated opening - - - - .
120 The J.Vord of the Doctrine

239. Smith's instruction (Pj 11. p. 683. s.v. kiira) to compare kammiira
suggests that, like Pische! (§ 167), he believed that kammiira is to be derived
< Skt karmakiira. If we accepted that derivation, we should hav~ to note that
the change -akii- > -ayii >-a- i..s very unusual in Pili. There seems, ho1,1,.cver,
to be no reason for rejecting the derivation< Skt karr.:iira which PED gives.
Udina-v 2.10 reads l:nnnJro and PDbp 163 bas l:ammiiro.
For the s.indhi of -assa + iva in rajatasse:va, see the note on 66.

240. Dhp-a III 344, 1-2: ayasO ti ayaro samufJh:'taf!1 taduf.thiiyii ti tiito
ughahitvii.
PDhp 160 reads ayasii tu ma/a samut{hito; Udiina-a 9.19 reads ayaso hi
ma1aJ:i samutthitaJ:r. The form ayasa is presumahly an ablative from ayasa, i.e.
ayas changed into an a-stem noun.. For such changes, see the note on 29.
In pad.ab PDhp 160 reads taro u!!hiiya tam eva khiidati, where we should
presumably scan tatO; Udana-v 9.19 reads sa tadutthiiya tam eva khOdari,
where sa refers back to pad.a a. Note the v.L tat' for tad in the yerse in the cry,
and tato in the explanation.
Dhp-a III 344,2-5: atidhonacarinan ti dhonii vuccati cattaro paccaye idam-
atthitaya a/am etena ri paccavellhitvii pan"bhuiijanapaiiiiii, !aJ?t atikkamitvti
caranto atidhonaciiri n&na.
Despite this explanatio~ I think that dhona is conected with dhura (see EV I,
p. 298 ad Th 1271) and I think that atidhonaciirin refers to the dhutangas.
~EV I, p. 245 (ad Th 844--56).
The metre is Vaiilliya. Pada bis one mora short, which probably accounts for
the insertion of ca in Udana-v.

241. Dhp-a Ill 347,10-15: ya bici pariyatri va sippa'!I va yasma

"cisajjhiiya»tassa ananuyun.jGlltassa vinassati va nirantaraf!1 va na upaf[i:iitf.
tasrrul asajjhiiyamala manta n· vutlaf!1,yasmii pana gharfiviisW!l vasan!assa
utfhiiy' u[!hiiya jhy.r;apa!isa'!1kharal)iidini akarontassa ghara'!1 n5ma
vinassati tasmii anuf!hfmamal.ii gharii ti vurtal!I.
The cty is explaining that anu[!hGna on the part of the inhabitant of a bouse
leads to the destruction of the house, and can therefore be regarded as a defect
of the house (ghara). Skt grha, however, is anested in the meaning "the
inhabitants of a house, family" (see MW, s.v. grha), and I do not doubt that
that is the sense intended here.
Dh.p-a Ill 347,16-I9:yasmfi"gihissa va pabbajitass_a vii kosajjavasena sarira-
pafijaggana1J1. vJ. parjkkhiirapa[ijagganWf1 _VO akarontassa kayo dubbar;.!)o
hoti tasma malarrz var;!lassa kDsajjan ti vultllJ!l.

242 For the:consonant-group _"-sm- in,asmiJ?l~ see the- note on 1-2.

243. Dhp-a IIL350, 14: taro ti heµhiivuttamalato.

In pada a tato would seem to be a quasi-ablative going with ma/ii.
 Notes 121

244. Dhp-a lll 353,IJ-14: paresarrr gu~arrr dharrrsanatJya dharrrsina.

For the meaning "importunale" suggested for dharrrsi(n) <Ski dhar[in, see
Andentn, 1904-5, s.v. dharrrsln. For the same meaning for dhamsi(n) < Skt
d/rvdllqa, ace Brough (p. 244 ad GDhp 221-22).
In pJda a ahlrfuna is m.c: to give the - - - • (pathyO) cadence.

245. For ca - tu ..but", see the note on 5.

For Jdrimatd m.c. in p~d.a a to avoid the opening" -- ·,see the note on 1-2.

246. Dbp-a lII 356,6-8: yo pa~am atimaperi 1i sahallhiklidisu chasu

payogesu ekapayugendpi parassa Jivilfndriyarrr upacchindati.
For atlmdp•tl u the causative of a// + ml "to destroy", see CPD, s. v.
atilio<!pdl., and MW, 1.v. 1.mf.
In pJda c there is teSO!ution of the seventh syllable. ThU probably arises from
the rqil!umtnt of l!detl by Miyalf. For the palatalisation of -a-> -i- before -y-
in Miyat/, ace the note on 20.

z.47. For the &andhi vowel -m- in <WI-m~o in pi!da c, see the note on 34.
In pJda d cditiorui differ in reading l:hanatl or IJia~ati. The correct form for
"dig"' ii khan, but there is confusion in Pili between fonm from Ski *;an- "to
destroy" and Sia khan- "to dig", because of the ovctlap of meaning in certain
contexts.
For the consonant group -sm- in -asmi(rrrl in pada c, see the note on 1-2.

248. Dbp-a Ill 356,23 foll.: cira'!' dulliiiya randhayum ti cirakiilarrr

nlrayadui:J;Mdin01!f a//hiiya ICU[! ete.dhmnmii mii randhanlu mii ma1hantii ti
aJtho.
For the negative injunctive ma+ randhayu'!f, explained by a third plural
imperative, see the note on 133.
For Sl:t randhayatt "deliver over to" with a dative, see MW, s. v. radh.
In pJda a there is teSO!ution of the fourth syllable.

249. Dhp-a III 359,13-17: tatlha ti tasmil!J parassa dQne maya appa'!J vii
laddJuu,i 1Ukha1!J va laddhan ti malikubhGva7JJ Gpajjati, samadhin ti so
pugga/o diva va ralli'rJ va upac5rappan1Nasena va maggapha!avasena va
samMfibri niidhigacchati.
Note tbat.Jano is constructed with a plllilll verb. PDhp 327 dadanti ... janii
and U<ilna-vt!0.12 dadanty eke .•. janii~ have plllilll subjects. W (p. 477
note 29) follow tbe reading aadiiti, giViriga·s-iii&ufu verb. - ·

250. Dbp-a III 359,17-21: yassa c' etan ti yassa ca puggalassa eva0 tarrr
etuu. tl:.a.nesu .mankubhavasal{JkhlittJJ[J akusala'!' samucchinnalJ1 mU!a-
ghaCCt1'11'1 katva, arahattaiiii1Jena samiiluzt<ll"[t so vuttappakfiralJI samiidhi'!J
adhigacchati If attho.
 122 The JYord of the Doctrine

Dhp-a Ill 390,2-J (ad 263): yassa c · elan ti yassa puggalass' <la'l'
issQd;dosajdta171 arahattamaggatitiT}ena samU!aka0 chinnt1f!l mUlaghdtarr
kalvd samUh.aJarp vantadoso dhammojopaiiiiiiya samanniigato siidhw-Upo ti
vuccalf ti attho.
Udana-lf 10.13 hJl3 talamastahr;ad dhotah, suggesting that mii/aghacca'l'
and PDbp 328 miilogghaccal{1 are adverbial accusatives. Cf. 263. For
ghacca, see Norman, 1989, pp. 22()-21.
For ca = tu '"but''. see the note on 5.
For uddh- > iih- in samiihata, sec Alsdorf, 1975, pp. l l{}-16, and cf. 263
and anilhata 338.

251. Dhp-a III 362.23 foll.: dososamo ti yakkhagaha-ajagarogaha-

kurrlbhilagafuldayo ckasmi'l" ye:va attabhQve gal)hiturr sakkonli dosagaho
paita ekantam eva gCD)luitl d dosena samo gaho niima n 'atthi.
Note that sama occurs four times in this verse, in contrast to 202.

152. Dhp-a III 375~13-15:paresaJ?1 hf ti ten' eva kiira1:1ena so pugga/o

SQJJlghamajjhadlsu paresCIJ?J vaffiini ucca!_fhane Jhapetva bhusfll!I opw;i.anto
viya opto)litt.
The particle hi is emphatic "the faults of others'';, there is no obvious reason
for so, and we might suspect that hi so is the equivalent of Skt hi f11lG. For so
< Skt sma, see the note on 172. PDbp 166 bas iha; GDbp 272 .,,,; the
reading ofU&na-v 27.1 is quite different.
For kaliJ?i, cf. 202 and see GD II, p. 268 (ad So 659).
For J:iJawl, see Norman, 1974, p. tJ8 .

.., 253. Dhp-a 377, l-4: ujj~aiiiiino ti eva'!I niviisetabba'!t eva pJrupitabban
· ti ·paresiu,t randhagavtsilfiya ujjhiinabahulassa puggalassa jhQniidisu
ekadhammo pi na vcu;f4hati, atlui kho iisavQ va¢¢hanti.
Th-a III 86,3-4 (ad Th 958): ujjhiinasaiiiiina ti pare hetfhato karvii olokana-
ciaO..; anujjhfryitabba.!fhane pi va ujjhannslfii_
For ujjhana, see BHSD, s.v. dvadhyiina.

254--55. I take pada to mean "footprint, trace, track .. not ..path". Sec the note
on 1. For the idea of apada in the Qk.isa' see
93 ..
For ca= tu ..but", see the note on 5. , ·

254. For papanca, see the note on l95. :.

For tathiigata, cf. 276 andseeEV};.P.. 118 (ad Th 3).

25'5. Dhp-a III 379,1-4: sOlpkhiirii tipaiica khandha, tesu hi elro pi sassato
nama n~ atthi, iiijitan ti bj«Jdhd!JaT[l pana la1J.hiimiinadi!{hi-iiijitesu jena
smpkhiirQ sassatO ti g~heyyuJ?J ·tdr[t ldca'!l iiijitam pin· atthl ti attho.
 Notes 123

From the root iiij- "to move", which is used both literally and metaphorically,
a noun qa "movement, disturbance, passion, craving" is fanned, and the
negative adjective aneja (414 422) means "unmoved, undisturbed. calm,
passionlcis, without desire".

256--57. There are no parallels to these two verses in the other traditions but
I assWJX:: that, since they both contain the verb nayati and draw a contrast
bctwcn sahasQ and asOhasena, they were intended to be taken together as a
pair.

256. Dhp-a III 381,12 foll.: yenii ti yena kilrar;ena allhan ti ofilJIJG'rf
vinicchdahha'!f attha'!f sahasii naye ti chandOdisu pati{{hito sii.hasena
musavadena vinicchineyya musa va/VQ iminQ jitmr: aya'!J pardjito ii parmrr.
niharati __ ayGlfl sii.hasena althaf?1 neti niima, so dhammaffho nJma na h.oti
ti a11ho.
J see nornuon to follow the cty in believing that sahara= sJhasena with the
meaning ..usavddena. I take it to mean "hastily". I similarly take qsahasena
in 257 iomcan ''without hastiness".
Dhp-a ill 382,5-6: ubho niccheyya ti yo pana pa~<;lito ubho atthdriatthe
vi~wuiati.
The cty supplies vadati, because piidas cd lack a finite verb. On the
assumption that 256 and 257 are to be taken together, I take nayati in 257b
as the V<Sbofthe relative clause.
For ca =Al "'but'', see the note oa 5.

257. Dhp-a III 382,6-9: asii.hasena ti amusQviidena, dhammena ti

vinicchayadltammena na chand&Jivasena, samen.Q ti aparMMnuriipen • eva
pare llaJtllijayiIIrt va parJjaylIJ!I va phandayali.
Forasaia:rena, see the note on 256.
There is a v.L pJpeli for phandayati, and C&P (p. 480 note 3) prefer this
reading. on the gi-ounds that "causes to move" is a more direct paraphrase of
nayati. MW (s.v. spand') does, however, list "to move (transitive)'" as a
meaningof spandayali.
It is di1licuit to believe that nayati is used of conducting an attha in 256, but
of leading people in 257. It is possible that pare refers to an unexpressed
atthe. «is an abbreviated compound= paratthe "others· cases'". For
abbrevu.ed compounds, see the note on 54. I translate the Pali as it stands,
giving """'fduct" for nayati because that can be used of both a legal case and
.people. - . . . . .
Dhp-a ill 382,J()..ll: dhammassa guuo Ii so dhammagutto_dhnmmaraklcJUto
dJumuntiapaniiaya samanniigato.
Note the past participle gutta constructed with a genitive in the sense of an
instrnmrma1
In piidabnayati is m.c.
 124 The Word of the Doctrine

258. We might have expected tdvatd withydvatd, as in 259; but this would
have been urunetrical

259. For the proclitic use ofna before ppamajjati, see the note on 8.
Skt pramad- is constructed with the ablative or locative. For the suggestion
that dhammam is an ablative, see Luden (1954, § 193). For ablatives in -am,
see the note on 49.
PDhp 32 pl:assaye, GDhp 114 pha,a 'i, and Ud2na-v 4.21 snS:et suggest thot
in pada d we should read p<h>assati for passati. See Brough (;>. 21 r ad
GDhp 114). It would l>e better still to read an optative phassaye. For phasseti.
see Norman, 1962, pp. 32-34. For the plph alternation, cf. paliia0lphalitam
260.
For ca= tu "but", see the notC on 5.

260. Udana-v 11.11 reads yavata pa/it"'!' iira~. For ilie plph alternation in
phalita, see the note on 259.
Dhp-a Ill 388, 1-2: paripakko ti pariji~1){NUddhib/W:vappatto ti attho.
For the use of vayo as a masculine noun in the phrase paripaklw vayo tassa,
literally "his age is ripe", cf. the words used by the Buddha of himself at D ll
120, 19•: paripakko vayo mayhaJ?i. Udana-v reads paripakvQJ?l vayas (il!ya. At
GDhp l 82 parivako is inconclusive, since -<> may be a neuter ending in the .
Ganclltlri dialecL
There are seven syllables in pi!da a. A trisyllahic spelling •thaviro in place of
thero would correct the metre, as in 26!. For other suggestions, >ee the v.11.
in OvH/KRN.

• 26 L Dbp-a Ill 388~10: dhiro ti dhitisampanno.

This suggests the meaning "firm. steady" for dhira. Cf. 208.
There are seven syllables in pad.ad. A trisyllabic spelling •thaviro in place of
rhero would correct the metre. as in 260. For other suggestions, see the v .IL
inOvH/KRN_

262-63. For riipa at the end of compounds, see MW (s.v. riipo):

.. sometimes used after an adjective or past participle to emphasise the
meaning, or almost redundantly". Cf. mii/ha-riipa 268_

262. For the violation of the two morae rule in vGkkaralJ.a, see the note on
70. See also Brough (p. 240 ad GDhp 186), and Norman (GD II, pp. 158--
59) on vyappatha. - -· -

263. For miilaghacca~. see the note on 250.

For -ii- in samiihata, see the-note pn 250.
For ca= tu_ ..but", see the note on 5.
 Notes 125

264. Dhp-.a III 391,12: abbato ti si/avatena ca dhiitavatena ca virahito.

For the cty"s interpretation of \.'ala, see the note on 95.
In pi!da b a/ika111 < Skt a/fka might be thought to be m.c., but note
alikavMi""'I' 223, where there is no metrical reason for -i-. It is probable .that
a/ika developed> alila by analogy with the co!l'.mon -ika suffix.

265. Dhp-a Ill 391.IS--17: sameti ti yo ca pariuDni vQ mahanrDni va pJp<ini

vUpasamt:ll so 1~(11(1 samiJatt6 samaf)o ti pavuccarr h··atrho.
Despite the historical inaccuracy, the cty's etymology of samalJO (< Skt
Srama1:1a) from samitatta (<Sb. iam) works in Pali, as does iamOIJO from
1ama1JII in PDhp 236. GDhp 189 iramalJO from fomadhare and Udana-v
I l.!4 Jramano fromJamltatvJt are less successful. See the note on 142.
For a1JU1T1-tliala, .be then~-< on 31.
For ca -111"but", seethe note on 5.

26H7. These two versC> are together in S l I 82, Mvu, GD hp, and
Udana-v.

266. Spkl266,6: .WO,,. dhamman ti duggandham akusaladhamma1!J.

Woodwanl(f.n.) says "vis"'?' sctmS preferable". The Cly cxp1"ru!tion seems to
be based cm Ski vUni.
Dhp-a ill 393,2: vinan ti visama1!J vissagandha1!J vii Myakammiidika1!J
dhammaJ!I samlJ.dliya. .
The second of these is til1lCh the same as the Spk explanation. The firs~ which
C&P translate as "uneven", might be regarded as an aberration were it not for
the rcad'mgvfyam41!fdJrannlbri samadi'iya at Mvu lil 422,tJ•. Udana-v 32.18
(B) has mmii111 anarmii111 samiidiiya, but in an Agnean bilingual
vdmadhm-mii.111 samiidiiya oc=. The verse is missing in (N). GDhp 67 has
ve.fma t.fltanna samada 'i, where we cannot tell whether the first two words
are a CO!J¥lUDd or not We sec then that the Dharmagup1'1ka, M2llasiirpgbika-
lokottaravaclin and Sarvastivadin vrnions, and part of the Pali cty tradition all
go bade ID a form with -s(i)m-.
For the mggestion that visamalvi~ama and vissalveSma depend upon an
earlier "'Ws:ma "domestic". which was assimilated to vissa (a western form)
and alsoJCSOlved towisama (an eastern form), see Norman, 1980, pp. 71-72,
Norman.19768, pp. 52-55, and Norman, 1997. For the form of vissa, see
OvH, SP. pp. 171-72 (OvH is wrong to say that vi~amiil!I is unmetrical in
Mvu -it shfws resolution of the first syllable).
In the S YCISion the verses are addressed lo· i brahman who w.S a beggar
(bhikkliata) and wanted, on that accoun~ to be called "bhikkhu". The Buddha
said that mere begging did not iiiake a bhikkhu, nor could someone who was
[a bralumn] still living at home [and who h;i.d not become apabbajita] be
called a blrikkhu. The second verse of the pair states that a bhikkhu wanders
(loke caran), with the implication that he cannot be a householder.
126 The Word of the DoctriM

The Dhp-a version of the s:ory (see Burlingame, VoL Ill, p. 145) add.! to the
background story, presumaoly 10 make it more intelligible when the meaning
of vissa had been forgotten... To explain why the brahman wa3 begging, it
stale3 that he bad gooe forth (thereby destroying the point of the verse). To
make it even clearer why he could not be regarded "' a b.\iillu, it add.! the
d=il that be bad gone forth in a heretic order (bON.rwamaye).
1 would suggest that the vene is a reference to brahmanical teaching about the
householder, and to its incompatibility with being a bhiilhu in the Buddhist
sense of the word. Cf. EV II, p. 86 (ad Thi ll2-16). For another anti-
brahmanical verse. see 344.
There arc only seven syllables in pada a. Brough (p. 191) suggesu possible
corrections. of which ltivatti for Jena is the most likely. Cf. GDhp tavada;
Udana-v 32:18 tfivatii; Mvu ITI 422,12• tiivatii.

267. Dhp-a Ill 393,4-5: yo "dha ti ya idlw.siisan<.

l believe that dha here is an old error for ca. For the alternation dhalca/tu ~
''but", see EV I 237, and cf. 409 412 415 416. Udana-v 32.!9 bas tu.
GDbp 68 du. Mvu Ill 422,w bas ca.
Dbp-a lll 393,5-7: uhhay"'JI p · etlUJI puiiiiaii ca piipaii ca maggabrahma-
cariyena biihitvii panuditva brahmacariyavii IWli
For the explanation of braJ..macariya by the root bah-, cf. bahitapiipa 388.
Dbp-a Ill 393,9-10: so tena iiii~ena kilesii=ip blWwzttii bhiklrhu ti vuccati.
For the explanation of bhiklrhu by the root bhid-, see the note on 142.
In pada b we should ignore the svarabbakti vowd in -c.u'yavii.

