Hi Bhante @Brahmali and Bhante @sujato
I agree with Bhante Brahmali. The privative sense comes through in some doctrinal statements. Taking eg “sabbe dhammā anattā”, I gave this analysis over at DW (with edits) -
This post will explore the meaning of sabbe dhammā anattā, particularly the issue of whether its anattā is the same in meaning as the anattā found in eg -
Rūpaṃ anattā, vedanā anattā, saññā anattā, saṅkhārā anattā, viññāṇaṃ anattā
Form is nonself, feeling is nonself, perception is nonself, volitional formations are nonself, consciousness is nonself : eg SN 22.14
First order of business, the atta in the latter anattā is from the stem attan (self, the Self). Here are its declensions –
Singular Plural
Nominative attā attāno
Vocative attā attāno
Accusative attānaṃ attāno
Instrumental attanā attehi
Dative attano attānaṃ
Ablative attanā attehi
Genitive attano attānaṃ
Locative attani attanesu (hypothetical form)
Warder, p.184
Quite clearly, the anattā in rūpaṃ anattā is a nominative singular, and the phrase rūpaṃ anattā would therefore be a predicative proposition meaning “form is not the Self”. It would not be a nexus between a substantive and an adjective, given that there is no numerical agreement between the substantive formations and adjective in “saṅkhārā anattā”.
What about sabbe dhammā anattā?
It should be noted that this phrase is part of the tag-team –
Sabbe saṅkhārā aniccā; sabbe dhammā anattā
All formations are impermanent; all phenomena are nonself
(per BB SN 22.90)
Both clauses are to be read as ending with adjectives predicating their respective substantive nouns, which probably explains why BB has rendered it as such. Other translators render it as -
all phenomena are not-self (Ven T)
all things are not-self (Mrs Rhys Davids & Woodward; Buddharakkhita)
It is this use of “not-self” for this phrase which creates the confusion in English, suggesting that the anattā in sabbe dhammā anattā and saṅkhārā anattā are exactly the same in meaning.
Taking my earlier point about “sabbe saṅkhārā aniccā; sabbe dhammā anattā”, the anattā here must be an adjective. But, adjectives must agree in case and number with their substantive noun.
And here, we see the point overlooked – since dhammā is the nominative plural of dhamma, its adjective anattā also needs to be in the nominative plural. But yet, as the declension table above shows, the adjective would need to be anattāno, instead of anattā.
What has happened? A sneaky little linguistic phenomenon has taken place, as described in Geiger’s “A Pali Grammar”, §92 –
Transfer to the a-declension often takes place as a consequence of the dropping of the final nasal.
In other words, an alternative form of attan would be atta, and this is inflected in the standard way for other –a ending nouns. For a-stems, the ā ending shows up only as the ablative singular, or the nominative plural. This clearly shows that anattā in sabbe dhammā anattā is a nominative plural. As an adjective in nominative plural, it is not another substantive noun standing in apposition to dhammā. It actually means “without Self”, instead of “not-Self”.
Where else can we find evidence that anattā in sabbe dhammā anattā means “without Self”? Take Dhp 279 -
Sabbe dhammā anattā”ti,
yadā paññāya passati;
Atha nibbindati dukkhe
esa maggo visuddhiyā.
Compare this to its Sanskrit parallel -
sarvadharmā anātmānaḥ (= nominative plural)
prajñayā paśyate yadā
atha nirvidyate duḥkhād
eṣa mārgo viśuddhaye
The Sanskrit does not have the nominative singular to read “not-Self”.
This also appears to be how the Chinese translated sarvadharmā anātmānaḥ. We see this in the parallel to SN 22.90 –
一切法無我
All dharmas are without self
It does not say 一切法非我 (all dharmas are not-self), which is untraceable on SC’s database.
It appears that depending on context, anattā needs to be translated either as “not-Self” (most of the time) and occassionally as “without Self”.
I think Harvey may have suggested that the 1st option read as “not the Self”. Might Bhantes discern any distinction between “not-Self” versus “not the Self”?