Stored-up Food

A discussion of Pācittaya 38

(Revised October 5, 2017)

Khematto Bhikkhu


Pācittiya 38 concerns the issue of how long a bhikkhu may eat staple and non-staple food after it has been formally received. There are three controversial points around this rule: 1) the interpretation of the word sannidhikārakaṁ, 2) the allowance given in the Commentary to re-receive, on a subsequent day, food that has been abandoned to a novice and eat it, and 3) the practicality of following the different interpretations of the rule. The following is a translation of that rule from the Thai Recension of the Canon with notes and passages from other rules and from the Commentary, showing that according to this rule, once staple or non-staple food has been formally received by any monk—regardless of his intention, whether or not the food has been cooked, even if he doesn’t know that it is food—no monk may eat it on any following day, regardless of what happens to the food in the meantime. Furthermore:

1) The controversy caused by assuming that sannidhikārakaṁ can be interpreted only as a ṇamul absolutive (explained below) turns out, on closer inspection, to be another case of a mountain without a molehill, as there is abundant evidence in the Canon that the word can also function as an adjective. Furthermore, the authors of both the Commentary and Sub-commentary were aware of that possibility.

2) The allowance given in the Commentary—to re-receive, on a subsequent day, food that has been abandoned to a novice and eat it—is not in agreement with the Canon.

3) The simple and strict interpretation of Pc 38 given here simplifies a monk’s relationship to food. As for the lay community, this interpretation also clarifies their relationship to food that has already been offered to monks.

Principles: Interpreting the Vinaya

A few points to keep in mind when considering this discussion are:

1) In formulating the rules in the Vinaya, the pattern is that the Buddha would first make a statement of the rule. Then, if any reason arose to make a major change in the rule, he would make a new statement. Finally, as the Buddha and the monks worked out the details of how to apply the rule, that information would be recorded in the Vibhaṅga. This consisted mainly of arriving at precise definitions of the words involved and defining the exceptions to the rule. In defining a term for the purpose of a rule, what is needed is not the most common idiomatic meaning, but a clear boundary between what does and doesn’t come under the term in the context of that particular rule. Thus, in the Vinaya, as in Western law, it is common for words to have specialized meanings, usually either a narrower or broader version of the common meaning.

In fact, the Buddha never seems to have intended that the rules stand on their own. He approved of the way the monks worked out their interpretations and the Vibhaṅga suggests that he participated in this process himself (See discussion of Cv.X.4 in BMC). In some cases, even though there is no recorded incident that prompted the Buddha to reformulate a rule, the final working-out of the rule in the Vibhaṅga differs significantly from the way the Buddha made the rule statement. In these cases the Buddha left the rule statement as it was rather than revise it to reflect the changes (assuming that he was still living when the changes were made). Also, as native speakers of the language, those monks would have had a sense of the range of meanings of the words involved, and of the range of usage that would count as grammatically correct, so they would have been in a better position to interpret the Buddha’s words than we, studying Pāli as a dead language with a very small extant literature. This means that the Vibhaṅga is the final guide to interpreting the rules.

2) The final part of the Vibhaṅga, the non-offense clauses, serves three purposes: A) to make exceptions to the rule, in which case the non-offense clauses override the word-analysis, B) to clarify areas where the explanation in the rule statement and word-analysis are unclear, and C) to clear up any doubts that an overly scrupulous monk might have about actions that clearly are not offenses in the rule statement and word-analysis, in which case the non-offense clauses simply confirm the word-analysis.

For examples of type A, a monk who is insane or the first offender inciting the formulation of that rule is exempted. All of the factors in the word-analysis are fulfilled, but the non-offense clauses, exempting an insane monk or the first offender, override that analysis.

An example of type B occurs in Pc 10, in which getting another to dig in the earth—an offense under the rule—is defined as aññaṁ āṇāpeti: “He commands another.” The word-analysis doesn’t give any more explanation of what “commanding” entails, such as whether the command has to be phrased using the imperative mood, or mention digging, to qualify as an offense. However, the non-offense clauses specify that saying, “Know this,” “Bring this,” “Give this,” “I need this,” or “Make this allowable,” are not offenses—meaning that they don’t count as commands. So in this case the non-offense clauses serve to clarify the meaning of the word āṇāpeti: “he commands.”

For an example of type C, the rule statement of Pr 3 starts, “Should any bhikkhu intentionally deprive a human being of life, …” and the non-offense clauses begin with, “Non-offenses: unintentionally, …” According to the rule statement, unintentional killing—defined in the word-analysis—would not be an offense, but the non-offense clauses repeat that fact, simply confirming the rule statement and word-analysis.