268-{i9. I follow the cty in assuming that these two verse> are to be taken
,together, and I take paMi/o·in 268d as the subject of parlvaffeti in 269a.

268. Dhp-a Ill 395,9-11: na monenii ti A:iimGJ!I hi moneyyapa{ipadii-

sarrikhiilena maggaiial)amonena muni n&na hoti, idha pana tiu;.hibhiivaf!l
san.dhaya monenJ ti vutlOJ!t, mii/harUpo ti tuccluuiipo.
Sk.t mauna means "the state of being a muni", and then the basic attribute of a
muni, i.e. "silence". lf muni is to be derived from man-/mna- (see the note on
269). then a muni is a think.er or knower. and the state of being a muni is
thinking or knowin~ so that mona could mean "'thought" or "knowledge".
The cry is explaining_ that of the meanings "knowledge.. and "silence" given
for mona, the second is int~ded here.
For ropa at the end of a compound, sec the note on 262-{;3.
For ciJ =_tu ..but•_·. se~_the no!C on 5. ·
Iri. pada a muni is m:t: to give the - - - • (p01hya} cadence.

269. "b~l\c• JU _3_9_6,&cl_l;yo mllnifii uhho loke ti yo puggafo imasmirri ..

khand!ladifoke tUla]rt:_iii-opetvQ _Tninliiit"'a ~YJYa- lme ajjh'1ttilca~-khandh11'-iirie -
bOhirO ti:.adTna na}'enli-;me ~bho pi bf1he miiiiiti.
 Notes 127

The cty is suggesting that the word muni is to be derived < muniiti. If this is
correct, then muni originally meant "thinker"; see Nonnan (1961, p. 350) on
muMti < •mnllt/. The cty (probably because of the idea of measuring in tu/a
in 268) explains munati as meaning miniiti "measures".
In p!da c yo munliti ubho loke can be translated, and is so translated by some,
as "who knows both worlds". I follow the cty in talcing /oke as a locative, and
I assume that ubho refers to varGJ?1 and piipQni.
In p3.da b muni is m.c., to avoid the opening - - - ". There is no metrical
reason for reading mun1- in pada d. h is probably on the analogy of muni in
pada b.

270. Dhp-a III 398, I: ahilp.sQ h. ahirrisanena, idarrt vuttarrt hoti: yena pii'}flni
hiTruati na tena karaJJ.ena anjo hoti yo pana sabbapal)ana'r' pQl)i-iidihi
ahifµsanena mertadjbhfrvanJya patfffhitattJ hfmsalO JrJ Ya {hito ayOJ'!'I ariyo
ti pawccati Ii attho.
For the truncated ending of ahimsii - ahimsiiya (instrumental according to n1e
cty; ablative according to CPD, s.v. ariya), cf. brahmacariyii 388, and see
Geiger,§ 81.1, and GD II, p. 171 (ad Sn llO).
The cty tclls a story about the Buddha telling a fisherman called Ariya that
those who took the lives of living creatures could not be called ariya, but the
fonn of p1das ab "he is not ariya because he kills" seems to imply more than
this. They seem rather to be a rejection of an etymological explanation of
ariya. pahaps based on an·= ..enemy", ·m the same way as the cty gives an
explanatioo based 011 ara "distant" for padas ed. ·

271. Dhp-a III 399,17-18: silabbatamattenii Ii catupiirisuddhisilamattena vii

terasadhiitagtDJOmattena vJ.
We may ttanslate sflabbata as .. virtuous conduct and vowsn or «vows of
virtuous conduct". I follow the cty in preferring the former.
Dhp-a III 399,J8-I9: biihusaccena vQ ti til)l)GI!J. pi{ak.ana'!' uggaT_Jhitamattena
vii.
GDhp, PDhp and Udana-v all have forms based on -sruta, and the -sacca of
the Pili form is not easy to explain. The cty's explanation implies that the
connection with Sruta was understood. It is possible that the expected
bf
•bahUsuccD < bahu.frutya was influenced sacca ..truth", but it has also
been suggc=:sted that bii.husacca is < • bahusmartya < bahusm[fi.
For vii puna in piida b, cf. vii pana 42.
For vivicazs:ayana a·s a syntactic compound. see the note on 24 and Norman,
_l29QB,pp-~24.----·-······- -- ... ·
In pada cthere is resolution of the first syllable.

272. Dhp-a ill 400,1: nekkhammasukhan Ii aniigiimisukhani.

For the ambiguity of Pali nekkhamma, see the note on 18 !. PD hp 272
nelcldun!una-; Vdan!'-v 32.32 sal'(lbodhi-.
 128 The Word ofthe Doctrine

Dhp-a III 400.)-4: bhiillU ti !esaJ?l aiiiialariJI!1 iilapan!o iiha, vissilla m/1pddi
ti vUsii.sa111 n.a. Qpajjeyya.
The cty U taking m<ipadi as a negative injunctive, which it explains by an
optative (see the note on 133), and this explanation is followed by most
translators. This interpretation can only be correµ if-'!' has been lost before
miipddl, and Rau (p. 172) suggests reading vimisa'!' miipddi. The loss of -m
cannot be m.c., since Warder (1967, §§ 242--44) suggests that in Pfili the
cpening .. - - - can occur with the vipula - • - - · in. a prior pada. It is,
therefore., not clear why Pali does not have -1!1-
PDhp 273 has bhikkhu vWiisam ilpadi; GDbp 66 reads bhikhu viSpasa
mavadi; Udana-v 32.32 reads bhik,ur vifviisam iipddyed; Mvu III 422,11 has
bhiqu vi.lvcisam iipadye. It is not clear why not one of these versions has
vissdscvp (oi the equivalent) with a negative, either na or ma. The facts that
the.metre does not preclude -a'!', and yet no version reads it. lead us to the
conclusion that the sentence is to be taken as positive and we should divide
the words as vissilsam iipadl. The negative which the sense Qear!y requires is
supplied by the particle na in the previous verse, which is to be taken with
both the verb in pada a and also with the verb in pada c. The fact that the two
verses occur together in Pali, PDhp, GDhp, Udana-v, and Mvu, confinru that
the transmission process in each of these traditions realised that the syntax of
the two verses made them inseparable. I would therefore translate, "Not by ...
do I reach the happiness of release; (not by ... ) has a bhikkhu attained
con.fid=. not having gained ... " ..
The BHS versions, with an optative instead of iipiidl, can be taken as
supporting this suggestion. The Mvu redactor replaced phusiimi by the
present participle sprhaya111. giving the meaning, "Not by ... would a
, bhikkhu, desiring the happiness of release, attain confidence". The reversal of
' padas ab and cd which we find in the Udana-v suggests that the redactor
understood it in this way, and by changing the order of the padas made the
sense very clear: ''Not by ... would a bhikkhu anain confidence, (or) woold
he reach the happiness ... ".

273. Note the sand.hi of maggiinCJJ?l + a!fhailgik.o > maggana!{hail.gii:o.

Dhp-a ill 403,l I-14: dipadGnali ca cakkhuma ti sabbeslll'{J pi devamanussadi-
bh~ dipadiinlJ111 paiicahi cakkhuhi cakkhumii tathiigato va se!!ho.
In pada b padil means "(religious) sayings", and in pada d -pada in dipada
means ·~foot". For pada, s~e the note on 1.

274-75: I take hi as the emphatic particle "This one'', although in 275 it

could be the explanatory "for".

274. Pada c has ten syllables. Even ·if we assume resolution of the sixth
syllable, we would still have nine syllables. The piida is possibly back-formed
from 275a, where we may safely assume resolution of the sixth syllable.
 Nor es 129

275. It is possible lha1 ve is< Sh vai, but Udiina-v 12.9 has vo ''to you·', and
so has PDhp 359. It is therefore probable that ve is the eastern form ofvo; cf_
315. See EV 1, p. 190 (ad Th 403). For eastern fonns, see the note on 32.
Dhp-a III 404,2-5: salJa5anthanan ti rGga!._a//Qdinarrz Sa{lthana111
nimmatliana'!J abhlihOna'!f etaT(I magga'!J mayii vinJ anussavii.dihi atta-
paccakkhato iiatva va ayaJ?1 maggo akk.hato.
Udana-v 12.10 (B) has ialya·kfnlanah; (N) has -.iranlhanah; PDhp 359 has
ia/lasaJ?Uano. Tnese are nominative forms and mu.st be in apposition to
maggo. I take -santhanairr as accusative, the object of a.Maya.
According to PED santhana is< Sam "appeasing" (Skt iiintvana). although
no reason is given for the aspirate -th-. This derivation is incorrect, and
Udana-v (N) shows that santhana is from the nasalised fonn of the root
.ira1hl.iran1h, i.e. iran1hana. See 1'' man. 1987, pp. 44-45. For Skt .irathana
"destroying, killing", see~, s.vv . .iralhana and sran!hana.
The cty seems to be taking aiinZiya as an absolutive (< Skt iijnfiya), and I
translate accordingly, although PDhp 359 has lll!'iidyye and Udana-v 12.9 has
ajnliyai (a mistlke for Jjiifiyw), i.e. a dative "to/for knowledge".
Jn pada a there is nosolution of the sixth syllable. Cf. 274e.

276. Dbp-a Ill 404,8-10: kevala1r1 hi akkhii!Jra va tathligatfi. tasmfi tehi

akkh5ravasena ye pa{ipannii dvihi jhiinehl jhQyino le tebliiimaka-
w1//as~t/l MJra/xmdhanli moillanti.
For tathiigata. see the note on 254.
The ctymakes no comment onpamokkhanti. PDhp 359 haspramokkhanti
with ·hO'!'dhanli; Udana-v 12.11 has prahiisyan/i with -bandhanam. For
pamalJJianti as the future passive ofpamuneati, see the note on 37.

277-79. Dhp-a !II 405,21-23: esa maggo visuddhiyZi ti visuddhatthZiya

vodanattltaya esa maggo ti attho.
The cty is taking visuddhiya as a dative of purpose "'road for the sake of
purity". I translate "road to purity", tzking visuddhiyii as a dative of the goal of
motion. See the note on 174.
In pada c nibbindati is m.c. to give the · · - • (pa1hy5) cadence.

277. Thcte are nine syllables in pada a because of the inclusion of ti by

analogy with 278-79.

279. Db{>-a,m 407.s-s: sabbe dhammii ti pancaillandhii va adhippetZi,

anattfi_fi--mii·--jfyantu mii mlyantii ---ti- vase- vattelu!!J na sakkfi ti,
avasawztta.nat{hena anana suii.iiii ass5mika anissarii ti attho.
PDhp 374 reads sabbadhlll!'mfi anZittfi; GD hp 108 sarvi dhama a~atva· di;
Udana-v 12.8 (B & NJ sarvadharmZi aniitmana~. The Udana-v tradition
clearly regarded anattZi as an adjective and replaced it by anfitmana~ "the
130 Tli• Word ofthe Doctrine

dha'!imas do nae have a self·. For a discussion of this, see Nonnan, 1981,
p. 25 note 2.

280. Dhp-a Ill 409,16 foll.: anu{{haMno ti anu;/hahanto aviiyamanto yuva

balf ti pa.thamayobban~ fhilo balmampanno pi hutvJ a/asabhiivena upeto
hoti.
For the present participle ending -dna in anuf!hahan.o, see the note on 20. For
u(lhahari, cf. utti!lhe 168.
There seems to be a contrast betwecoyuva ball and GlasiyCU[I upeto, which the
cty shows by the inclusio~ of pi in !he explanation.
Dhp-a III 410,4-5: paiiiiJya daflha~ ariyamaggmri apassanto na vindati
na pa.!ilabhati d attho. ·
The cry is taking pnnniiya as an imtrumental "does not find the road by
wisdom", although a number of translators take it as ·~o wisdom", probably
influenced by maggo visuddhiyi1in277-79. I follow the cty, in the belief
that the usage of pannaya here is parallel to that of panii(iyam 277-79.
Tne metre is Tnwwh abc; Jagati d.

28 L Dhp-a Ill 417,9: visodhay• ti """'11 visodhento.

This agrees with Liiders' view (1954, § 228) that visodhaye is a present
participle, with_,, as an eastern foll!l of-mr. See the note on 371. For eastern
forms see the note on 32. PDhp 278 has vilodhiya (an absolutive); Udana-v
7.12 viiodhayann (a participle). There is, however, no objection to taking it as
an optative: .. One should pwify, •.. one would attain", Le. "if one purifies,
one would attain". It is an optative in 165 and in 289, where PDbp 36&-69
has visodhaye and Udana-v 6.15 vi.fodhayet. The similar, but not identical
pada at Un 24. l Id {not l lc-:u stated in OvHIKRN), has ee tinni visohaye, i.e.
an optative.
For (su-)s07J1vuJa with the insmunerua.I case, see the note on 225.
The metre is Tri~tubh b; Jagati acd. In pada b we should ignore the
svarabhakti vowel in /raYrii. In pada b we should read ca rn.c.

282. Dbp-a III 421,2: bhUri ti pa{havisamtlya virthatiiya pafifiiiy' eta'!1

nt1maf!l.
Dhp-a III 421,J--5: elGJ!I dvedhiipathan ti efa'!I yogaii ca ayogafi ca bhaviiya
vibhavtiya ca ti va<jcj.hiyO. ca awuftPr.iyii ca.
If we take -path a literally· as "path", then bhavaya and vibhavaya are datives
of goal of motion. See the note on I 74. For the idea of "two-fold", see the
noteon384. ·- ____ _
In pada a we should read jiiyati m.c. to give the - - - • (pathyii) cadence.

283-84. The cty tells a story of the Buddha exhorting bhikkhus to cut down
the forest of riiga, dosa and moha: and Of Uttering these two verseS to-explain
that the word vana was not being-used in its literal sense. The difference in
 Notes 131

rnctn: suggests that they were not originally together, but have been asSociated
because of the word vana!ha occurring in both of them. The association must
be old, since they are together in GDhp 93-94, PDhp 361--{;2 and Udlna-v
18.3--4.

283. There is a word-play on vana "forest .. and vana "desire ... Cf. 344. For
v.'ord-plays sec the nore on 9-10.
In pad.ab jtiyati is m.c. to give the - - - - cadence.

284. Dhp has narisu where PDhp has nlitisu and Udina-v bandhu;u. The
idea of "relative" is supported by the simile of the calf and its mother.
The metre is Vaitiiliya. ln pada a we should read y2iva('!I] to go with rava.
This avoids the need to follow FausbO!l's suggestion ofremoving hi.

285. Dhp-a III 428,21-22: siiradiko.n ti saradakiile nibbattG'!'.

The cty offers no comment on pliQinii. PDhp 363 has piiQina; GDhp 299
praniQa; Udina-v 18.5 (u)daJ:ar. For the suggestion tbatpliQlnii is an old
mistake for piiniya'!I "water", see Norman, 19718, pp. 219-20, a";id Smith,
1938,p.2.
For sugala, suNorma11, EV!, p. 161(adTh185), EV II, p. 90 (ad Thi 135)
and 1994A,p.224. Cf. 419.
For hr- not making position in briihaya, see the note on 75.
The metre is Vaitfiljya. Note the syncopakd opening-• --in p.ida c.

286. Dhp-a m 431,10-11: idha vas.san ti ima.smi'?I /haneidaii c' idaii ca

karonto catwniisal!J vassal]l vasi.s.s5mi.
GDhp 333 mids va;a kari;amu; PDhp 364 va.i.M kari;yiimi; Udiina-v l.38
var~OJ!l kar#yami. These all mean "I will make the rainy season dwelling".
Because of the confusion between vassa ..rain" and vJsa "dwelling" in Pali,
the two words are used almost interchangeably in some contexts, and here
vassa1!J is used almost as a cognate accusative "I shall dwell a rain
(dwelling)".
Dbp-a III 431,14-16: antarGyan ti asukasmirr: n&na kale va dese va va;.·e va
marissiim.I ti attano 1Tvitantartiya1!J na bujjhati.

287. Dbp-a III 433,4 foll.: byasattamanasan ti hiraiiiiasUVG1J1Jadisu va

pattacivariidisu va b"iicid eva labhitva taro uttaritartllft pal!hanatJya .satta-
miinasaflf ___ yaJ?J yaf!t _laddha'71 h_q~i _ la~tha lattha lagganatGya satta-
mQnaSaf!J.
For the consonant group by- in byruatta-, see the !)Ote on 1-2.

288. Dhp-a ID 434,21: antakenodhipannassii ti mara~ena abhibhiitassa.

1:-i pada a taf}fiya is a dative of purpose. See the note on 164.
 132

289. Dhp-a III 435.3-5: e!am auhwvcsan ti elaf!J -fesOlfl annamannassa

Jii')Drrr hhavi.twri asama!fhabhiivasa'!1ki:iilaf!1 .kflral)GfT1 jiinitvii.
Dhp-a Ill 435,7: virodheyya.
For visodhay~ as an optativc:, see 165_
In pad.a c PDhp 368-69 has saggugan:anam miJggaT(I, but UcLl.na-v 6.15 has
nirviil)agamanmµ.

290. Dhp-a III 449,4-5: taffha matliis'..ikhan ti pamCu:zasuJJw'!l parillasukha'!l

\!UCWfi.
Dhp-a III 449,12-13: ath · etam vipulaf!l sukhaT(I samma passanro pa':-ifito ta.7!1
maniisukharrr cajeyya.
It is difficult to accept that mii.trii-sukha can mean "a small amount of
happiness·~. which should rather be sukha-mJ.trO, and I would rather believe:
that mattii means "material things" (see Brough. p. 229 and MW, s.v. miitrif).
To abandon a small pleasure for a large one does not seem to me to demand
great wisdom, whereas to give up the happiness gained by material pleasures
in this world for a larger (spiritual) happiness is indeed the mark of a wise
man. We may assume that this meaning of matt ii was unknown to the cty
tradition. Udana-v 3030 reads miitriisukhaparitygiid; PDhp 77 miittiisukha-
paricciigii; GDhp 164 (ma)trasuhaparica ·i. If \Ve accept the presence of
miitrii here, then there is no need to follow the suggestion in PED (s.v. rr.attif)
that mattii- is m.c. for matta- (to avoid the • - - " opening; see the note on 1-
2). There is, in any case. no reason for such a lengthening in piida c.

291. Dbp-a III 451~: vera so na parimuccaii niccalil/tlJ!I vercrvasena

dukkham eva piipw;iiff ti attho.
--: We might think that ve:ra was a mistake in pada d, since we should expect a
referenceto dukklra. Cf. GDhp 179 duha, PDhp 117 duUhii, Udiina-v 30.2
duhkhiin.
In view of the support for pari- given by PDhp parimuccati, GDhp
parimucadi. and Udina-v parimucyate, we should probably have read
parimucctJJi, with the v.l., as suggested in the critical apparatus to OvH/KRN.
The readingpari-would show resolution of the founh syllable.
There are nine syllables in piida b. \Ve could change to yo icchati suk.ham
attano (with resolution of the third syllable), or read icch.e. Cf. PDbp icche,
Udana-v icchet, GDhp ich~ ....

292. PDhp 266 includes hi in pada a, but GDhp 339 and Udana-v 4.19 both
omit it.'GDhp, PDhp and Udiina-v allinclude ta(d), and w.e should probably
have followed the v.L and included rad in OvH/KRN. If we follow PDhp in
reading both hi and tad in pad.a a, theie would be resolution of the sixth
syllable. It is possible that the omission of tad in Dhp and hi in GDhp and
Udana-v is- due to normalisation. ·
For a discussion ofunna/a, see Brough (pp. 27'1--80).
 Nota 133

Jn paC. b we should ignore the svarabbakti vowel in kaylrati. For the cadence,
see the note on 9-10.

293. Dhp-a lil 452,24-25: slit=Mrino ti satataklirino a({hitakiirino.

For ca= tu "but", set the note on-5.
The co= after susamaraddM in OvH/KRN 1994 should be removed.

294-95. It is probable that these venes were t.ake .over from a brabmanica!
source, although it is not clear whether the literal meaning, that whatever a
brahman docs be goes scot-free, was ever intended. It was inevitable that in a
Buddhist context they would be explained metaphorically, whatever the
origiml meaning.
T.. position of ca in pMa b is unusual, and we should expect to translate
''mother, father, kings and twokhattiyaslsotthiyas, the kingdom/tiger" but this
seems to be ruled out by-paiicamtvir in 295. We must therefOrc assume that
kha11iya and sotliya apply to the l::ings.
Dhp-a lJI 454,11-13: anigho ti nidduillo, briihmano Ii khiniisavo, e/esa'!'
1anhiidf1UI1[1 arahattamagg~ hatatta khinlisavo niddukkho ~utvii yiiti.
It is possible that there is a word-play underlying hantvli and anigho, based
on-a word-play on ni-han (Sb ~erj and a-nigha (cf. nigha "destruction"
Thi 491), similar to that on anigha and ni-han at Th 744-45 (see EV I,
p. 23 J). For word-plays, see the note on ~JO.
For hr- not making position in broizmano, see the note ?n 75.

294. Dhp-a III 454,1-9: slinucaran ti iiyasiidhakena iiyullakena sahital[l.

euha hi la!J-hd Janell purisQJI ti vacanaJo tfsu bhavau sattanaf!'J jananato
ta1:1ha miita nama; ahaT[t asuk.assa nfima raniio vii riijamahamattassa vii
pul!o ti pitara'!I nissOya asmimii.nassa uppajjanalo asmimdno pita nGma;
/oko vl)ia riijiina1!J yasma sabbadi[!h.lgalQni dve sassatucchedadi!fhiyo
bhajanti. tasma sassatucchedadi[(h.iyo dve khattiyariijlzno nii.ma;
dviidasiiyaJaiuini vilthafarthoui raf!luzsadisattd ra{fha'fZ niima, iiyasadhako
iiyullakapuriso viya la'!J nissito nandirago anucaro nama.
The cty interprets sanucaral{Z as referring to a single anucara, who is
identified as the tax-eo!lector. I see no reason for following the cty in this.
I asswnc that raff ha means "the people of a l::ingdom, a nation, subjects", as in
Skt. See MW, s. v. rifyfra.

295. Dhp-a'11! 454,21 foll.: dve ca sollhiye ti dve ca briihma~e, imissii hi

g&thiiya ·sattl:a attano dhammissarataya · desaniividhikusalatiiya ca
sassatucchedadif!hiyo dve briihmm;ariijano katvii kathesi.
The concept of briihmana killgs seems unlikely, and I assume that sollhiya, if
it applies to riijiino as khattiye does in 294, has a more general meaning
"learned". Ud.ana-v 33.6!-<i2 avoids the problem by reading riijiina'!' dvauca
.Srotrijau in both verses, although this makes nonsense of vyaghral!I ca
134 The ~Yord of the Doctrine

paMcama~ in JJ.62c. The reading of GDhp l2 is ambiguous. There is no

equivalent of pJ.da bin PDhp.
Dhp-a III 455.l-5: veyyagghapancaman ti ettha vyagghGnucarito
sappa(ibhayo duppapj>ajjo maggo veyyaggho n&na~ vicikicchanivara!)am pi
/en a sadisataya veyyagghaJ?! ndma, /QT'(l paiiCamLJ!fl assQ fj nf v0TOl)O-
paJ1cakal'(J veyyagghapaiic.amarri n.Gma.
The obvious way of taking veyyagghapaiicammp is as the fifth after the four
already mentioned, i.e. mother, father and two kings.. The.cry, however, takes
it as implying that there is a group of five of which four have not been
mentioned, and in.tc.rprcts this as being the five n~. Since all the other
victims are per.;ons, it seems strange to speak of killing a (tiger-infested) road,
and I take veyyaggha (< Skt vaiyilghra) to have the same meaning as
vyaggha, and l follow Max Miiller ( 1881; p. 71) in believing that this refer.; to
an eminent man. Cf. Skt vyiighra "any pre-eminently strong or noble person,
'a tiger among men'" (MW, s.v. vyiighra). The cty seems not to know the
usage with reference to a man. PDhp and GDhp have no parallel to pada c.

296-301. Note the usage of the cognate accusative suppabuddhal1l with

pabujjhanti. For the analysis as a proleptic acoisative nsod adverbially, see the
note on 157.

302. Dhp-a Ill 462,9--14: duppabbajjan ti app<»?I vii mahan/0l7l vii bhoga-
kkhandhaii c' eva iiii.tipariva!{aii ca pahiiya bnasmbri so.sane urwri datva
pabbaj}a'!l niima duillOI!f, durabhiraman ti eYW?I pabbajitentlpi bhikkhr5.-
cariyaya jivitavullirp gha/antena aparimDl}QSilaillandhagopana-
dhammiinudhammapafipattipiiranavasena abhirami/JD/! du>J:hGr[I.
For duppabbajja as a neuter noun "(there is) a difficoh going forth", see EV I,
p.147 (ad-Tu I I I).
Rau (p. 172) suggests readingpabbajjal1l durahhirt1J1UU/I, but since pabbajja
is normally feminine. this would cause difficulties of grammar and
interpretation. See also Brough (p. 256 ad GDhp 262).
Dhp-a lil 462,21 foll.: dukkho 'samiinasal1lviiso ti gihino vii hi ye jdtigotta-
kulabhogehi pabbajita yQ si/Qciirabahusacciidihi sam.Gna pi hutva ko si fVCII!J
ko ahan ti Gdini vatva adhikaral)apasutii honti JI! asamiinti niima Jehi
saddhi'!l SOJ?1V'2so nilma dukkho ti attho.
Dhp-a III 463,.....,;: dukJ:hiinjtpatit' addhagii ti ye va{f11SGl!lkhiital1l addhdnal1l
pa{ipannO.na addhagU te dukJ:he anupatitQ va.
The cty is taking addhagiJ as plural although, as it is clearly singular in pada e,
this seems to be ~ecessary._Jn OvH/KR.i'l _we Punctuated -iinuparit'
Dddhaiu, ··as-sllrn.ing the sandhi of -o + addh- > -addlz-. For this sandhi, see
the note.on 10. For anupatati, see the note on 221.
For the sense of ca ... ca in pad.as ef, see the note on 235.
 Notes 135

There are nine syllables in p3.da a. We could assume resolution of the. fifth.
six:h or seventh syllables. I have listed it as sixth (-abhi-). In pada b duJ:hii is
m.c. ln pada fthere is resolution of the first and fowth syllables.

303. The traditions '.iiffer in their parallells to p"fijito. PDhp 331 h.S pfijiyo
(futi.:re passive participle); GDhp 323 puyidu (past participle); Udana-v 10.8
pUjyar~ (finite verb}. The past participle/future passive participle variation
depeods upon a tly alternation. Cf. 227 and see Norman, I 989C, pp. 38&-89.

304. Dbp-a III 469,t6-18: tartha rnnto ti riigadinarri santataya buddMdayo

sanra ndma, idha pana pubbabuddhesu l:atJdhikarJ ussannakusalamM!ii
bhJvitabMvan.G sattJ santo ti adJdppeJii.
n cry seen1s to be confusing san/o with santa, i.e. the nominative plural of
the past participle of the verb Mm- "to be at rest", although the alternative
explanation (idha pana) suggests that santo has been recognised as the
nominative plural of the pr=t participle of the verb ru- in the sense "good".
See the note on 54. Note asani' for asanto •"bad". For the sand.hi of -o + e- >
-~ in 05antettha < Q5anto ettluz. cf. 174.

305. Dhp-a !TI 472,13-14: eJillmulI[I tlaseyyDJ?1 ca bhajet/uJ ti attha.

It would be possible to take d:iJsana'!' and tJ:auyya'!' as adverbial
accusatives, as they must be in PDhp 313 and Udana-v 23.2 (GDhp is
inconclusive), but in Dhp they can be taken as objects of caram.
for the pleonastic -anta in vanante, see GD II, p. 174 (.ati Sn 127).
Udana-v 232 reads ramayec caikam iitmiina'!' van~ eka~ sadii vaset; PDhp
313 eh> ramayam iittiina'!' vaniinte ramitii siyli; GDhp 259 eku ramahi
atvana araEi eka 'o vasa. Although the other traditions have fonns from the
root ram- in p3da c, no P3li source known to me reads ramayam for
damayam. for a discussion of this, see Brough (p. 255 ad GDhp 259) and
C&P (p. 490 note 29). For the d!r alternation, see GD II, p. 160 (ad Sn 81).

306. Dhp-a Ill 477,9-1 I: katvii ti yo vii pana piipakamma'!' h>tvli niilw'!'
e/GJ?I karomi ti ii.ha.
Vida b scans correctly but, in place of c' iiha, we should have expected cc(<
ty < [1]n) aha. The double consonant has probably been simplified to -c- m.c.
Dhp-a III 477,s• reads: yo viipi karvii na h>romf ti ciiha, which is unmc:trical
and presumably represents an attempt to solve this problem. The other
versions 4ea1 with the problem in different ways. PDhp 114 has na karomiti
iiha, which does not scan. GDhp 269 na h>rodi liha agrees with Udana-v 8.1
(N) na karoti iiha in omitting ti and changing the verb from first to third
person.-Brough 's suggestion (p. 258) that a past tense is required here is
rejectedbyC&P(p.491note1).
Here pi emphasises tlie totality of the number two. See the note on 100-1.
For the dv alternation in viipi I cii.pi, see the note on 50.
 136 The IVord oflheDoctrine

The metro is T~\Ubh.

307. Tull vc:rn: recurs at Vin Ill 90.25-26 and It 43,J--O.

Dbp-a III 480. l: /WsiivaJ:mrhii ti hisiivena pafivethitakaQ/ha. ·
Sp 486,.22-25: .tdsdvaka~;i:a ti kiisiivena ve/hitaka!)/}•ii. eltakam eva ari;·~­
ddhajadharal}Gmattmri sesarµ. stlmal1iia'!1 n 'allhi Ii vuJtal{l hori.
It-a I 177,3-4: Usiivakcl)Jhd ti ka.sii.varasapftallii kO.savena vatthena
pa/ivefhitaJ:m:l/M.
PDbp l 13 J:i4iiyalwmfha bahavo; Udina-v 11.9 J:D.,oyaka Qfhii bahava~.
For a discussion of tho fact that a scrap of yellow ar the throat does not make a
man a bhikJ:hu, sec OvH. SP. pp. 92-93.
For the third person plural indicative middle ending -are in upapajjare, see
the note on 225.
Th~ is resolution of the first syllable in pada d.

308. Dbp-a Ill 481,7-10: /alto iiditto aggivaQQO ayogu[o bhutto seyyo
sundara!aro ti, J:i~ ? tappaccayii hi eko va aJtabhiivo jhiiyeyya, kussila
pana .saddhadeyy~ bhunjitvii anekiini pi jiitisatiini niraye pacceyyQ d attlio.
For the use of the verb bhuiijati with raf!hapiQifa, see EV!, pp. 239-40 (ad
Th 789) and EV IJ, p. 53 (ad Thi 2).
Foryan ce"tban if", cf.106.

309. Dbp-a Ill 482,19-21: na nikiimaseyyan ti yathii icchatl evQJ/'I seyycuri

alabhitvii anicchitQJ/'I pariaakam eva kiillll!' seyyG1]1 labhati.
The construction of nikiimaseyyaT{I with na is unusual. We might have
expected a'. Cf. PDhp 210 anikiima.leyaT{I; Udana-v 4.14 anikiimasayyiiT{I;
GDbp 270 ani"ama-saya.
·· The metre is T~pibh. In pada b iipajjati is m.c. and in pada d tatiymri is m.c.,
although this may be a historical form< Skt l(tiya.

310. Dhp--a ill 482,22-25: tena ca apuiiiiena nirayasa:rrtkh.ata pOpika gati hoti
ti, rati ca thoJ:ika ti tassa bh.itassa bhitliya itthiya saddhil!I rati sapJ thoJik.ii
parittii hoti.
The metre is Tri~tubh cd; J2gati ab. In pada a gati is m.c. and in pad.ab rati is
m.c.

311. Dhp-a llJ 484,19-20: nirayiiya upakaefifhati niraye nibbattiipeti ti aJtho.

PDbp 296 reads upaka!fati; Udana-v 11.6 upakar;ati. For kaififhati see
Norman, 19938, pp. 149--54, and cf. BHS kaffati_fsc~ BHSD, s.v. ka/taJ1).
For nirayiiy' as the dative of goal of motion, see the note on 174. ·

312. Dhp-a lIJ 484, 2!}-21: sithilan ti oliyitvii karq~ena sifhilagiihaT{I katvii
kataJ'{tya1!f kiiici kaml1U11!J. , -- -- -
For safhila in a moral sense, cf. 313, and cf. sithi!a 346.
 Notes 137

For mahapphalaf!l, showing a development from mahat- rather than mahJ-

(i.e. < maha1-phala), cf. 356--59, and see the note on 123. Contrast -pph-
m.c. in ka(ukappha/af71 66.
In p3da c we should ignore the svarabhakti vowel in -cariymp. See EV I,
p. 175 (ad Th 277). ·

313. Dhp-a III 485,9-12: kayiraii ce ti lasmQ. ya'?1 knmma"!' l.areyya la'!l
kareyyQth' eva, dafham ena'!l parailame Ii tliirataram eva ln.tvii avatta-
samQdiino hutvJ etal'{1 kayirJ.
Spk I I 07 ,23-25: kayiril (S I 49, 10• kayirali) cdayirolh' enan ti yadi viriya'!'
ka.reyya kareyyJ1ha 71al'f1, viriy11J?1 NJ osaile;ya.
Both cties seem to be taking kayiran in pad.a a as an optative. I follow
Andersen (1904-5, s.v. karoll) in believing that it is a future passive
participle< Skt kiirya. Udiina-v 11.2 reads kurvii~o hi sadii projiio, which
gives no help with the probkm.
In view of Udiina-v 11.2 drdham eva, we might think that eva is a better
reading than ena'!' in pada b. It has probably been repeated in error from pada
a. Hereda!~ is presumably an advabial accusative. Cf. 61.
Dhp-a Ill 485,12: paribbiljo Ii safhilabhiivena lato l:ha'J4iidibhiivappallo
samal)adhammo.
Spk I I 07,26 (ad S I 49,11 •): silhilo hi paribbiijo ti siJhila-gahitapabbajjii.
The cty tradition seems to be unaware of the fact tfurt paribbiija exists in the
sense ofparibbiijaka, and assumes that it is the equivalent ofparibbajjii. For
paribbiija ~ paribbiijaka, see Sn 134. ·
For sa/hila in a moral sense, su the note on 312..
Dhp-a III 485,13-16: bhiyyo Ii abbhantare vijjamiiJuur riigarajiidifri evaropa
sama!Jadhammo apanetu1!1 na sakkoti, alha klio lasso upari aparam pi
rGgarajG.di'!' Okirati ti attho.
There seems to be a word-play on rajas "dust" which a bad ascetic might
sprinkle on himself (sec 141) and rajas "defilement .. which such conduct
increases in.stead of decrea.o:;L.1.g. For word-plays, see the note on 9-10.
In pada a we should ignore the sv.irabhakti vowel in kay'raii (< Skl kiirya).

314. Dhp-a III 486,13-14: duilalan Ii siivajja'?1 apiiyasaf!1vattanikaf!1.

kamma'!I akatarn eva seyyo var£V?I uttaml1IJ1.
Lilders (1954, § 189) suggested that aka1aJ?t was an example of the ablative in
-a1]1. For such ablatives, see the oote on 49. For an alternative explanation
taking~! as a nomillative, see Noonan. 197IB, p. 20 (ad GD hp 337).
For ca= tu "but"; see the note on s.- -- --- --
315. PED says that -gopetha is the potential, i.e. optative. PDhp 234 bas
~ ;jl_
evam
rakkhat ha iittiinal]l, i.e. imperative; GDhp 131 has bhodha in a different
·construction; Udana-v 5. I 7 reads gopayala. It would appear more likely, then,
that gopetha is the second pen:on plural irnperatiy~ of gope1i.
 138 The Word of the Doctrine

Dhp-a Ill 489 ,6-7: so kha!)O tum he ma atikkamatu.

This confirms that in pada d ve ~VO, which is supported by GDhp 13 l yu:
PDhp 234 vo: Udana-v 5.17 vo: cf. kho~o taf/1 ma upaccagd, Thl 5. This is
therefore the eastern form of the second person plural pronoun, rather than the
particle ve < vai. Cf 275. For eastern fOrms see the note on 32.
For ma upaccagii as a negative injunctive, see the note on 133. For the form,
cf.412417.
Dhp-a III 489, 7-9: kha!)Otftii ti ye hi faJ?l kha!JGJ!J afitd ye pa puggale so khn00
atfto te nirayamhi samappitii hutvii tatrha nibbattirva socanti ti attho.
This seems to be explaining khar;.iitltii in two ways: "those who have pass.sed
the moment" and "those whom the moment has passed". I take it as "those
passed by the moment", since that seems to follow from the sense ofpada d.

31§--19. The compounds ending in -di!{hisamddiiniI could be possessive

compounds, but I take them as ablatives of cause. Uda.na-v 16.4 reads mithyii-
dr~tisamodiiniit; PDhp 169--70 has -samiidona.

316. Dhp-a III 490,14: alajjitiiye (E' alajjita ye is a mistake) ti alajjitabbe

lajjitiiye(E•/ajjitay<)ti ___ _
For the future passive participle ending -ttlya < -tayya < -tmya. see Geiger,
§ 203 (and note). GDhp 273 has alajidavi and lajidavi; Udina-v 16-4
alajjitavye and /ajjitavye-, PDhp 169 alajjitawe and lajjitavve.
For lajjanti ... na lajjare, cf. Udana-v 16.4 lajjante ___ alajjma~; GDhp 273
lajadi ... na /ajadi (probably singular); PDhp 169 Iajjanti ... na /ajjatha.
For the third person plural indicative middle ending -are in lajjare, see the
note on 225.

' 317. For-bhayadassino, cf. bhayadassiva 31-32.

319. For ca= tu '1Jut.. , see the note on 5.

In pada d suggaliJ?I is m.c. Cf. 18.

32~21. For the consonant group -ky- in -viikyaJ?I, see the note on 1-2. For
the violation of the two morae rule. see the note on 70.

320. Dhp-a IV 3,9-11: cO.piito patitan ti dhanuto mutfm!t, ativiikyan ti a(!ha-

anariyavohG.ravasena pavlittal!l vitikkamavacanal!J.
PDhp 215 capatipaiite .Sare; GDhp 329 cavadhivadida sara; Udina-v 29.21
as
capa.d utpatit0'!1 Sariin. I therefore translate SQT(ll?f a_pluraL For -_al[I as an
- accusative plural ending, see the-note on -64-65. i.s Brough (p. 274) says, it
seems well nigh essential to asstime that the original verse had a causative
form pdtilaJ?I, and it is noteworthy thet PDhp and Ud.ana-v agree with Dhp in
having a non-causailve fonn.- GDhp is inconclusive.
 Notes 139

PED states (s.v. cJpa) that cJpQlo is m.c. There seems to be no reaso'n why
this should be so, since Cllpato would not be unmetrica!. The ablative ending
-a(t)o is conunon in Pkt (see Pischel, § 363), but less so in Pali; see GD II.
p. 187 (ad Sn 198). The reading in PDhp suggests that capato patita'T' may be
an old error for cJpGtlpatitarri. Cf. the explanation ciipGto atikhinQ for
cQpQrikhi1J.G in the note On 156.
I ass\nne th.at ncJga is a special sort of elephant, like kuiijara in 322. Cf. 322
324 329-30.