In non-offense clauses of type C, the presence or absence of the non-offense clause doesn’t make a difference in interpreting the rule, as it merely reinforces and restates what is in the rule statement and word-analysis. The compilers could add these in places where they felt monks might be overly anxious.

However, in non-offenses of type A and B, the presence or absence of the non-offense clause does make a difference. Consequently, for any action that either 1) clearly would count as an offense or 2) falls in a gray area not defined as an offense or non-offense in the word-analysis: If the action isn’t specified as a non-offense in the non-offense clauses or Vinīta vatthu (Precedents), then a bhikkhu wanting to be scrupulous should go on the assumption that the action would count as an offense. And we should assume that the compilers, being scrupulous themselves, intended it to count as an offense.

3) The Canon began as an oral tradition, in which repetitive passages are easier to memorize and variants stand out. Also, as every word had to be memorized, efficiency was a high priority: There wasn’t room for unnecessary explanation, and the style was necessarily terse. Because of that, when a passage doesn’t follow the pattern found in other passages, it should be assumed that the difference is intentional and meaningful. Anyone memorizing the text would definitely notice the difference, and one who understood how to interpret such differences would know what it meant. What this means in this case is that when something is listed as a non-offense under one rule, but not under a similar rule, it should be understood as an intentional difference between the rules, meaning that there is no exception under the second rule.

4) For one interpreting the rules for the purpose of living in line with them, one should make the following assumptions: A) The compilers of the Vibhaṅga were intelligent enough to be consistent within the discussion of each rule. Any explanation based on the premise that they were not consistent should give way to an explanation assuming that they were. B) The compilers were well enough acquainted with the contingencies surrounding each rule to know which factors were and were not crucial in determining what is and is not an offense. Any explanation that adds or subtracts factors from those mentioned in the Vibhaṅga should give way to one that follows the Vibhaṅga’s analysis.

5) Any attempt to use the Great Standards [See Mv.VI.40.1] in taking the explanations for one rule and applying them to override the explanations for another rule should be rejected, inasmuch as those Standards are meant solely for issues where nothing has already been explicitly forbidden or allowed. The Vibhaṅga for one rule cannot be used to rewrite the Vibhaṅga for another; otherwise there would be no end to the rewriting of the rules.

6) The origin stories to a rule cannot always be assumed to be a guide for determining the range of actions that would fall under the rule: That is not their purpose. Their purposes include providing a human context for the rule, as monks are human beings, not robots, and need to understand the purpose for living in line with the rules. In some cases, they convey Dhamma lessons that are only marginally related to the rule, or give the monks who memorized the Vinaya grounding in other areas of Dhamma (as in the allowance for a monastery, which includes the account of the Buddha’s awakening), or simply make their work of memorization more enjoyable and keep the Vinaya monks from becoming grim and humorless (Pc 51). In some cases the events in the origin story wouldn’t even count as an offense under the rule (NP 4, Pr 4). If the details of the events in the origin stories were taken as defining what constitutes an offense under the rule they lead up to, then one would break Pc 57 only if one were bathing in a king’s bathing spot, and if one carried salt around in a bamboo tube, rather than in a horn, then one could add it to one’s food as one liked. [See BMCI: Food Chapter]

Translation: Pācittiya 38

Note: Passages from other rules or from the Commentaries are in blue. Passages from Sub-commentaries are in green. Numbers in square brackets are from the Thai edition of the Canon. Numbers in braces are from the Myanmar edition. In the PTS edition, the passage starts from Vin IV 86.

[512] {252} tena samayena buddho bhagavā sāvatthiyaṁ viharati jetavaneanāthapiṇḍikassa ārāme.

At that time the Buddha, the Blessed One, was staying near Sāvatthī, in Jeta’s Grove, Anāthapiṇḍika’s Monastery.

tena kho pana samayena āyasmato ānandassa upajjhāyo āyasmāveḷaṭṭhasīso araññe viharati.

And at that time Ven. Ānanda’s preceptor, Ven. Veḷaṭṭhasīsa was staying in the wilderness.

so piṇḍāya caritvā bahuṁ piṇḍapātaṁ labhitvā sukkhaṁ kūraṁ ārāmaṁ haritvā sukkhāpetvā nikkhipati

Having gone for alms, having gotten much alms-food, taking dried boiled rice to the monastery—having dried it—he kept1 it.

yadā āhārena attho hoti tadā udakena temetvā bhuñjati

When he needed food, having soaked it in water, he would consume it.

cirena gāmaṁ piṇḍāya pavisati.