321. Dhp-a JV 3,18 foll.: samitin ti uyyanakf/ama!)cja/Qdisu hi mah0.jana-

majjha'!1 gacchantJ dantam eva gol)ajGtirrr vQ assajiiti'!1 va yOne yojervJ
nayanti.
It is strange that the cty does not give the meaning "war'' for samitf_ This
might be thought to be the most likely meaning in the cootext, and I adopt it
here..

322. Dhp-a JV 4,11: malianagd ti bDijarasankhiita mahaha11hino.

I tal:e lruiijara as a parallel to ntiga, and assume that it is a particular sort of
elephant, perhaps, as PED implies. a "trumpeter". Cf. matanga 329-30. For
ntiga, see 320.
For iijiiniya (<iijiineya). c(purisiijaniia 193.
Dhp-a IV 4, l3-I.S: yo pmta catwnaggasai!Hii5tena attano dantatOya attadanlo
nibbisevano, aywri tato pl \.'a1W?l- .
PDhp 91 has atta diinto talo VarGJ?1; Udin.a-v 19.7 arma diintas lato varam,
although the alternative version reads iitmadiintas; GDhp 341 atvadada is
inconclusive. 'lamed self' seems to ~ demanded by the sense (and is
implied by attanii sudantena in 323), and we should perhaps read atta danto
with PDhp.

323. Dhp-a IV 6,9 foll.: koci puggalo supinantena pi agatapuhballii agatan ti

sanilltiiJO'!' nibbanadisO'!' ta!hii gaccheyya.
The cty is taking agata to mean "not gone to, not visited"_ I take it to mean Llie
place where there is no gati, i.e. it is not a place of rebirth (cf. 420 and
Norman 1994A, pp. 22-23), it is a place where no one is reborn.
Dhp-a IV 6,12-13: Im!' agatapubOO,,, dis"'!I gacchati dantabhimii'T' piipuntiti.
For disa in the sense of"region, country", see MW (s.v. dif) and cf. bhW,,im
in PDhp 92 and U~-v 19.8.
Note the hndhi ofyatha.+ attanJ > yathauanii in pad.a c in some editions.
·This is-thereS\Jlt of a Sanskritising type of sandlll. For the Violation of the two
morae rule, see the note·on 70.

324. Dhp-a IV 13,13 foll.: kapJ):appabhedana ti tikhi~amado, hatthinaf!! hi

madakiile ka~naciilika pabhWanti pakatiyii pi hatthino rasmif!! kiile ankuse va
/..."1.1.n/atomare vJ khary:fanti CalJcjQ bhavanti, so pana atical)cjO yeva.
140 Th• Word ofth• Doctrine

For ndga sc:c 320.

The metre is Vait.lllya. In pada a we should scan -p<llako as -pdlakJ m.c. In
pada b we should scan -pabhdano 33 -pabhedanJ m.c.

325. Dhp-a IV 16,lL mlddhI ti thlnamiddhiibhihhuto.

For middhl see EV I, p. 140 (ad Th 74).
For mahagghasa. showing a development from mahat~ rather than mahd·
(i.e.< •mahad-ghasa), sec the note on 123.
Th-a I 73,26-l9 (ad Th 17): nlddiJyit/J ti supana-sl/o. samparfva//asayf ti
samparivallak"'!' samparivattaka'F' nipajjirvd ubhayena pl S<yJasukhD'F'
phassasukha'!' middhasukha'F' anuyutto ti dsmetL
Dhp-a IV 16,14--18 "Th-a I 73,29 foll.: nivdpapu((ho ti ku~qak/Jdlnd
siikarabhnllena puffho, gharasilkdro hi bdlal/Jla!o pa((hdya poslyamdno
thii/asarirakiile geha bah/ n/Wam/llD/t alabhanlo h•(/hd mai!cl!dlsu
samparivattitvG assasanto passasanto saya1• nia.
In the explanation in Th-a, samparivattakarp is a ~mul aboolutive in -a~.
See the note on 156 and EV II, p. 66 (ad Thi 26). ·
Dhp-a !'v 16,18 foll.: /daJtl vutta111 hotl: yaJd. purlso middhl ca hot/
mahagghaso ca nivdpapu((ho m~ >(ya ca ai!il61!a lrly/Jpathena
yiipetu'F' asakkonto nlddllyanasllo samparlwzllasdyl ladll so anlcca'F'
duill!U/1 analt/J ti ti~/ /aJ:J;Juv;4nl ~ 114 sa>J:otl, lts4'1' amanas/-
hiriJ mandapaiiiio punapplUJQ gabbAam upcd pb/iaYJ.sato na parlmuccaJ/
d.
We may agree with Rau (p. 173) that this verso about a mahdvardha is not
appropriate to the mgiivagga. Tho compiler pre;wnably thought that this
vagga was the most suitable for the inclusiollofa ~about a largo animal.
It is in the Yugavarga in Ud!na-v, which bu no mgavarga.
The metre is TrlWibh. Cf. Th 17.

326. Dhp-a IV 24,14-16: ajja ah4'F1 pabhbulaip manahatth/Jrl hat1h/Jcarlya-

sankhiito cheko onkusaggaho all.tusena •iya yonlsomanasJJ:iJnna
niggahissiimi, niissa vitiJ:kmnitu'F' dassdml ti.
For the alternation 'F'Plpp (NC/CC) in hattAl-ppabhlnna'F' (• hollhi'F'
pabhinna1'J), see the note on 31 and EV I. p. 141 (ad Th 77). Th 77 read•
niggahissiimi, which is a v.L in Ms L.
For the cognate accusativ~ acari cdrjkm[t, see the note on 108.
For -d as a historic form in a fossilised sandhi position in tad - t4'F1, see the
note on 72.
"D!~ Jl!etre is Jagati

327. Dhp-a IV 26,J0-!4: duggii ti yathii so pculh sanno kuiijaro hatthehi ca

piidehi ca vQyiimQl]I katva palikadugga/o attOnaJ!I uddJuiritvd,thale pad((hito,
eva,,. tumhe pi kilesadugg:ato attiin4'1' _uddharatha nibbiinathale pati{/hapetha
ti artho_ - -
 Notes 14 l

328-29. For the meaning "zea:ous" for nipalrn. see EV l. p. 143 (ad Th.85).
Dhp-a IV 29,5: s[Jd}urvihQridhiran ti bhaddaviharirrt paTJ¢ita17J.
I take siidhuvihJridhira171 to be a dvandva compound, made up of sadhu-
vihQri and dhira_ This seems to be the way the cty is taking it. TI;ic gloss
palJefitlY!J, continns the meaning "wise" for dhin:urr-

328. Dhp-a IV 29,5 foll.: parissayiini ti ___ sihavyagghtldayo piika{a-

parissayt:, r<igabhayadosabhayiidayo pa[iccha111U1parissaye cti Ii sabbe va
parissaye abhibhavitva.
PDhp 9 reads pariJrav[aniJ; Udana-v 14.13 parisravQl)i. Cf. ASokan
parisava and BHS pariSraya and parisrava (see BHSD, S.YV .). It would
seem that the original form of the word vras parisrava (< sru "to flow") and
the idea of ..flowing around" must be very similar to ds"ava ·~flowing in". It is
possible that parissaya is connected with Jain parissaha, where -h- is a glide
consonant, replacingy/v. For the alternationy/v,scc the note on 40.
For the eastern masculine accusative plural ending--&U of parissayQni, see the
note on 82. For eastern forms. see the note on 32
Dhp-a IV 29.9: attamano upaf!hitasati huJvii.
The metre is Tri~bh. In piida c there is resolution of the first syllable. In pada
d satimD is m.c. to give the cadence ¥ - - - •

32~30. Dbp-a IV 30,7-9: matango ti laddhanamo imasmi'!I aranne aya'!I

hatthiniigo yiirhOTfl pahtiya sahbiriyQpathesu elaJ:o suklzmp carari. evam pi
eko canyya ti attha. ·
Ps IV 206.12 foll. (ad M Ill 154,2J•): matangaraiUie va nago ti mataliga
araiiiie nQgo va. mOtaitgo ti harrhi vucca!i. niigoti m.ahtmtadhivacanam elfll!J.
In this view, therefore, matailgo niigo means "large elephant" and in pada d
matalig' araiiiie is for miilarlgo araiiiie. Most translators ignore the cty's
explanation, and either omit mC.ranga, or translate miiJJ::zngaraiiiie as ..elephant
forest ... i.e. the equivalent of nagavanassa in 324. C&P. however, who
translate: "like the elephant in the Miitatiga forest'" (p. 48), point out (p. 494
note 15) that in so doing they are going against the cty. They justify their
rejection of the cty on the grounds that the sandhi of -o +a-> -a- seems
unusual, and to support their transla'tion, they quote PDhp 10-11:
miilaf'!Jgtlranne, but this is no more conclusive than Pali mOtaligaraii.ii.a, since
it too can be taken in both ways.
For a discussion of the sandhi and other asped>ofthese verses, see Norman,
1993B,p)l- 157~2 and for an objection to my suggestions, see Wrigh~ 1996,
-- . p. 5R I believe that the evidence forthe inteiprd>tion of miitairg"aranne is not
sufficient for us to be able to say decisively whether we are dealing with two
separate words or a compound, and much· depends upon the value which we
place upon the commentarial tradition. In the translation I take nago and
ma.tango as being in apposition. as I have tal::ea kuiijarli and niigii in
apposition in 322, and I assume that miitanga is a parti~ar sort of elephant,
 142 The ~Vord of the Docrrlne

perhaps named after a geogrcphical location. See Norman, 196-0, p. 273. ~or
nJga, see 320. ,
In pada ct there is a redundant fifth syllable. See the note on 141.

329. Dhp-a N 29,10 foll.: rdjd va raf;han ti ralfhaT[I hitvd rajjato riijisi viya,
idam vutta171 hori: yathd vljitabhiimippade.So rdjJ idarrt rajjarr nOma
mahanla![I pamddaJ[h&iG.f!l, kif!l me rajjena kiin.tenJ ti vijitara[fham pa}Uiya
.. ekako va cara!l.
The cty is therefore taking tbis of a vlctOrious kirig who-later deCides to
become a wanderer. Cf. GD IL p. 149 (ad Sn 46). It seems, however, equally
possible that it refen to a king who has been defeated and is obliged to flee
from bis kingdom when it has been conquered by another king.
The metre is T~\ubh.

330'. Dhp-a IV 30,15 foll.: aya'fl hi sahayatd nCima yaT[I ba/am nissCiya
adhigantu.m na sakka ti. n • atthi bale sahiiyatii.
Ir is not cl~ar how the cty is interpreting biile - perhaps aS~ a locative "in a
fool". For the interpretation of bale as an instrumental plural~ Skl balais, see
the note on 61. PDhp 11 reads bale bitiyara; Udana-v 14.16 has Mia~
sahayaka~. with biilasahiiyata in the alternative version_
The metre is Anui\Ubh ab; Trii\ubb ed. In piida d vie should read
appassu[k]ko m.c. The confusion between -kk- and -k- arises from the fact
that the word may be a compound of appa with ussuka < utsuka or appa with
ossuila < autsukya. In pad.a c we should ignore the Sval"abhakti ·YoWi::l in
ka)ira.

331-33. I translate sukha .in these three verses as "pleasant". "Producing

11.appinesS:' would be better, but is more cumbersome.

331. Dhp-a IV 34,7-8: tasrr.5 ya itaritarena parittend va vi"pu!ena va tiltano

santakena santuf/hi ayam eva sukha ti artho.
For itaritara, see EV L p. 147 (ad Th 111).
Note the internal sand.hi of -a+ i- > -i- in irari!ara. Contrast BHS itilietara
(BHSD, s.v. itaretara).
The metre is T~\Ubh.

332. Dhp-a IV 34,12-15: miiueyyata Ii miitari sammDpafipatri,petteyyatii ti

pitan" .sammiipafipam·, ubhayena pi mtittipitunnGJ!l upaffhiinam eva kathilGJ!f.
Cf. Sv 920,31-32 (ad D III 145.27): matteyyataya ti miitu-lratabba-varre:
Dbp-a IV 35.J.-S: s&nannara ti piibbcijitesu. sanunDpafipatti. briihmailidtiFti -
bQhitapfipesu buddhapaccekabuddhastivakesu sammiipafipatti yeva,
ubhayena pi lesOJ!l catUhi paccayehi pafijagganabhiivo kathito.
We should note that stimaniia and brahmaiiiia are abstract nouns in their own
right: ·~he state of being a samal)a" and ''the state of being a brGhmal)a". The
 Notes 14]

addition of -ta (Udana-v 30.21 ir3mar;yatii and briihmanyatti) would ffiake

double abstract nouns. Such formations are not unknown in Pali, e.g.
piiramitii, where pJraml is an abstract noun from parama. It is, however,
possible that this double formation implies a different meaning, and MW
quotes brahmal)yatii from MBh with the meaning "friendliness towards
Brahmans. piety"_ Simil3.rly we could take siimannara as "friendliness
towards sall1al).as" _The origin of the f onru matteyyatii and petreyatii (PD hp
66 mtitreatii and p~tt~ata) is not clear. In Udana-v 30.21 mOrrvyatii aod
pitrvyaJO cannot be historically correct. The Pali forms are perhaps 10 be
derived< •ma.treya-td and •paitreya-tii. They presumably represent sinUlar
double abstract fonnations from mlztr and pi tr- The ldea of "proper attention
to, respca for" is supportod by the cry.

333. In pada c there is resolution of the sixth syllable. Jn pada d there is

resolution of the foulth syllabic.

334. Dhp-a fV 44,2-4: so palavati huriihuran ti so ta!}htivasiko puggalo

b/unie bliav< uppalavati dMvati.
For hurdluua111. =EV I. p. 189 (ad Th 399).
For miilwii, S<C 162.
I take icdsa'!I to be from ips- ·~o wish to obtain", although Udiina-v 3.4 mids
icchann iva. PDhp 137 reads phalamqi.
The metre is Vaitaliya. In pad.a c palaVati is m.c., but the pada is one mora
short. We could read piilavati m.c., although the lengthening of a svarabhakti
vowel m.c. would be unusual. In EV I, p. 189, I suggested reading_up/avati.
PDhp 137 reads sii (LC. ta~hii) priiplavate; Udiina-v 3.4 reads sa hi
saf!Jsarate. Rau (p. 173) reads plavati and omits manujassa and vanamJd as
glosses, giving a SJoka verse. In p2da d the omission of final -'!l in vanasmi is
m.c. Sec the note on 143A_

335--36. In piida asahatiis m.c. to give the - - - • (pathyii) cadence.

335. Dhp-a IV 44,15 foll.: yathG niima vane punappuna vassantena devena
abhivaflaI!t biral)lU[t tiJJG:f!l vacf<.f.hati, C"V01?1 tassa puggalassa antova{!a-
miilakii sokii abJUva441umti ti attho.
Note the v.11. abhiva/fal!t, abhiva¢efha'f', abhivu!fha'!' for abhiva{{ha (< Skt
abhivffia). PDhp 138 has ova{{hii; Udiina-v 3.9 bas avavnfii.
Forvisattilii as an epithet of ta~ha. see the note on 180.
FOi1ilfuiiQ, seetlic..note on 337. . .. . .

336_ Dhp- IV 45,5-8: tamhii puggalii va{{amiilakii sakii papatanfi, yatha

n_iima pokklrare paJwnapatle patilarrt udabindu na patif!hOti, evaJ!I na
patitfhanti ti aitho.
144 The Word ofthe Doctrine

For the reference to the phenomenon that water docs not cling to a lotus leaf,
cf. 40 I.
For ca= tu "but", see the note on 5.

337. Dhp-a IV 45,8-9: 1117[1 vo vadamf ti tena .l:ara~ena aha!f1 tumhe vadami
bluuidam vo ti bhadd117[1 twnhaka!f1 hotu.
PDhp 140 reads ta'!' vo vademl bhadraJ?l vo; GDhp 126 ta yu vadami
bhadranu; Udiina-v 3.11 tad val vadaml bhadrQJ[l vo. The Udana-v reading
suggests that the first vo in pad.a a is for ve < Skt vai, but GDbp yu supports
the view that both occurrences of vo in pad.a a are the second person plural
pronoun, as is vo io pada e. See Liidm, 1954, § 24.
For bhaddam vo as a quasi-vocative, see Lii.dcn, ibid., and Brough (p. 264 ad
GDhp 286).
Dhp..a IV 45,15---18: mQ vo na/af!J va soto va Mii.ro bhaiiji punappunan ti mi5
twnhe nadlso!e jiitQJ"(l nafa'?1 mahiivegena G.gato nadisoto viya Kilesamiiro
Maral}amdro DeVapunamiiro punappunClJ?l bhaiijatii ti attho. ~ .
Note va twice in pada e: "May Mara like a stream not break you like a reed".
For the negative injunctive ma bhaiiji, see the note on 133.
As Max Muller states (1881, p. 81 oote ad 335), usira (Skt uiira) is the
fragrant root, said to have medicinal properties, of biral)a grass (Skr viri7Ja)
Andropogon muricatum.

338. Dhp-a IV 48,6-8: chedanaphalanapii.'f)akavijjhaniidina1!J kenaci

upaddavena anupaddave.
Because the final -r in pwuir is historically correct, we printed punar e:va in
OvH/KRN. To be consistent with analyses made elsewhere, however, I now
regard it as puna-r-iva. For the sandhi -r-. cf. Grogge-r-iva 401; siisapo-r-iva
'407, and &ee GD II, p.142 (ad So 29).
For -U- in aniihata, see the note on 250.
The metre is JagatL -In pada d we should read nibbattatf m.c.

339. Dhp-a IV 48,13-18:yassa puggalassa iii a;]hattiknss' upiidiiya

a!fh&-asa !aIJhOvican·rani biihirass' upOdii.ya a{{hiirasa ta!)hii.vicaritGni ti
imesa'?1 !a!Jhiivicaritii.naf!t vasena challif!tSO!iyii sotehi samanniigatii.
maniipesu riipiidisu assavati pavattati ti maniipassavani! !a7Jhii bhusii.
balavati hoti.
Foc the 36 streams, see C&P (p. 497 note 9).
Th 76-0 reads kuddi!fhbp in place of duddif{hilfl.
ln_pada a chattirrzsati _is m.c.

340. Dhp-a IV 49.8-.11_: ~an ca disvQ ti Ia1J1 pa_na taT)hiilataT(I ettha sii.
uppa_fjamii.nii uppajjatf_ti jiita{!-h__iinavasena disyii, pa_iiiiilyii ti .satthena vane
jatalata'!1 viya maggapamifiya mU/e" chindathii ti attho. _
Th 761 reads sabbadhi, which is a v.l. here.
 Notes 145

34 l. Dhp-a IV 48,11 foll.: saritiinf ti anusafiini payiitGnl, sinehitGni ti

civarddisu pavallasinehavasena sinehitfini ca !aJJ.hiisinehamakkhitilni ti allho,
somana.ssQnl bhavanti tar;hiivasikassa jantuno evBriipiini som,anassiini
bhavanti.
I take somanassilni as the subject, and sGritiinl and sinehitlini as adjectives
describing somanassO.ni. The cty's explanation anusafa supports a derivation
< sar- "to flow'' for sarita, rather than < smar- "'to remember".
Dhp-a IV 49,15-18: te siitasitiJ ti le ta1JMvasika pugga/a sGtanissitii
sukhani.ssiIJ ca hutva sukhesino sukhapariyesino bhavanti, te ve ti ye evariipO.
narti te jiitijarO.maraJ)O.nl upagacchantl.
For siita "pleasure", cf. Skt lex. siita, BHS Jiita, siita, and see Brough (p. 240
ad GDhp 186), and cf. sacca 224 408.
In p3.da c instead of te UdJ.na-v 3.5 and PDhp 148 read ye, which seems
preferable, since there is nothing to identify those referred to as le. I translate
ye.
The metre is Vaitiliya. In p3.da b we should scan somanassiini as - - - - - , i.e.
s6ma/ltlllQni m.c.

342--43. Dhp-a IV 49,19-22: paja ti ime sattiJ tiisakara~ena tasi~ii ti

sallkhQ'!I gatilya ta~hiiya purakkhata parhii'lri!a hutva, biidhito ti luddena
arafiiie baddho sasa vrya S"'T'Sappanti bhO:yanti.
For pura1:khata - "subjected to", see EV I, p. 130 (ad Th 37).
For biidhita see Norman, 1979C, pp. 37~38.

342. Jnpada d ciriiya is the dative of time.

The !lldre is Vaitfillya abc; Aupacchandasaka d.

343. Dhp-a IV 50,6-8: attano viriigaf!J riigadivigama1!J nibbJna1!1 patthen"to

iikarl.khamiino bhiillu arahattamaggena tatrr tasir;a1]1 vinodaye nuditvii
niharitvii cluJ#eyya ti aJtho.
The reading iikankha looks like an imperative, but the v.11. akankhanta and
iikaiiklri, and the cty's explanation by patthento akankhamDno, suggest that
iikmi.klra is m.c. for aka.ilkhal!l. i.e. the masculine nominative singular of the
present participle.
The !lldre is Vaitfiliya. In pada d we should read bhik[k]hu m.c.

344. Dhp-a IV 53,7-D:yo pugga/o gihibhiive iilayasankhiita'!I vanathani

chatf.r!dva pabbajitataya nibbanatho vihGrasaiUliate tapovane adhimutto
_. .g_~_ariivGsab4ndhanavan_asarikhiitii ta1:1hiiv_an~ mutto hutvii puna tarri
ghariivii;~&mdh~Mta~havana;,. eva dhaWiri evdiii tani puggalai;i prusatha,
eso ghariiviisabandlumato mutto ghariiviisalxvuihanam eva dhiivati.
The cty tells a story of a young man who entered the Order, but later returned
to domestic life, and explains the verse in the same way. This does not,
however, take account of the word-play on the meanings "forest" and "desire"
 146 TJ-:e tVord oflhe Doctrine

for vana. Cf. 283. For word-plays see the note on 9-10. It seems more
likely that the Vcr5e is an attack upon brahmans, here the viina-prasthas, who
leave thel!" house and family and live an ascetic life in the woods. Because of
the word-play, the going to the forest can be interpeted as running to desire,
and therefore running to bondage. For such attacks upon brahmans, see the
note on 266. For adhimutta "intent upon", er: 226.
C&P read etlui for e:va in piida c (p. 498 note 13), as does Ch. PDhp 151 has
etha; Udana-v (N) 27.29 has eva (there is no(B) version): Dhp-a read.s eva in
the verse and ev<IJ?1 in lhe expla:.'1ation, but m::ords the v.1. etha in both places.
For etha passaIJ..a see 171 We may assume that eva has replaced etha as a
result of em in padas bd. and I translate etlui.
The metre is Aupacchandasaka a; Vaitaliya bed. This is very unusual, but it
seems genuine.

343--47. Dhi>-a JV 56.1-2: dhira ri buddhodayo pan<fitapuris3.

This gloss confirms the meaning "wise'" for dhfra.

345. Dhi>-• TV 56,J-<: babbaja1inehi ... rajjlD?' katv3.

It is possible that pabbaja should be derived < Sid parvan; cf. parvayoni
"growing from knots or joints, a cane or reed". Alternatively, it may be
derived< babbaja (Sid balbaja) "Eleusine indica (a species of grass not liked
by cattle)". If we assume that pabba-ja is parallel in structure to d&u-ja, then
we are dealing with babba. For the p!b alternation, see the note on 149.
Dhp-a IV 56,6: siiraJtaraitii Ii siira!lii. hutva MfharOgaua ti attho.
For siirattarattii., cf. EV I, p. 162 (ad Th 187).
The metre is T ~\ubh.

' 346-47 • .Dhi>-a IV 56.20 foll.: kilesabandhanaf!1 n3~amaggena chinditva

anapekhino huJvri.
The metre is Tri~~h. There is resolution of the first syllable in pad.ad.

346. Dhp-a IV 56,13: sirhilan ri baddha!{h&ie chavicammama1!f.sii.ni na

chindati lohitOJ!I lUl nfharati, baddhabhiiva'!l pi ajanapetvO. 1halapa1ha-
jalapaihadisu J:.ammiini katu'!I na detf ti sithilllI!I.
Translators have had problems with .silhiial!f as an epithet of the fetter, and
have had to devise alternative meanings to the usual one to make it fit the
context. The expected meaning is "slack", and it is used of slack vf~a strings
at Ja II 249,10, and of a saddle being loosened at Ja I 179,!8. The cty tries to
give a meaning which is appropriate.. Udana-v 2.6 has susrhiraf!l, which was
accepted by Lilders (1954, § 82), but rejected by Brough (p.230 ad.GDhp
170) following Mehendale (1955-56A, pp.66 foll.). PDhp 144 sukhumam
goes against susthira. GDhp 170 has .iiiila which must stand for .iithila.
Brough (p. 231) suggested taking-sithi/a- as a compound with the following
word. but was doubtful-about the metre. This problem cari be overcome by
 Noter 147

assuming that sithilaJ?'I- is a compound with-'!!- m.c., to give the break - .- - .

See the note on 49. Cf. the use of safhila with reference to an action in 312
and to a paribbaja in 313.

347. Dhp-a IV 59,t-4:ye sallJ ragararrJ dosadu((hJ mahamu/hJ sayam-

kata'!1 lal)hdsolaf!l anupa.tanti, te la~ samarikl:nmillUfl na salionti, cva'!J
duratikkama'TJ etam pi chetvii.na vajanti dhirQ.
For the suggestion that a word-play is intended on the two meanings of .solt1J!1
(< Skt srotas "stream" and JVa-ota "self-woven"), see Brough (p. 233 ad
GD hp 171 ). For word-plays, see the note on 9--10. Brough ca!Jllot be right in
thinking (p. 232) that ye in pada a was intended as a neuter accusative. For his
argument to be correct the "original .. reading would have needed to be Y'21!t.
which was changed to ye by someone who did not undc tand the verse.
Since ya1?1 followed by etarrs would be a more elegant construction, it is hard
to see why anyone should wish to change il See also C&P (p. 499 note 22).
For the meaning "cutting across" for chetviina in association with sota see
Mehendalc (I 955-56A, p. 72). Cf. 383.
The metre is Trinubh. In pada d we should read du[k)kham m.c:to match
sukha'!' in 346d.

348. Dhp-a IV 63,1-5: muiica pure ti atitesu khandhesu alayam nikantim

ajjhosiina1f1. pattlumarrs pariyur!hiinal{t giiharrs par&nasmri taTJ,ha'!' muica,
munca pacchato ti aniigatesu pi khandh.tisu ii!ay&/fni mwlca, majjhe muliru ti
paccuppannesu pi tani munca. .
For paragii, see the note on 8.5. For-gii < Skt -ga, see the note on 7.
For the future in -h- and the ending -isi in upehisi, see the note on 236.
For pure ... pacchato ... majjhe, cf. 42!.
The metre is Vaitaliya

349. The metre is Vaita:Jiya. In pada a we should read -(pa]mathitassa with

Ch and Ms E to get the syncopated opening - - - - - . In pada b note the
opening - - - - - - . In pada d we should read klto to get the syncopated
opening----- ...

350. Dhp-a IV 69,2-3: vyantikiihin- ri esa bhikkhu fisu bhavesu uppajjamika-

ta1.1ha1!1 vigaltI7Ji kari.ssati, Miirabandhanan ti esa tebhiimak.ava{(asarikhii!af!1
MiirabandhanlUfl hi chindissati ti attho.
For the future (c)c,l,echati < Skt chetsyati, cf. the discussion of bhijjati 148.
Forca~tu~ut",seethenoteon5. -- .
For the consonant group vy- in vyanti-, see the note on 1-2.
For the furure in-lz- and the ending -iti in -kiihiti, see the note on 236.
The metre is Vaitfiliya_ In pada a we should read -upasame, to get the
syncopated opening· - - - - . In pada c we should insert a svarabhakti vowel
and read viyantikiihiti m.c., as Fausb0ll suggested, to get the syncopated
 148 The Word of the Docrnne

opening - ~ - - _ In p!d.a b bhQvayatl is ITLC. In p3<la <l cch- in cchecchati is

m.c., to make the socond syllable of <.ra long.

3.51. Dbp-a IV 70,14-18: ni/(ha711 gato ti imasmi'J1 siisane pabbajitJna171

arah.a!!al'fl ni11M n&na, ta1]1 gato patto ti auho, asantiisi ti abbhantare
riigasantdsdd!na111 abhclvena asantasanako. acchidda bhmiasalliini ti
sabbanl pl bhavagi!mlnl sal/iinl acchecchL
For acchidda, cf. acchidJ Sn 357. Ms E ofDhp-a has acchinna in the cty.
Presumably-dd- i> m.c. to avoid the opening· - - • .

352. Dhp-a IV 70,19-21: niruttipadakovida ti niru11iyaii ca sesapadesu ell ti

catilsu pl pafi=z1bhidasu cheJ:a Ii attho.
I take pada b~ IO mean "word". See the note on I.
Db~-a IV 71,i-12: maMpaniW Ii save esa ko(iyaf!l [hitasarira mahantlinl17!'
atthadliammaniruttipa(isambhidlllll1J/I silakkhandhiidinaii ca pariggiihikiiya
pafiiiGya samanndgatattd mahapaiiiio, vimul!acitta171 khviihmp Sariputta
ma/Uipurlso Ii vadilmI Ii vacanato vimuttacittatiiya ca mahiipU:riso ti vuccali.
In piida f there are thirteen syllables; mahiipuriso is probably redundant,
although the cty includes both mahiipaiiiio and mahiipuriso in the
explanation. Possibly mahiipuriso has been included in the text because of its
ocrurrcnce in the cty. It is not in PDbp 147.
In pada e -sllriro is m.c. to give the· - -· (pathyii) cadence. Cf. 400.

353. This is the verse which the Buddha uttered when asked who his teacher
was.
For the consonant group -sm- in asmi in piida a, see the note on 1-2.
The metre is Triwibb. In pilda b aniipalillo is m.c. In piida c the metre is
'improved-by reading -[k)khaye.

354. The= is Tr4!Ubb. Note the opening - - - - in pada a. If we followed

the pattern of padas be and read sarva'T' dlina'T' we should still have an
unus~ opening.. The openings of the two versions of the p.3.da at Udana-v
26.31 (jayate diVuvrz and sarvaJ?I diinDJ?f) are also problematical.
In pada c -ratiis m.c. In pada ct the metre is improved by reading -du[k]kham.

355. Dhp-a IV 80,1-2: no ve pi3ragavesino ti ye pana pD.ragavesino puggalii

na le bhogli hananti.
I should prefer to read ve force in piida b (cf. the v.L). For the cJv alternation
see the note on 50. Alternatively we could read ca (cf. the v.l.) which seenu
to be supported by the cty's pana. This agrees with. Udana-v 1.16-n·a .,.,--
ihiitmagav~i1)£1m. Th 771-73 reads-no-· ca paragavesino, but the rest of the
verse differs.
Although Udana-v has (iitma-)gave;i~am, the existence of Pali piira-gjj, etc.
(see the note on 85) suggests that piira-gavesin is in fact to be derived from
 Not"' 149

-gamesin, i.e. it is a future active participle. For such participles, sec Geiger,
§ J 9JA. wh= the reference to Dhp 55 is an error for Dhp 355. For the mlv
alternation, sec the note on 183.
Dhp-a I\' 80,2-4: afiiie va attanan ti _bhoge nissJya uppannJya Ja!Jhtiya
duppanno puggalo pare ·Jiya attJnam eva hantf ti attho.
For a11ana7rJ m.c. in pAda d, possibly helped by the existence of the weak
grade form found io atrana, see OvH, SP, p. 166. Udana-v 2.16 retains the
correct Skt form by changing the word order and reading hanry ii.tmtinarn
atho paran. Cf. artam 379.

356--59. For mahapphala'!'I < •mafwt-phalaf!1, cf. 312 and see the n(){e on
123. There is a play upon words in these verses. Just as a seed planted ia
fields ·hich are free from defecU, Le. free from weeds, brings a rich rrsu!t, so
a gift to those who are free from the defects of passion. hatred, delusion and
desire brings rich fruit.
For the consonant group -sm- in t.asma in pad.a c, see the note on 1-2.

357. Nore that in dosadosa we have both meanings of doso, i.e. <Sia dvqa
and Sia daµi_

360--61. For sa'!'varo with the instrumental case. see the note on 225.

362. The metre is Vaitfiliya abc; Au!"'cch•ndasaka d. The metre of ab is

irregular. In pilda a we should scan hattha-sa[Ji]fiato, and in pada b so[li]iioto.
See the ..note on the scansion of saiinatassa in 24. PD hp 52 reads
vacasa1!Jyyato in pada b, which would need to be scanned vlicJsaT?tyyatO.
Udana-v 32.7 reads vacdsaJ!IY<l~; GDhp 53 wzya-saiiodu.
Reading samlihitalto in pada c would give an Aupacchandasaka pada to go
with pada d, although this is probably not necessary, since mixed pairs of
Vaitiiliya/Aupacchandasaka padas arc common. We find samiihilatto at Th
981 (for the erroneous prefix su-, see EV L p. 261), but none of the versions
of this vase supports such a suggestion.

363. For manta-bhii~in < manda-bhii~in (~Sia mandra), see EV I, p. I 17

(ad Th 2). and cf. mita-bhii~in Tl.1. For the ntlnd alternation see Geiger,
§ 61.L
Dhp-a N 93,8-14; a1tha111 dhanunaii ca dipeti ti bhlisitalthiiii c' eva
desanadhafunafi ca katheti, madhuran ti evariipassa bhikkhuno bhasilarrz
madh.urCll!I ndma, yo-pana attham -eva .sampOdeti na piili'tl~]JQlirri yeva
samp&:leti na attharrz, ubhayal[I vii pana na sampiideti tassa bh0sital]1
madhurtlJ!l niima na 1ioti.
For the opening•· - •in pada a,~ the note on 1-2.
364;"-Dhp-a IV-95,I-2;-.nivllsanatfhena samathavipassanlidhammo iirlimO
assii ti dhammiir&no.
 150 The Wordofth•Do~

There are nine syllables in pJ.da d because of the negative.. See the note on 7
We can correct the metre by asswning resolution of the fourth syllabk:.

365. Dhp-a IV 97;J: saliibhan ti attano uppajjanakaliibha'!I.

The cty is, then, taking sa as< Skt sva.
For ihe sand.hi of na + ai1iiesa1]1 > ni1Riiesa17f, which violates the two morac
rule, see the note on 70.
For pihayaf!i, see the note on 94.

366. Dbp-a (E') IV 97,6• prints nMhigacchatl for niilimaiiiiati by mistake

367. Dbp-a IV 100,1-7-18: mamQyitan tiyassa ahan ti vd maman /j va gt5Jio

n' anhi.
It ~possible to take mamJyita as a past participle used as a noun, i.e. as
"cherishing" rather than "cherished". For past participle3 used as a nouns. sec
the note on l 04.
Dbp-a TV l 00, 18-24: asatd ca na socatf ti tasmin ca nOmarilpe khayavayflI!I
pa.Ite mama riipairt khfr;a'!' ... pe ••• mama viiiii&;u1J?1 khfl)On ti na socati na
vihaiiiiati, k.hayavayadhamma1f1 me khil).an ti passati, save ri so evariipo
vijjamiJne pi nJmarUpe mamiiyitarahito, asatii pi Cena asocanto bhikkhU ti
vucca.ti.
In pad.a c asata is the negative of the instrumental singular of the present
participle of the verb as- "because of something not existing". For grieving
because of something which does not exist, see MI !36,11 foll.
For the consonant group -sm- in -asmi('!I) in pada a, see the note on 1-2.

, 368. I take pada tom= "place". See the note on 1.

· Padas bc<I ~ 381bcd.

369. Dbp-a IV 108.9 foll.: siiica bhiklchu imG!!' n.7van ti imam attabhiiva-
saJikhG.tlJ-'!' niiwzi?i micchiivitakkiidakm/i siiicirva cha<f4ento sinca, sit!tl te
lahum essati ri yatlia hi mahii.samudde udakass' eva bharitanGvii chiddtlni
pidahitva udaknssa sit!attlya sittii sal/ahuka hutvii samudde anosiditvii
sigha'!" supa{!antJl!f gacchati, eYaJ?1 ••• nibbanmri. gacchati.
For siiic- ·~o bail our, see GD 11, p. 298 (ad Sn 771).
For the future in -h- and th; ending -isi in ehisi, see the note on 236. Udana-v
26.12 has bhavcyati I cyasi; GDbp 76 has bhqidi I qidi; Mvu ll! 421,I0°-
11• has bhl!!)'ati I bhqyati; PDhp 57 has hehiti I ehisi, i.e. the other four
versions have forms from bhii- in pada b. The Pili version has probably lost
h- from hessati: "Then it will be light for you".
I take /ahum to be an adverbial acusative, but versions with forms from bhii-
presumably have lahu agreeing with navii; cf. Udana-v laglnf, PDhp iaghu.
We may assume that the Pali red<ictor added the atil.i.SY3ralo turn 'it into an
adverbial accusative.
 Notes 151

370. Dhp-a IV I 08,20 foll.: panca chinde ti he{{hti apiiyasamplipahini pane'

orambhGgiyasaf'(fyojaniini pOde baddharajju'!I puriso satthena viya he{fhQ
maggallayena chindeyya, pai.ca Johe ti uparidevalokasamplipalani paiic'
uddha!]"lbhJgiyasarriyojan.Jni puriso g1-v3ya baddharajjukalf1 viya arah.aua-
maggena jaheyya pajahe;ya chindeth' evG Ii auho,paiica vuttaribhiivaye ti
uddha'!'bhGgiyasa'!'yojanJna'!' pah3natth3ya saddhiidini pancindriyOni
ultarif?l bhO.veyya. paiicasailgii1igo ti eva'!l sante paiicanna1!1 riigadosamoha-
mQnadi((hisaTtgQna'!' atiilamanena paiicasatigGtigo hutvJ bhiillu oghali'}IJO
ti vuccati.
For v- in vuttaribhlivay<, see EV l, P- 123 (ad Th 15).