He entered the village for alms (once) in a long time.

bhikkhū āyasmantaṁ veḷaṭṭhasīsaṁ etadavocuṁ kissa tvaṁ āvuso cirena gāmaṁ piṇḍāya pavisasīti.

The monks said to Ven. Veḷaṭṭhasīsa, “Why do you enter the village for alms (once) in a long time?”

athakho āyasmā veḷaṭṭhasīso bhikkhūnaṁ etamatthaṁ ārocesi.

So Ven. Veḷaṭṭhasīsa reported the matter to the monks.

kiṁ pana tvaṁ āvuso sannidhikārakaṁ bhojanaṁ bhuñjasīti.

“But friend, are you consuming stored-up food?”


“Yes, friends.”

ye te bhikkhū appicchā .pe. te ujjhāyanti khīyanti vipācenti

Those monks who were modest … criticized and complained and spread it about:

kathaṁ hi nāma āyasmā veḷaṭṭhasīso sannidhikārakaṁ bhojanaṁ bhuñjissatīti .pe.

“How can Ven. Veḷaṭṭhasīsa consume stored-up food?” …

saccaṁ kira tvaṁ veḷaṭṭhasīsa sannidhikārakaṁ bhojanaṁ bhuñjasīti.

“Is it true, as they say, Veḷaṭṭhasīsa, that you consume stored-up food?”

saccaṁ bhagavāti.

“It’s true, O Blessed One.”

vigarahi buddho bhagavā kathaṁ hi nāma tvaṁ veḷaṭṭhasīsa sannidhikārakaṁ bhojanaṁ bhuñjissasi netaṁ veḷaṭṭhasīsa appasannānaṁ vā pasādāya pasannānaṁ vā bhiyyobhāvāya .pe.

The Buddha, the Blessed One, rebuked him: “Veḷaṭṭhasīsa, how can you consume stored-up food? This neither inspires faith in the faithless nor increases the faithful …

evañca pana bhikkhave imaṁ sikkhāpadaṁ uddiseyyātha

“Monks, you should recite this training rule like this:

{253} yo pana bhikkhu sannidhikārakaṁ khādanīyaṁ vā bhojanīyaṁ vā khādeyya vā bhuñjeyya vā pācittiyanti.

“Should any monk chew or consume stored-up staple or non-staple food, it is to be confessed.”

This is the rule statement. The word sannidhikārakaṁ can be formally either an adjective, ‘stored-up’ or an absolutive, ‘while storing / having stored’. However, only by treating sannidhikārakaṁ as an adjective does it fit with the Vibhaṅga’s explanation, so that is how it is translated here2.

Interpreting sannidhikārakaṁ as an absolutive, on the other hand, would yield translations such as these: “Should any monk, having stored it up / while storing it up / with the making of a store / in the manner of making a store, chew or consume staple or non-staple food, it is to be confessed.”

What follows is the word-analysis—the precise working-out of the rule that the Buddha and the monks arrived at through putting the rule into practice.

[513] {254} yo panāti yo yādiso .pe. bhikkhūti .pe. ayaṁ imasmiṁ atthe adhippeto bhikkhūti.

“Should any” Any (monk) of any kind … monk … “Monk” is meant in this case.

sannidhikārakaṁ nāma ajja paṭiggahitaṁ aparajja khāditaṁ hoti.

‘Stored-up’ means received today and eaten (chewed) on a later day.

One of the controversies around this rule centers on the definition of this word, so it is worth looking into in detail. Here are the English equivalents of the individual words:

sannidhikārakaṁ: stored-up (adj.); having stored up; with the making of a store (absol.).

nāma: name; what is named (used in definitions in the same way as “means” in English).

ajja: today.

[paṭiggaṇhāti: he receives (used as the technical term for a monk to formally receive something).]

[paṭiggahita: past participle of paṭiggaṇhāti: received.]

paṭiggahitaṁ: neuter form of paṭiggahita, matching khādanīyaṁ and bhojanīyaṁ in the previous sentence.

aparajju: the next day; a following day.

[khādati: he chews; eats.]

[khādita: past participle of khādati: chewed; eaten.]

khāditaṁ: neuter form of khādita.

hoti: it is.