371. Dhp-a IV 109,14-16: ma bhamassU ti paiicavidhe ca te hi.magu'}e

cittllI!' m.ii bhamatu.
There is a MS tradition supponing rhe reading bhavassu, but in view of rhe
explanation by bhamalu \l.'C must assLUTie that the verb is bham-. although
Brough (p. 195 ad GDhp 75) favours bhave5SU. PDhp 33 reads ma YO
kiimogu~ii bbrame'!Uu di/OJ!!, which agrees with GDhp 75 ma de kamagu~a
bhametsu cita: "may the kiimoMas not make yous mind wander'". An earlier
form of the Pali reading must have been bham"J!1SU (third person plural aorist
of the causative bhramayati) which became bhamessu by'"the common
interchange of ·'!15- and ·55· (for NC/CC see the note on 31), and was then
changed to bluimassu, because the -essu ending was unusual in the dialect.
When the subject changed from third person to second person, k.amagur:iO.
was changed to kamagu1Je. The Udina-v reading suppOrts the view that the
kamagu1Jas were originally the subject of the clause. For the mlv alternation
see the note on 183. For a discussion of some of the .difficulties in this verse,
see C&P (p. 503 note 9).
In pad.a a the metre requiresp&nado, i.e. the second person singular aorist of
pamajjati, but all editions known to me readpamiido. See EV I, p. 149 (ad Th
19). The required reading is now attested in a Ms from North Thailand.
For ma piimado, ma gili and ma kandi as aorists used as negative injunctives,
see the note on 133.
For gu~a at the end of compounds, see the note on 53. I usually translate
kamagu1Ja as "'strand of seD.SU2l pleasure". as though gul)a had the meaning
''rope, cord" here, but this is not correct, strictly speaking.
The metre is Vaitallya a; Aupaccbandasaka bed. Jn pada a there is the
syncopated opening - - - • . The metre of pada d is incorrect, but can be
corrected ht exclucling mii m.c., and either assusning that the force of the
particle ma is carried over from pada c. ·ot th:afkana7 is in origin a present
participle "aying out.., with -i < -e. For the present participle in -e. see the
note on 281. If mii is retained, the metre can be corrected by reading kandi
du[k]kham.
 152 The iVord of the Doctrine

372. Dhp-a IV 109,.11 fo!L: paiif.Q n' auhi ti ajjhiiyantassa sam6.hiro bhiillu
yathiibhi1Jl1I'(l jO.niili passari ti vuua/ak.khal)ii. paiiiiii n 'atthi.
In pii<Ll b ajhayaJo is the genitive singular of the preseot paniciple ofjh[Iyati,
with the negative a-. The expected -jjh- (< Skt a-dhyiiyato) is simplified to
-jh- m.c., to give the cadence - - - • . PDhp 62 has ajhayato, but Udiina-v
3225 has to change the word order to accommodare nii.dhyiiyaJo.

373. Dhp-a IV 110,5 foll.: suiiiiiigiira'!I paviffhassii. ti kismicid eva

viviitokiise kiiyagh.ii!U1J?1 vijahitvii kammaf!hGnamanasikii.rena nisinnassa.
Dbp-a IV 110,8: samm.a ti herunQ kiiral)ena dhammam viprusantassa.
In piida d Udana-v 32.9 bas the plural dharmam, and PDhp 62 the singular
dhanni:zlfl. For dhamm.arri as a possible accusative plural see the note on 64--
65. Since PDhp supports the Pili reading, I translate dhamlnlU[I as a singular.
Db.p-a IV 110,9-10: amiinusi ti a{{hasamiipattisankhiitii djbbii pi rati .hoti
uppajja!i Ii attho.
[n piid4 c raJi is m.c. to give the - - - • (pathyii) cadence.

374. Dhp-a IV 110,10 foll.: yato yato sammasati ti ... yasmi'!I yasmi'!I
attana abhirucitahi.le ahltiru.citakammaf[hiine knmmClJ?I karonto sammasati.
!n piid4 c labhanis m.c. to avoid the opening· - - ·.See the note on 1-2.
For amaJa "death-free-, see the note on 21.
For the consonant group -l)'- in --..yaya, see the note on 1-2.
Note the plural viji'lnatm/t, in conrrast to the singular vipassaJo in 373.

375. Dhp-a fV II 1,10...13: mitte bhajassu kalyii~e ti vissaf/hakammante

appatiriipasahaye vajjetva siirajfvitiiya suddhiijive jarighabala1r1 nissiiya
jivitaJ.:appaniiya obmte a1andite kalyal)lI111itte bhajassu sevassil Ii anho.
"': For !a/ydl}a mitt.a see the note on 78. Udana-v 18.6 reads mOtraI]t bhajeta
pratiriipaf!J iuddhEjfvo bhavet soda; PDhp 64 mitre bhajetha ka/liil)e
Suddhii:Jivi atandrito; GDhp- 60 mitra bhaye 'a patfiruva S-.idhayiva atadridi.
The Udana-v reading mii.tra1!% pratiriipa'!I ..suitable measureH seems to be a
ref=ce to knowing the limit (cf. bhaja~amhi mattaiiiill'!I in 8) rather than to
knowing friends. There seems to be no reason why an original mitra/mina
should become miitral(marta), so the Udina-v version is inexplicable if it is
secondary. In the ASokan inscriptions and some Pkis. however, mitta is
found in the sense of mDtra, preswnably being derived from a weak grade
form< "'mi-tra, cf. the pasf participle mi-ta <ma-. If an earlier version of this
veise had the equivalent of pratirUparrr "'mitram. the'::l we can see that GDhp
retained the reading, the_ ~dana-v redactor co~_i!y_ in_tC!J?t:~!_~~--•mj!!q_ap_d
replaced it by matra, but the Pili and PDhp re<Llctors knew mitta only in the
meaning "mend". Tuey therefore changed pa[irlipa to kalyiil)ll, which is more
usual in this context -
In support of the.view tlJ.at suddhajive and atandite are eastern nominative
singular forms in -e (in -agreement with the stibJeCt of bhajassu), cf. the
 Notes 153

singular forms suddhafivi'li and atandita'fJ in 366. For eastern forms see the
note on 32. ·
In pada b PDhp 63 iha pramiiassa supports ldha, but GDhp 59 tadha
praiiasa and Udina-v 32.26 tathii prajiiasya suggest a reading tathii.
For the consonant group -ly- in kalyJ1:1a in pad.a e, see the note on 1-2,

3 76. Dbp-a IV 111,IJ-16: patisanthGravuu' assa ti ii.misapa{isanthiire ca

dhammapafisanthdre ca sampannavuttitGya pa{i.santhdravutti assa
pafisanthdrassa kJrako bhaveyya ti attha.
For Sl1. pratisaf!t.Slara MW gives the meaning "friendly reception" and
BHSD "distribution severally (of gifts)". Brough (p.190 ad GDhp 60) is
probably correct in seeing a combination of the two and in suggesting: "the
s~reading out of gifts for mutual exchange, and in particular between host and
guest".
The reading -vutt' is supported by PDhp 64 -va{!i and Udana-v 32.6 -vrttih. It
may mean "having distribution as a way of life", i.e. it may be a
bahuvrihi rather than a tatp~a compound. GDhp 60 -guti may 1x: the
correct reading. or it may be a backformation guess from -vutif, on the
assumption that -v- is a glide consonant< -g-. It may, however, be influenced
by-gori in GDhp 59. Some editions read vatty-assa orw(Y-assa. where the
sandhi of -tti > -{t)ty lx:fore a vowel is a Sanskritism.
Note assa < • asytlt in pada a. but siyQ m._c. in pada b. See the note on 10.

377. Dhp-a IV I 12,19: maddaviini ri miliitiini.

378. Dhp-a IV 114,16: ahhijjhiidi1JaJ11 ahhiivena santamano.

It is not clear whether santava in pada b is a past paniciple with -van/ , or<
-vat. We should expect santamano, which is what Ms L reads, and which the
ct'f seems to 1x: reading in the explanation.
The tradition is confused here. PDhp 53 reads iii.ntakayo Sii.ntacitto iii.ntavG
susamiihito; Udana-v 32.24 (B) Siintakiiyah iiintavii.k <ca cittena>
(conjectured by La Vallee Poussin, 1912, p. 376) susamiihita~. (N) confirms
siintaviiksusamiihital) (pada a is missing). Balk(I988,p.476) states that the
Tibetan suppons a Skt version itintakayal:z Sii.ntacittal; i5ntavii.k susamQhita~,
which agrees with the PDhp version (although the order is usually body,
voice. mind). Doubtless santava was thought to end in -va < -vant, and
susam.aJUta was thought to reflect the expected sanram.ana. so santavQco was
insatal to ~ve the usual triad.

379. Dhp-a IV 117,J-2: coday' attiinan ti attanii va attiintll/I codaya siiraya,

j.Xl/imiise ti attanfi va attiina'!I parivimCIJ?Jse.
Thect'j is taking coday' as the imperative codaya, but in view of the optative
pa[imiise, we should probably take coday' as the optative cadaye.
 154 The IYord o/Jhe Doctrine

Foe the future in -h- an<l the ending -isi in vihiihisi, see the note on 236 and
cf. GDhp-60 vihfllisi < •vihar~yasi "you will dwell".
For so with a second person singular future verb in pad.a. c. see 1he note on
134_
In piida b note at!a!JI for attiir.al!' m.c. Cf.. attan.aJ]2 m.c. in 3.55. In pad.a b
there is resolution of the first syllable.

380. Dhp-a IV 117,9-10: a/Iii va altano gati pali(!hii sarlJf!lVfl.

PDhp 322 iittQ hi iittano niitho OJtii hi iirtano gatf; Udan.a-v 19.14 iitmaiva hy
iitmano nG.tlralJ titmii iara1_1.am iitmar.alJ. For the canon in one tradition
agreeing with the cty in another. see the note on 66.
Dhp-a IV 117,15: attiinGip sGipyamaya gopaya.
In pada c saniiamay' is preswnably the second person singular imperative of
the causative of yam-. There is no menical reason for -ii- (cf. sainamessanti
37)_ PDhp 322 readssGipyyamay·; Ud.ana-v 19.14 SGipyamay' (which cannot
be an optative). MW lists only sGipyamayati for Sh
For the consonant group -sm- in tasrrul in p3da c. see the note on 1-2.

381. I take pada to mean "place". See the note on I.

Padas bed~ 368bcd_

382. Dhp-a IV 13 7,6: yunjati Ii gha{ari viiyamaJL

For yunjate (middle) with the locative, cf. ayogeyWijGip a~ 209.

383----423. For br- not making position in briihmm;a, see the note on 75.

383.-For akaJaiiiiii. see the note on 97. Here the reference to blowing the
destruction of the sarikhiiras (Sb saf!!Skara. where the connection with a-
lq-ta is more obvious) possibly hints at nibbiina being the place where there
are no lqta things, i.e. no sankhiiras "conditioned things". For -iiiiii < Skt
-jiia, see the note on 7.
For the meaning "cut across" for chindati when associated with sota, see the
note on 347.
Dhp-a IY 139, I J: mahantena parakkamena paroll.amitvii.
This would appear to be the explanation of parakkamma, ic. an absolutive,
but the cty reads parakka'l'a in both verse and lemma. Cone (1986, p. 112)
draws attention to S I 226,28* ito pafikkamma and Ja V 138,26• ito
parakkamma explained as imperatives (Spk I 346,6: ito pal):ama; Ja V
138,29·; parakkama apeh1), with -mm, m.c. (see Geiger,§ 136). PDbp 34
parfikriimma and Udana-v 33.60 parakramya suppon the view that it is an
absolutive. GD hp_ 10 parakamu is inconclusive.

384. Dhp-a-lV · 140, IJ ·foll.: yadii ti jiasmi'!I kiile dvidhii/hitesu .samatha-

vipassanCidhammesu abhiiiiiiipiiragiidivasena ayGl'[I khi')ii.savo paragii hoti.
 Notes 155

For pGraga, see the note on 85. For -gU < Skt -ga, see the note on 7. The
usual phrase is sabbadhammtJ.na~ piiragU. and it is not clear what c!Vayesu
dhammesu means. The introductory story tells of Siriputt.a asking the Buddha
wh.at the dve dhamml!. are and being told that they are samatha and vipassanii.
It is noteworthy, however, that dve. a:nd dvaya do not have the same meaning,
since dvaya means ..twofold" rather- than "two". Sn 886 states: saccam mu.rd
ti dvayadhammam Ghu, on which Nidd I 295,7 foll. comments: mayharr:
saccaJ!l ruyharri musti ti evam <lhaf!2su. This would seem to be a reflection of
rhe meaning "twofold nature, falsehood" found for dvaya in Skt (sec MW,
s.v. dvaya). If this is the idea here, then dvaya dhamm.11 would mean
somcthing like "pairs of opposilcs", of which tnnb and falsehood would be
one example. Another pair might be indicated by dvedhilpatha (to bhi:va and
vibhava) at 282. '·
The other three versions similarly have forms from dharma, but Udana-v has
a different accompanying adjective from the rest: GDhp dva 'qu, PDhp
dayesu, Uctana-v svqu. The Udin.a-v reading bas a parallel in Pili in Ud 5,&•-
9": yadii sakesu dhammesu paragii hoti briihma~o. which is explained (Ud-a
69,23-24):yadl!. saktu dhammesii Ii yasmi111 kii/e sal:a-attahhirvamnkhiitesu
pancasu upiidanallhandhadhammesu.

385. I talcc pl!.ra and apara to mean the next world and this world. For
difficulties arising fom the interpretation of piira as nibbana, see Brough
(p. 202 ad GDhp 86).
Dhp-a IV 141,12-13: kilesadaraJhiinam vigamena vitaddmwr<.
For the meaning "free from fear (or distress)" for vfta-ddara, see the note on
205. If dara is connected with daratha "suffering, pain, distress", then d >
dd is m.c. to avoid the opening•·- - • .

386. Dbp-a IV 142,IO-l l:jhiiyin ti duvidhena jhanena jhayanta111, kamena

viraja'f!I, vane ekakm,,.- 0sillf17!1.

387. Here atha ~"but". See the note on 85.

We need the sense of tapati in pada b, so \Ve may suspect that the finite verb
Jbhiiti (also in Udana-v 33.74 and PDhp 39; GDbp 50 bas avha 'i which is
no! the expected finite verb form) is a later addition.
There is resolution of the seventh syllable in pada c.

388_ Dhlj-a IV 14S,IO-t4: samacariyii ti sabbiikusa/iini samdvii cara~ena

tasmii ti yasmii biihitapii.patO:ya briihmaT)o __ akusaliini sametvd caraiJ.ena
samal)o ti vuccati tasmG yo attano riigiidima/af!t pabbii.jento vinodento carati
so pi tena pabbiijanena pabbajito ti vuccali ti attho.
Note that the etymology briilunw;a < biihita must have been made in a dialect
where briihma~a had become •biihana. Cf. the note on 267. ·
156 The Word of the Docirine

For samacariyd a3 the. instrumental in -a of an a-stem or i-stcm noun, i.e.

-carya or-carya, see Nonnan, 1974,pp. 172-73; and the note on 270.
The cty, followed by PED, gives an etymology based upon Skt foma, but
GDhp 16 has sama'irya, not Jama-. Asoka's Rock Edict Xlll(O) has
samacariyC1I[l at ShahbizgarhI, where the scribe woWd have written Sama- if
be had recognised that it stood for Sh sama-. See the note on soma in I 42.
This shows that the "etymology" was constructed in a dialect in which
frama(lll had already become SG11JlV!IL
1n view of GDhp parvahl'a, Udana..Y (B) pravrOjayitvfl, (N) praviihiya, \VC
might have expected an absolutivc "pabbajiya "having made ... go forth" in
pada c, rather than the present participle pabbajaylll!', although Th 349 has
sabbapapam pavahayirri.
Tue metre is Vaitaliya. :

389. Dbp-a IV 147,24 foll.: paJiareyya ti khi~iisavabriihma~o 'ha'!' asmi ti

j&umto khbJ.Gsavassa vd aMassa vd brahmal)OSSa na pahar_eyya.
See Brough (pp. 179-80 ad GDhp 11) for his explanation of this verse. I
follow him and C&P (sec p. 507 DOie 12), and the cty, in thinking that
brahm~o is the subject of both palrDrr:yya and muiU:dlra.
Dhp-a IV 148,1-3: nG.ssa muiiceJhO.ti so pi paha/o klu-JJ.asavabrahmal)o assa
paharitva fhitassa verwp na muiicetJa tasmirri kopal!l na kareyya ti attha.
For the sandhi of na + assa > niissa in pada b, and for the violation of the
two morae rule, see the note on 70•.
For the sandhi of yo + assa > yassa (y' assa) in pada cl. see the note on 10.
PDhp 46 reads ya ssa va su na mu<aJti; GDhp 11 tada vi dhi yo na mujadi;
Udana-v 33.63 dJliJ; ta'!I yd ca prai-ncati.
There is resolution of the sixth syllahlc in pada a.

390. This verse presents major problems, in that we do not know if in pada a
the two negatives (na and a-kiiia) Inilke a positive. or merely emphasise each
other, or if in pada b piyehi is positm: or negative (see below). Of the vme
Brough says (p. 180), "an editor··- will. if be is wise, refrain from trying to
translate the untranslatable".
_'Qhp-a I:V_ 148,7-11: eJad akinci sqyo ti YGJ?1 khil)iisavrusa akkosan!aT!J va
apaccakkosan(lJ!1 paharanlaJ?I Vii.- apap."ppaharal}-aJ!I eltu[J tassa khFl}ti.sava-
brohma~assa na lrinci seyyo (C&P [p. 508 note 18] read na akinci seyyo) ti
appamaltaka'!I seyyo na hoti adhimanam eva seyyo ti attha.
Piida a is corrupt. but GDhp 15 basbltodi and Udina-v 33.75 has asti. This
suggests that Sf<Y.YO is also from the mot as-, possibly a development< siya,
lengthened tosiydm.c. See Brough(GDbp, pp. 182-1!3).
1 follow the other versions (GDhp 15 y' asa; Udiina-v 33.75 hy asya) in
believing that there is no reference to hi'!lsa in pada c. and 1 Understand
hirrisamano to stand for hi ssa (-< tbcenclitic sma) mano, with the change of
hissa > hif{lSa. For the CO'NC alternation in ssll{IS. see the note on 1_9.
 Notes 157

Pada b is ambiguous, in that piyehi may stand for (a)piyehl, with a- .elided
afw manaso. The same is true of Udilna-v 33.75 yath?l priyebhyo and of
GDhp 15 ma1:1asa pri'a1:1i. We cannot therefore be certain whether it is
exclusion from pleasant or unpleasant things. Since there would appear to be
J;ttJe merit is r·1ming one's mind from unpleasant things, I as.Sume thal
pleasant things are meant.here.
For -d as an historic form in a fossilised sandhi position in etad"" e/Cl!f, see
the note on 72.
There is a problem with sammati-m-eva duilla'f1 in p3:da d. GDhp has
samudim aha saca~ Ud.ina-v (B) 5GJ?1V[!am ed duhkham; (N) SQ..1'[1V[tim aiti
satyam. I translate sammati as the passive of the verb Jam-, although the
other versions suggest that some traditions took it as the noun sammati (more
commonly sammuti), which was back-fonned'into Sl::t as saf!WTfl or SCJ1fVrfa-
For the sand.hi consonant -m- in sammati-m-eva in pad.a~ see the note on
34. Against Brough 's belief that there are no e:umples in Pali of sandhi -m-
after an indicative verb, see Norman, 1974, p. 172
In piida b I translate nlsedho as a noun, although Brough suggests (p.181) that
it was originally an optative nlsedhe.
The metre is Tri~\ubh abd; Jagati c.

391. For sa'!Mlta with the instrumental case, see the note on 225.
I take ;wsa as the genitive of the agent with the past participle -l:afa'!L

392. Dhp-a IV 151,13-t7: aggihuttal/I vii ti yathii /Jrahmano aggihutta'!'

sammiiparicara'l)ena c' eva balikammadihi ca sakJ::accarri namassati. eval!l
yamhii Gcariyii tathiigalappavedilal{I dhaml1U11lf vijlineyya ta7l1 sakkaccarri
namasse;-ytl ti attho.
Note that briihma~a is used here (somewhat inappropriately in the brlihma~a
section) of a caste brah.man. Note the development of -hutta < Sl::t -hotra via
-h6tta.
It appears that sakkacca'!' is an absolutive with -I?' added. PDhp 35 reads
sakkacca; Udana-v 3.6 sat/qtyainal!J (< sat/qtya enaT{J). For the extrnsion of
an absolutive by -l?f, cf. upapajja'!' (see CPD, s.v. upapajja),pecca'!' (see
PED, s.v. pecca), and see EV I, p. 294 (ad Th 1242).

393-94. Forja!ii, cf. 141.

393. We liliould pertaps have read sucifor sukhiwith the v.l. Udana-v 33.7d
reads sa iucir briihmal}'lh sa ca,- and C&P-{p. -508 note 20) read .suci here in
place of sukhi, despite Dhp-a JV 152,JO• and 152,14. I translate suci.

394. Dhp-a JV 156,12-15: abbhantaran ti ahbhantara'l' hi te riigiidihlesa-

gahanaf!t kevala'!I hatthilarp:jal[J assa)alJ.efa'!I viya ma[{ha1']1 bal:ira'!1
parimajjasi ti.
 158 The Word ofrhe Docrrine

Rau (p. 174) suggests reading garaha'T' for gaha~a'T' , but GDhp 2 read.
gaha~a and Udina-v 3.6 gahana'T'.

395. For the consonant group -sm- in -asmi('?') in pad.a c, see the note on 1-
2.
For a discussion of the tenn dhammani-sanrhata, see Hara. 1995, pp. 381
foll.

396-423. These verses also occur at Sn 620--47.

396. Dhp-a IV 158,llH9: bhovi!di ti yo pano iimantaniidisu bho bho ti VOJVii

vicaranto blwvadi niima so hoti.
Pj II 467,l-6 (ad Sn 620): yasmii bho bho ti vocanamattena aiiiiehi
salifican~hi visighattii bhovadi niima so hoti, sace hoti sakiiicano, yo
pant1ycuri yatzha katthaci kule jtilo pi riigiidikiiicaniibhiivena akiiicano ... tam
ah~ briimi brahm.tllJ.CITf1, kasmii:yasmii biihitapOpo ti.
Sec GD II, p. 263 (ad Sn 620). Translators are divided in their interpretation
of bho-vi!din, some tiling it as a sign of disrespect. The cty is non-committal
about this. In GD IL p. 71 l translated bho-viidin as "one who addresses
others disrespectfully". I am now not certain of this translation. In Skt bhavaJ
seems to be nsed respectfully and, despite PED, I would now follow C&P is
believing that bho-viidin means "one who merely knows proper etiquette in
greeting others, one who addresses others respectfully". At Ja VI 21!,J•
bhoviidi occurs as a synonym of briihma~a.
This verse occun at Sn 620 (see GD II, p. 263) with the reading save, and
Dhp-a reads save in both the repetition of the verse (IV 158,JJ•) and the
, explanation (IV 159,19), with ce as av.!. in both places. For the c/v alternation
~ see the n6te on SO. The version of this verse at GD hp 17, however, which is
written in the ~thi script where the a.Jq;aras ca and va are not con.fused.
reads sayi. This might be thought to support the reading sace, since
intervocalic -v- would not be elided, but it is not impossible that yi =the
emphatic particle ye-, which has the same meaning as ve. Udana-v 33.15,
however, reads sa ced and this is supported by the Tibetan Udina-v reading
gal-re. See Brough (p. 183). l translate ce.
Dhp-a IV 158,16-17: maltisambhavan ti bTiihmal)1"yQ miitu santike
udarasmim sambhiitam.
Pj !I 466,i• foll. (ad Sn 620): yviiya'T' ca/usu yonisu yatrha katthaci jiito
latriipi vfi visesena yo brQhnral)asSa saf!rVGl)l_Jitiiya miitari sambhiito, lt11[f
:f<]nijg_lJf_maltisambhaya'!f. Ihe cty includes-~h~ .word miitisampatti in lhe
explanation at 466,30, which supports PED's suggestion that matti- is <miiti-
= •miitri < miitari -~ miitr-. Udana-v 33.lS rea;4.s· marrsa!1ibhaV'1QJ._With
mattisambhava, ri pettikasambhava in Ja Vr485,2o• and .see Brough
(p. 183). - . .
 NOtes 159

397. - Sn 621.
For paritassati, cf. parirtase in Sn 924.

398. Dhp-a IV !61,I-4-Pj ll 467,1.1-14 (ad Sn 622): nandin ti nayhana-

bhGvena pavalta1!f k.odJunrr, varattan ti bandhanabhtrvena pavattarr tpr;ha'!l.
sandGna'!1 sahanuk.J:aman ti anusayJnuUamasahila'!l dvGsa{{hidi{fhi-
sandJna'!I.
These words arc used metaphorically. The cry tells a story about two oxen
which broke these various things.
Dhp-a IV 161,5: avijjiipafighassa uillittatllya (Pj ll 467,15: vikkhiuauii)
ukkhittapaligha'71.
In Skt parigha is a beam for locking a gate. One who bas lifted the beam up
has therefore removed an '.obstacle in his path. Udana-v 33.58 reads
urlqiplapan"k.ha. ln Skt parikhd is a moat or ditch. It makes no ... cnse to speak
of someone raising this... The Skt tradition see~ to have confused parikha
and parigha, doubtless through a dialect where both words became pari(g)ha,
and has therefore wrongly backformed the word, confusing it wi1h
sa'!1li1)1Japan·kha ..one who has filled in the moat.., Le. removed an obstacle in
his way. Both words arc epithets of an arahant
For pa/ikhalpa/igha, sec LUders (1954, § 130), and for the eastern -r-/-/-
alternation, sec the note on 32.
Dhp-a IV 161,6 (= Pj II 467,11-16 ):.catunna'71 sacciina'r' buddhattii
buddha'71.

399. Dhp-a IV 164,1-3: adu!lho ti <10'7' dasahi akkosavatthiihi akkasana'r'

pii~i-iidihi pa{hanaii ca anubandhaniidihi bandhanaii ca ya akuddhamlinaso
hutvii adhiVO..Seti_
Udiina-v 33.18 reads iikrosii'71 vadhabandhii,,.S ca. For akkosa'r' and vadha-
bandhaii as possible accusative plurals sec the note on 64-65.
Dhp-a IV 164,4-6: khantiba/ena samannagatatta khantibala'!I punappuna
uppattiyQ anika.bhuten. eva khantibaliinikena samannagataJtii baliinikaT!f.
The repetition of bala is awkward, although it is also repeated in GD hp 28.
Both versions ofUdiina-v 33.18 include vrata: (B) /qlintivratabalapetam; (]'.')
lqO.ntibiila vratiinika'!J- It is possible that the original version of the verse had
vrata (perhaps in the form vada or vala) in place of the second ba/a. Rau
(p. 175) suggests reading biilanik0'7' instead of baliinikaJ!I.
In piida c khanti- is.m.c. to avoid the opening• - - •. Pj II 467,2C>-21 (ad Sn
623) reads khanli- in the lemma.
•
400: Dhp-a IV 165,21 foll.: ta~hii-ussb-Yiihhlivena anussut0'7'.
For an-ussuta, cf. an-avassuta- 39. These are forms from the root sru "'to
flow". CPD "not overflo)Vll" is perhaps misleading. Udiina-v 33.19 has
bahuirutam. Sn 624 reads anussadam. Pj II 467,24-25: ta~hli­
ussadiibhiivena anussad0'7'. For ussada, see GD II, p. 245 (ad Sn 515).
 160 The iVord ofthe Doctrine

Dhp-a IV 165,20""' Pj II 467 ,2J: vatavantan ti dhutavatena samanniigaram.

For the dhu1angas, see EV I, p. 245 (ad Th 844-56).
In pad.a c -sOrfra'!' is me. to give the - - - "(pc.rhyQ) cadence. Cf. 352.

401. - So 625. For the image of water aod a lotus leaf, cf. 336.
For sandhi -r- in iiragge-r-iva, see the note on 338.
For the image of the mustard seed and the poi.."lt of an awl, cf. 407.
For the usages of lippati with the instrumental and locative cases, see EV I,
p. 121 (ad Th 10), and cf_ 336.
For p.lda c, cf_ BAU !V.4.28: na kanr.a~ii lipya!e piipakena_

402. Dhp-a IV 168,8-9 = Pj II 467,JO (ad So 626): pannabharan Ii ohita-

khandhahhiira'!'. For pannabhiira, see Norman, l 979D, pp. t"--48.

403. Dhp-a fV 169,12-14 = Pj II 468,3-5 (ad Sn 627): l1J""!' duggatiyii ayam

sugatiyd ayal'(l nibbG.nassa maggo ayaJ[J amaggo ti evt1J?1 magge ca amagge
ca cheka!iiya maggJmaggassa kovidam.

404. Dhp-a IV 174,1-2: ubhayan ti gihihi ca aniigtlrehi ca ti ubhayehi pi

QSGl!lSaf!ha'TI aniilayacaran ti attho.
For ubhay<V{I as a_n adverbial accusative, see CPD (s.v. ubhaya).
For aniigiira. cf. 415 416. It should historically be anagiira.
Dhp-a fV 174,2-J "' Pj II 468,9-10 (ad Sn 628): anokasiirin Ii aniilaya-
ctlrinaJ!!.

405. Dhp-a fV 175;12 foll. ,. Pj II 468,11-12 (ad Sn 629): tasesu thiivaresu

"'- cii ti _tm;)UjJiisena tasesu, tal}hiibhiive:na thiratiiya thiivare.ru.

406. Dhp-a fV 180,5-8,. Pj II 468,17-19 (ad Sn 630): hallhagate dande va

satthe vfi virujjhamtlne pi paresal{1 pahQradilnalo aviratattii artada!Jcjesu
janesu nibbutaJ{J nikk.hinadal)cj..al[1.
For nibbuta'!' as the opposite of attadal}¢esu. see Norman, l 994A, pp. 25---
26. For a discussion of danda, see Caillat, 1993, pp. 208-16.

407. =Sn 63 L-For sandhi -r- in sGsapo-r-iva, see the note on 338.
For the image of the mustard seed and the point of an awl, see 401.

408. Dhp-a JV 182,19-20: akakkasan ti apharus<V{I, viiiiiiipanin ti attha-

- _viiiiiapGJJil!', saccan ti bhiitattha'!J.
For the suggestion that sacca is< Skt •satya "pleasant... rather than·satya
..true"', see the note on 224.
Dhp-a fV 182,20-22: nabhisaje ti yiiy,a giriiya annal!' kujjhiipanavasena na
laggiipeyya~ khil)Qsavo niima evariipam eva girC17!J bhiiseyya. -
 Notes 161

For abhisaje. see CPD (s.v. abhisajati), and cf. Sn 386. For kanci, see the
note on 133.

409. For "dha (Dhp-a IV 184,1 • reads so) GDhp 19 ha3 du and Udana-v
33.25 hL. This suggests that in the exemplars followed by the rcdactqrs of
those two ta.ts there was a reading ca which they interpreted in the meaning
"but"_ C(412 415 416. For the dha!ca/ru alternation, see the note on 267.
Sn 633 rads ca, but in GD II, p. 264 I read and translated 'dha in place of ca,
followingB' and seeing a parallelism with Sn 636 639 and 640, all of which
begin will yo 'dha. I now think that I was wrong to read 'dha.
For a~. see the note on 31.
Note thewl nddiyari for nCJdiyate, and cf. adiyati 246. For the palatalisation
of -a- >-i--before -y-. see the note on 20.
In pida a 1lll is m.c. for va. In pad.a c there is resolution of the seventh syllable.

410. ~IV 185,15: asa ti ta~ha. niriisayan ti nittanhaifl.

Pj II 469.t (ad Sn 634): niriisayan ti nitt~haT[1.
For the--1-p!ay on iisa and niriisaya, and for the change in my translation
ofthe1-r(GD II, p. 72), see GD II, p. 216 (ad Sn 369). Udan.a-v 33.43
reads Dm4a'?1, which makes a better word-play with iisii, and supports the_
reading llir0s0.Sa1J1 in Ch. C&P read nirdsaya1J1 in the verse, but nirdsaJ?1 in
the lcnma. without commenL For word-plays, see the note on 9-10.
Fortheamsonant group -sm- in asmif!1 in ·pada b, see the note on 1-2.

411. I:li[>-a IV 186,6-8 ~ Pj II 469,3-5 (ad Sn 635): a!aya ti ta~hii., aiiiiaya

akathaid:athi ti a(fha vatthiini yathabhiitarp jiinitvii atfhavatthukaya
vicikia:ii9<z nibbidkiccho.
Dhp-a IV 186,8-9 ~ Pj II 469,5: amatogadha'?' anuppattan ti amataT[1
nib~ ogahetva anuppatta:J?l.
I take ogudha to be a by-form of ogadha, and therefore retain the translation
"fum !mis, foundation" which I adopted for Sn 635 (GD II, p. 72).
For anlllfa ...death-free", see the note on 21.
In padacthere is resolution of the first syllable.

412. F« 'dha Udana-v 33.29 has tu and GDhp 46 du< tu. Cf. 409. Here
Sn 636 ... 'dha_ For the alternation dhalcalru see the note on 267.
Dbp-alV 187,7-$ -IJ II 469,8 (ad Sn 636): ubha ti dve pi puiiiiiini capapani
ca~ ti attho. sarigan ti rQgiidibheclmµ sarigOJ!1.
Udam-. 33.119: ubhau sangav upatyagat. This suggests that we should take
- ---- ·-·sanga.as· a-masCUline· accilsative plutaL-For·accusative plurals· in -am, see
the nOll:OD 64-65.
For ~gii, cf. 315.
 162 The Word ofrJ..e Doctrine

413. Dhp-a lV 192.8-9 ~ Pj ll 469.14 (ad Sn 637): nandibhavaparilckhinan ti

tisu bhavesu pariillil;afa1Jhaf!l.
The cty is. then, explaining the compouud-as though it were bhavanandi·
pariillil)G'TJ. There is, however, no rc2son why the compound should not
have been written in this form, if this \\'Cre· the author's intention. since it
would be metrical. Cf_ the similar compounds in 415 and 416, which are
explained as dvandva compounds. I t:-anslate all three compounds as
dvandvas.
Udana-v 33.31C (B) has nandibhavaparU.,Uµnp. but a.a conjecture. (N) is
missing at this point. GDhp 36 has nanibhavapari4ina, and GDhp 33
sarvabhavaparik..fi1Ja. The latter suggests that we should take nandibhava,
ere., as "existence of joy/desire, etc.".
For nandi- m.c. in pada c to avoid the opening" - - •,see the note on 1-2.

414. Dhp-a IV 194.7-8 = Pj 11469.16-17 (ad Sn 638): ldlesaduggaii ca

SQ!!!Siirava!{aii ca catunnarp sacCiifUU!1 apa[ivijjha:n..aJ:amohaA ca atito.
For the eastern rll alternation in pa/ipatha, see the note on 91. For eastern
features. sec the no.tc on 32.
Udiina-v 3.41 has S"1!1Siiraugham upatyagiit, and Bernhard's note (ad loc.)
suggests the reading sa1pSiiram oham accagii here. A different punctuation of
this suggc5tion would give the compound s"J?U'iro-m-oham (< ogham). with
the meaning "the flood of s"1!1Siira". The nasalisation of -a- before -m- would
result in sa'!lsiira1!J-m-oltam, which would be interpreted as sal{Jsiira1!J.
moham. For such nasalisation. see the note on 25.
DhJ;>-a IV 194,IO = Pj !1469,19: /a~hiiya abhiivena anejo.
For aneja, cf. 422 andiiiji/a 255 and see EV!!, p. 98 (ad Tb 205).
, For piiragato, see the note on 85.
'In pada a there is resolution of the fourth syllable.

415. For 'dha GDhp 20 has du <tu (there is no eQuivalent in Udina-v

33.35). Cf. 409. For the alternation dha!caltu, sec the note on 267. Sn 639
reads 'dha.
For the first Jong -J- in aniigiira, see the note on 404.
Dhp-a IV 198,15 = Pj II 469,23: fG7!J parikkhi{)akamaii c' eva parikJ:Jiil;a-
bhavaii ca.
The cty, then, is taking lciiTI'}abhava as a dvandva compound.. Radhakrishnan
takes ir as a reversed tatp~a compound, and translates Dhp 415 "in whom
all craving for existence is extinguished" (1950. p. 185), but I do not think
that this is possible. There is no reason why bhavakiima- ·should· n·ot have··
been written, if this were the author's intention, since it would be metrical. Cf.
the similar compound in 413, which is explained as a tatpuru$a compotr'.ld..
Udiina-v 33.35 (B) has kiimiisravavisa0yuktaf!1. (N) is missing. GDhp 20
reads kamabhokapari~ina. Cf. 416.
 Note.s 163

for kiimiibhava- nt.c. in pada c to avoid the opening~ -- ~ , see the noJe on
1-2.

416. There is no GDhp parallel and Udina-v has no equivalent to 'd~a, but
we may assume that 'dha here is the same as in. 409 412 415. Sn 640 reads
'dha. For the alternation dha!ca!tu, see the note on 267
For the first long-<!- in anagdra, cf. 404.
Dhp-a JV 221.17-19 11111 Pj II 469,26-27: ta'!hiiya c' eva bhavassa ca
pariW.inafld ta~hlibhavapariillirµI'!'-
This is explaining ta~hlibhava as a dvandva. Cf. 413 and 415.

417. Dhp-a IV 225,1-S: manusaka'!' yogan ti mlinusakiiyan (Pj II 469,27

foll. ad Sn 641: lhanusaka,,i Qyu) c' eva pai1ca J:iimagu'!e ca, dibbayoge pi
e.s' eva nayo; upaccagd ti yo tnanu.saka1?1 yoglil!' hitvO dibbayoga'!1 atikknn.to
ltI1!1 sabbdri catiihl pi yogehi viscvriyunaJ?1 ahL1J?1 briiJima-'.'laJ?l vadiimi ti attho.
Udina-v 33.45 has kamd7{l and kiimlin in padas ab instead of yoga'!?. This
would suppon a suggestion that yoga1!1 is an accusative plural form. For
accusative plurals in -aJTI, sec the note on 64--6 5.
For upaccagd, cf. 315.

418_ Dhp-a IV 225,18-21 ~ Pj II 469,JI foll. (ad Sn 642): ratin ti

paiicakiim.agw:iaratirri, aratin ti araEiiaviise uli.aJJ.!hitarta'!f (Pj ukka7J{hila1!f),
siribhiitan ti nibbutmp, niriipadhin ti niiupakkifesaJ!!, viran. ti ta'!1 evarUpal!f
sabbaldumdhaJoka'!' abhibhavitvii {hitlil!' viriyavantam:
Fausbell reads dhirarrz for virmrz,*but the cry's gloss viriyavanfa'!f confirms
viral!'. Focthe dh!v alternation, see the note on 193.
There is resolution of the sixth syllable in pad.a a. or alternatively we could
read carannin place of ca aratUi.
In pada b nirUpadhi'!' is m.c.

419-20. These verses recur at Sn 643-44, and are quoted at Th-a III
181,21•-Jo•. They are in the reverse order at GD hp 43-44, and also at
Udana-v 33.46,48 (with another verse in between).

419. Forvedi, cf. Skt vetti from the root vid-. There is one example of veli in
Pali (Th 497); elsewhere veti > vedi, perhaps by analogy with the aorist, or
from a belief that t)ie root vid- ought to retain -d-. See Geiger, § 140.1. Cf.
vedi 423.t
· Dhp-a IV 228,18 = Pj II 470,S-6 (ad Sn 643): pafipattiyii su//hugatattii
sugat~.

For sugata, see the note on 285.

Dhp-a IV 228,18-19 = Pj II 470,6: (ad Sn 643): catunnam sacciina'!'
buddhatiiya buddha'!f.
For thesuffu -so in sabbaso, see GD II, p. 204 (ad Sn 288).
 164 The Word o/Ihe Doctrine

For the suggestion th.at GDhp 44 vavati ca 1.-'i sarvaSo supports a view that an
earlier pi has been omined in p3da b, because there were nine sylbbles, sec
Brough (P- 183). We could have overcome thi~ problem by assurning
resolution of the first syllable.

420. In pfoa b we should, perhap•, read a compound with Udana-v 33.46,

i.e. devagandhabbanuinusii. GDhp is ambiguous.
The cty (Burlingame, Vol. Ill, pp. 334-36) tells the story, which is also found
in Th-a III 18G-84 ofVa/igisa. who was able by tapping on a skull to say in
which of the five gatis (hell, animal world, world of pettis, world of men,
world of gods) the owner of the skull had been reborn at death, but who was
unable to do this in the case of the skull of one who- was pari1'libbu1a, i.e.
khi~iisava, b.ecause he had not gone to any place of rebirth (gall}. See Berglie,
199.,5, p. 43. Cf. the note on 92-93.
Dhp-a IV 228,11-22 = Pj II 470,8-9 (ad So 644): ilsaviina0 khi~atiiya
khi~iisava'!J, kilesehi Jrakatta arahan!CU!f..
For the opening· - - •in pada a. see the note on 1-2.

421. Dhp-a IV 230,10-22 "Pj ll 470,10-1 t (ad So 645): pure ti atitesu

khandhesu, paccha ti aniigatesu khandhesu, majjhe ti paccuppannesu
khandhesu.
Note pure in place of the expected puro < Slct puras. For the eastern
alternation -el-o, see the note on 32. For pure ... pacchii. ... majjhe, cf. 348.
Cf.jassa n 'atthi pun1 paccha, majjhe tassa Jruo siyii,Ayar L4.4.3 ..
Dhp-a IV 230,22 foll." Pj II 470,11 (ad Sn 645): kiiicanan ti yassa tesu
fhiinesu tal)hagatasa:nlMtt11[2 liiicanarti n • atthi. "

'422. Dhp--a N 231,11foll.,,,;PjII470,14-18 (ad Sn 646): acchambhitattena

usabhasadisatiiya usahha.J?'I uttamaffhena pavara'{l viriyasampalliyii vira'!f
mahan!iinaf!I silakkhandhadinmr: esitaua mahesi'!' ti')l}O'!l lvf0.rr5.l)a1!1 vijitaaa
vijitGvinarri naharaJilesaMya nahiiraJ:mri.
Note the v.I. dhira'f!I for vlraJ?t. The cty's gloss shows that this cannot be
correct. For the alternation vldh. see the note on 193.
For aneja, see the nocc on 414.
For sniitaka in the brahrnanical technical sense, see M\V (s.v. sniltaka). For
the Buddhist metaphorical interpretation of this term .. washed clean of
defilements by the water otthe eight-fold path", see EV I, p. 126 (ad Th 24)
and cf. GD ll, p. 57 (ad Sn 521).
Other editions read a svarabhakti vowel in n° hGtafcmri in pad.a c, which must
be ignored m.c. For the· nh:/nan: alternation, see GD ICp. 246 (ad Stl518).
We should perhaps read niitairm/J with the v.l. in Sn 646.

423. For vedi, see the note on 419".

For pubbeniviisa, see EV II, p. 76 (ad Thi 63).
 Notes 165

Dhp-a IV 233,!l foll.' Pj II 470,20 foll. (ad Sn 647): yo pubbenivdsom

pfikn[a'!1 karva jGnGti, chabbtsaridevalokabheda'!I sagga'!1 catubbidha'!I
apfiyaii ca dibbacaklc.hunii passati, a tho jGtik..khayasa'!lkhJtaf!! arahatta'!1
polio ... tam ah<»!f hrGhma!Jalll vadi1mi.
These three items arc the three vedas of Buddhism
Udana-v 33.47 has svOrgCip6yiirri5 ca, i.e. an accusative plural. For
saggJpJyan as a possible accusative plural, see the note on 64-65. It could
also be regarded as a dvandva compound which is a neuter singular collective .

•
 INDEX OF WORDS DISCUSSED OR QUOTED IN Tl!E
NOTES .

.akata 50 arnbbisaJTJ 153 as.a.tarn 73

akatarjl 3 14 an!gha 294-95 asara 367
akataiti\u 97 anucara 294 asanl' 304
akiilci 390 anunhahana 280 asadhu 223
akubbato 5 !-52 anunhanamala 241 asahasa 257
akkocchi J-4 anutthunaJTI 156 assa ( ~ yassa ?) 179
agata 323 anupadiyano 20 assa (~ siya) 10
aggilruna 392 anuyuiijctha 27 assaddha 97
aggi 31 anuvicca 229 ahirpsa 210
aggbzti 70 anussuta 400
acchidda 3 51 anUhata 250 akamkha343
acchiddavuni 229-30 antaka 48 ajafuia 193
ajjatanam 227 anvcti 1-2 ajaruya 322
ajjhag:ii 154 apajita 105 adiyaii 246
aiiilanlu 74 apada 179-80 .adisa 202
l!I)wµthiila 31 apara 385 lipadi 272
atanditd75 api (emphatic) 63 Abhassara 200
atikbllµ 156 api (concessive) 65 iiriya 208
atidhonaciri(n) 240 api (totality) 100-1 livudha 40
atimaiiiicyya 36 5 appafu1attika 89. lisava 89
atimapeti 246 appabodhati 143 lisli°410
ativilya 32<>-21 appamaiiiiati 121
altat!t 379 appasmi 224 icchar[l 334
attagbafililiya 164 appcssukka 330 itara 85
attaja 161 abbata 264 itaritarena 331
attadl!l)<)a 406 abhahha 32 iddhi 175
attadanha 166 abhittharetba 116 (i)dha 267
attanaI!1 3 55 abhimatthati 161 indakh:ila 95
attasambhava 161 abhisajc 408 iva 227
attahdll 84 abhisambudhano 46
attli 104 amata 21 lresi 134
attanuyogi(n) 209 amatapada 21
atha "and' 55 amatogadha 411 uklhitta·paligha 398
atha "but'' 85 amatta.iiiiu 7 ujjugata I 08
addbagii 302 amha 161 uttama-porisa 97
adhimn«a 226 ayasa 240 uttinhe 168
anattli 279 arahati 9-10 wmaja 292
anattbiya 72 ariya 270 upaka<lQhati 311
anav.$.hitacitta 38 alajjitaya316 upaccagli 315
anavassutacitta 39 allipiin' 149 upanisJ 75
anindita 227 alik.a 264 upapajjam 392
168 The JVord ofthe Doctr_ine

upapajjare 307 lcimaJcima 83 cir.lya 342

ubbhata 34 JcimaguIJe 371 cc 229
ubhayarn 4zµ lci.maratisanthava 27 cod:iy· 379
uyyogamukhe 235 -Llmii 83 ccbccchati 350
usira 337 lcimabhava parikkhi l)Jl
415 chindati 34 7
ekanta 228 kasava 9
ekaseyya 305 lclsavak:u)\ha 307 jayam 201
ekasana 305 -kahiti 350 -jata ss
etad 390 kirn (+instr_) 394 -ji- 103
etta 196 Jcim "why" 146 jil3 179
etha 171 kitavii 252
edhati 193 kuilj ara 3 22 iiarta 72
eva (= iva) 149 kenaci 196 -Iliii1 7
eva (= et ha) 344 koci 143A
°'a 134 konca 155 Qahantam 71
esano 131-32 kodha 223
essati 369 takkara 19
ehisi 236 khanatita 315 la!)haohavaparikkhina 416
khattiya 294 !3tiya 309
oka "watery" 34 khanati 247 tathagata 254
oka "home" 87 khanuoala 399 tad 326
ogadha 411 khandha202 taiµya 288
ogha 25 khandhadisa 202 tadi 95
otata 162 tava 284
gati 92
kailci 133 gavesi(n) 355 thavira 260
ka\\haka 164 gavessatha 146 thera 260
k:u:i~ 164 giithiipadarp IO I
kaJ:i\ha 307 gili 371 ctajja 224
kaJ:iha 87 -gul)a 53 dandha 116
kata 74 -gii 7 dandheti 116
katakata 50 gocanup 135 damayanti 80
kadariya 223 damayam 305
kandi371 ghacca 250 dametha 159
kanuna 15 .ghaiilia J64 dara 385
kammakili\jha 15 ghara 241 dajhaqi 61
kanunavisuddhi 16 diiruja 345
kammara 239 c' 306 disa "enemy"' 42
karomi 306 ca "but' 5 -disa "like" 202
kali 202 ca·-· ca 235 dipa 25
kasanivi\lha 143B capatikhii)a 156 duppabbajja 302
kasam 143A capato 320 durannaya 93
kasiiva 9-10 cittakata 147 dosadosa 3 57
 !ndcr of Words Discussed 169

dvaya 384 nrutaka 422 piivaka 71

d·.-edhapatha 282 pi (totality) lOC>-1
pJccssati 44-4 5 p iLllrarjl 2 94-9 5
dha 267 paccattaJjl 165 pithiyati 173
dharpsi(n) 244 pacceti 125 piyappiya 211
dhamanisanthata 39 5 pajjalita 146 pihayati 94
dhamma 1-2 paiicama 295 piti 79
dhammapada 44-4 5 paiina 28 pitibhakkh.a 200
dhammapiti 79 pa\isanrharavutti 376 puthujjana 59
dhammani 82 panditamani(n) 63 puppha 47--48
dhira "jinn" 208 patita 320 puppharp 49
dhira "wiu" 23 pada 1 pubbeniviisa 423
dhiro 207 pannabhara 402 purakkhata 342-43
dhuta!lga 240 papaiica 19 5 purisajai\ila 193
dhuvarp 147 papupphaka 46 purisiidhama 78
dhona 240 pabbaja 345 purisuttama 78
dhoreyya 208 pabbatiini i 88 piijii 73
pabbiijayam 388 peccarp 392
na (proclitic) 8 pabhanguI)l! 148 petteyyata 332
narp 1-2 pamokkhanti 276 posa 104
nandibbavaparikkhi~a -paraJjl "higher than"
413 202 phassati 259
namayati 80 parakkamma 383 phunha 83
niiga 32D pariiiiiiita 92 phuseyyu 133
nikiimaseyyarp. 3 09 pariddava 195
niggayba-viidin 24 paribbaja 313 babbaja 345
niijcja 148 parimuiicati 291 - bala 399
niddara 205 pariyodapana 183 baliinika 399
ninna 98 parissaya 328 badhita 342--43
nipaka 328-29 parihiiniiya 32 bii!e 61
rubbana 283 palavati 334 biihitapiipa 388
rubbanatha 344 pajigha398 biihusacca 271
rubhlnag2mjnj 75 pasavati 201 brabrna 142
nira!lba41 pasuta 166 brahrnafuiatii 332
nirayifya 311 passati 259 briihmal)a 142
nirasaya 410 passatha 344
nirasisa 410 pabiitave 34 bhaiiji 337
nivigha ~3B paceti 135 bhaddarp VO 337
_nisamma--karj(n) 24 - p~iIJii2ll.5 bharrtaS~ 371
nisedho 390 piimado 371 bhaviitha 1438
niyanti 175 piiragata 414 bhaviiya282
nekkha 229-30 piiragavesi(n) 355 bhikkhu 142 266
nekkbamriia 181 piiragiimi(n) 85 bhijjati 148
netavc 180 piiragii 348 bhiita 142
! 10 The JVord of the Doctrine

bhti1ni 93 yattha-lcima-nipati( n) visJ.lllJ 266

bh<:cch2ti 1·13 24 visuddhi l 6
bhovJ:di 396 yamapurisa 23 5 visodt-.2::e (n;::) i6S
yamfunase6 visodh.J.ye (pres_ part.)
maghava 30-a yiiva 284 231
martku 249-c yiivad 72 vissa 266
maccudheyya 86 ye 159 viss.asa 272
maMantu 74 yo=- si qui; 76 vih:iliisi 379
=tta 290 yogakkhema 23 vitaddar.! 335
maniisukha 290 vuttanobavaye 370
mattisambhava 396 rakkhati 231-33 vutti 229
matteyyata 332 randhaywµ 248 ve = vo ..you" J.15
madhuva 69 ramai)iyaru 99 ve=Sktvai 7
Il1'lilUSSa 182 ramita 305 ve =cc 396
mantabhiir)i 363 rfullaI)eyyaka 98 vedi 419
mamiiyita 367 -riipa 262-63 veyyaggha 295
rnabagghasa 325 vera 291
rnahaddhana 123 Iajjare 316 verim42
mahapphala 312 lajjitaya 316 VO 337
mahanaga 322 lapayanti 83 voca I33
mahapurisi 352 lahum369 vyantikjhi ti 3 50
rnabiivariha 325 Jabhupanisa 75
matailga 329-30 lippati 401 . =vas:. 207
mfilararp 294-95 lokava4\)hana 167 sarpsira 414
mana 221 smphil:a 100
-miinin 63 vajanti 83 sakubbato 52
mfiluvii 162 va!)113gandha!p 49 saklra=irµ 392
mitabh.a!)i(n) 227 vana283 saggassa 178
middhi325 vanatha283 saggaya 174
mukha 235 vananta 305 sarµk:ha!adhamma 70
muccati 71 vaniidhimuna 344 samkhaturp 196
mucchati 71 vantasa 97 sai!khara 203
muncati 389 vasissiimi 286 sailg3majuttama I03
mui\ceyya 127 vassa 286 sacca -pleasant' 224
muniiti 269 vahato 1 sajju 71
muni 269 viihana 175 sailiiamaya 380
me3-4 vijita 329 sathila312
medhaga6 vijessati 44-45 sataiµ54
mokkhanti 37 vi til)I)ll 17 6 sataljlSamalJlJ 06
mona268 VJohaviiya 282 sadattha 166
moha414 vivicca-sayana 24 sadisa 202
visaiplchara 154 sadevaka 44-45
yarp. =si quern 76 visrupkhita 154 santa "al res(' 96
yaii Ce l 06-7 visattikii 180 santa "weary" 60
 !nde:x ofJYords Discussed 171

santava 3 78 suggati 18
S.:lnlavJca 3'73 sucarita J68
santo ··goo .....-· 15 i 30.l suci 393
santhata 395 suddhajive 375
santhana 27 5 subbata 95
sandh3vissam 153 sumedhasa 29
sandhicchcda 97 surakkhita J 57
sabbhi J 51 scl:ha 45
Sam.a "equanimity" 142 scti 201
~ .. impartiality" 257 scyya1r. 61
samacariya 388 scyyaso 43
samarya 142 scyyo 390
sama 106 , 0 (~ Ski snu) 172
sarniigai\chi 210 S-0 J34
-samadana 31&-!9 soc.are 225
samiiijati 8 l sota 347
samitatti 265 sonhiya 295
samiti 321 s.omanassa 341
samirati 81
samilhata 250 hativalcisa 97
samparivattakarµ 325 hatthippabhima 326
samrnati 390 hasa 146
sayarpkata 347 hi 62
sarita 34 J lllipsamano 390
savahanaJ)l J75 huta J0&-7
sahasa 256 lrurarp 20
sahita 19--20 hw"iliuraljl 334
sata 341 betu 84
sadhurupa 262--03
sadhuvihiiridhira 328
sanucara 294
sfunaiiiiatii 332
siirattaratta 34 5
sahasa 256
siiica 369
sithila 346
sinehita 34 J
silabbaf! 2 71
SU 59
sukka 87
sukha 331-33
sukh3Sal)lvasa 207
sukhi 393
sugata 285
 INDEX Of GRAMMATICAL POINTS DISCUSSED
IN THE NOTES

abbreviated compounds 54 dhlth 6

ablative in -a'!' 49 dhlv 193
ablative in ...Jto 320 dissimilation of aspirates 94
absolutive in -a'!I (r:ramul) 156 double negative 390
absolutive with -'!1 392 d!r 305
absolutives in compounds 24 d!t 145
accusative plural in -a'!l 64-65
-a. instrumental in 270 -el-ais 61
-a +-iva 66 Eastern fonns 32:
-am> -Gm 143A -iini, acc. pl. masc. in 82
-am< -iin (acc. pl.) 64-{)5 -c, oom. sg. masc. in 375
-iini, masculine plural in 82 -e, pres. part. in 281
Ur91
blp 149 ni~ 32
ve- l'0275 ·
ca~tu5 vly40
ca ... ca235 -e-1-i- 149
cadence (sloka): - - - • 7 evaliva 149
with svarabhakti 9-10 esa + 2 sg. verb 134
canon ~ cty 66 -esin, fiinE act part. in 355
cdnc 127
CC/NC 19 fossilised '3Ddhi 72
ddh 261 future act. part. in -esin 355
change of gender I 04 future in -h- 236
dj!y44-45 future passive 37
cognate accusative 108 157
cognate nominative 159 gender. diange of I 04
compounds: genitive of goal of motion 178
abbreviated 54 glk6
split 49 ghlk34
syntactic 24 glv316
dv50
h, future m236
dative: hlylv 'lfJI
ofgi%ig 151
. - ofpurpo.SeY64. -i-1-e- 149 ·
of goal of motion 174 -i+Na>-eva 149
of time 342 infinitive of purpose 34
dative plural 151 injunctive. negative 133
dhlc 261 instr. sg. m-ii 270
 174 The IVord o/th~ Doctrin~

instr. pL in -e 61 plb 149

-ir-1-ijj-l-inj- 81 phlp 159
-isi. -iti, future in 236 plconastic -anttJ 305
ivaleva 149 plmlv 183
plph 159
jldy44-45 pres. part. in -e 281
proclitic na 8
k!g6 prolcpm 157
k!gh 34 puns 9--10
kld111k 19
k!vlylt 162 quasi-voe. 337

labialisation-of vowels 105 rid 305

locative with jiinati 6 redundant syllables 141
long vowel in place of short 6th and r/191
7th 19-20
l!r91 sandJU: -o +a-> -<l- 56; -o +a->
-a- 329-30; -o + aCC- > -aCC-
masc. acc. pL in -iini 82 1o; -a + aec- > -ace- 10; -ani
metathesis of syllables 82 + aCC- > -aCC- 273; -a'!l + u-
,,,klkk 19 > -U- 4; -i + u- > -u- l 03; -o +
mlvlp 183 e- >-e-174
sandh.i consonants:
l)amul absolutives 156 -d- 51
nasalisation of vowels 54 -m-34
-nc-1-cc- 127 -r- 338
NC/CC 19 Sanskritisation of consonant groups
' negative, double 390 1-2
'}In 32 so+ 2 sg.. verb 134
nom. sg. in -e 375 split compounds 49
nouns, past participles as 104 slss 146
n-stem >a-stem 197 subjunctive 1438
nt!nd 363 syntactic compounds 24

openings: tld145
Sloka "--" I-2 thldh 6
trigubh - - - - 142 -tt-1-tih- 116 209
- - - - 354 tti > (t)ty insandhi 376
-ol-u 7 truDcated forms in -a 270
-o-1-u- 392 two morae rule, violation of70
tly 227
palatalisation of vowels 20 t!ylvlk 162
paradoxes 97
past participles.as nouns 104 -u/-o 7
 Index o/Grammatical Poinls 175

-11-l-o- 392
-u < -o~ 84
uh- < uddh- 250

vie SO
VC/VCC 148
VC/VNC 106
vldh 193
velvo 31-5
vlg 376
v!hly 201
vfmlp 183
volve315
vfy40
vfy!tlk 162

word-play 9-10

yldj44-45
ylt 227
ylv40
ylvlh 201
y!v/Uk 162