Discussion: sannidhikāraka

The absolutive3 (also called “gerund”) is a type of indeclinable, called such because it does not decline, i.e., change its inflectional ending depending on case, gender, or number. (To be a declinable, the stem form would change its ending: for example, a final ‘-a’ in a stem would become ‘-o’, ‘-aṁ’, ‘-ā’, ‘-āya’ or any number of other endings, depending on the word’s role in a sentence.) Its function is similar to that of an adverb, and it is most commonly formed with endings in ‘-tvā’ or ‘-āya’. Some common absolutives are vatvā: “having said,” and ādāya: “taking.” There is also a rare form ending in ‘-aṁ’ or ‘-akaṁ’, called the ‘ṇamul absolutive4’, which is typically (maybe always) formed from an action noun (a noun formed from a verb), in the accusative case. Examples include: lihati: “licks,” > patta-nillehaka: “one who licks the bowl,” > patta-nillehakaṁ “while licking the bowl,” or karoti: “does/makes,” > kāraka: “a doer” > kārakaṁ: “in the manner of a doer” or “while doing.” That means that, in principle, every ṇamul absolutive will be identical with the singular accusative form of the action noun upon which it is based, although that action noun may or may not occur in the texts. (In the case of kāraka, it does occur, as we will see below.)

If the sentence also contains a plural noun in the accusative case, acting as the object of the verb, then because of the difference in case-endings, there is no ambiguity:

From MN 129: ... tiṇāni dantullehakaṁ khādanti: “They chew on grass (tiṇāni, plural noun), grinding it with their teeth (dantullehakaṁ, absolutive).”

dantullehakaṁ must be an absolutive in this sentence because if it were an adjective, it would change its ending to ‘-āni’ in order to match the neuter plural tiṇāni. If it were meant to be an action noun modifying the subject, te, (“The tooth-grinders chew on grass”) it would have been dantullehakā.

However, if the noun in the accusative is singular, or there is no noun in the accusative, then the sentence can be read in two ways, depending on whether the action noun is interpreted as a ṇamul absolutive or simply as a noun/adjective (in Pāli, nouns and adjectives are considered one grammatical category). For example:

santi kho tassa bhagavato sāvakā ... sattha-hārakaṁ pariyesanti (SN 35:88, variant in Pr 3) could be interpreted either as “There are disciples of the Blessed one who search for a knife carrier [i.e., an assassin] (action noun/adjective),” or “There are disciples of the Blessed one who, in the manner of knife carriers (absolutive), search.”

na jivhā-nicchārakaṁ bhuñjissāmi (Sk 49) could be read either as “I will not eat (while) sticking out the tongue (absolutive),” or “I will not eat one who is sticking out his/her tongue (action noun/adjective).”

In the first sentence, the first interpretation is more convincing: The monks search for an assassin. In the second sentence, the second interpretation makes no sense at all, thus the meaning of the passage is unambiguously the first reading, and the grammatical interpretation follows that.

Even looking just at compounds built from -kāraka, we find that it can form compounds of both grammatical types:

bhaṇḍana-kārakaṁ kalaha-kārakaṁ vivāda-kārakaṁ bhassa-kārakaṁ saṅghe adhikaraṇa-kārakaṁ: (Pc 17) “makers of quarrels, strife, disputes, dissension, and issues in the Saṅgha (accusative noun/adjective).”

capucapu-kārakaṁ: (Sk 50) “making a chomping noise (absolutive).”

But the present case is more ambiguous than either of the above two examples. “Should any monk, while storing it up, chew or consume food ...” isn’t implausible.

So it’s a legitimate question: Is sannidhikārakaṁ in fact a ṇamul absolutive? As noted, absolutives, being indeclinables, differ from noun/adjectives in that noun/adjectives do decline. Another difference is that when absolutives act as non-final members of compounds (extremely rarely), they still preserve their distinctive absolutive ending5. (I know of no cases of ṇamul absolutives being used this way.) Using these two criteria, in some cases it is possible to distinguish between absolutives and noun/adjectives.

In the phrase sannidhikārakaṁ paribhuñjitabbāni, in NP 23, sannidhikārakaṁ must be an absolutive because sannidhikāraka still takes the ending ‘-aṁ’ rather than ‘-āni’, which it would if it were an adjective, in order to match the neuter plural paribhuñjitabbāni. On the other hand, in the Vibhaṅga to this rule, in the phrase sannidhikārake sannidhikāraka-saññī, the word must be an adjective, as sannidhikāraka takes the ending ‘-e’ to indicate the locative case in the first occurrence of the word, and in the second occurrence it is reduced to its stem form as the first element of a compound.

At first glance, this might appear to be a grammatical contradiction, but the explanation is simple: sannidhikāraka can function either as an adjective or as an absolutive.6

Here is a complete list of the 32 occurrences (9 discounting repetitions), aside from the definition in question, of sannidhikāraka in the Thai recension of the Canon:

Cases in which sannidhikāraka is inflected (or reduced to its stem form) like a noun/adjective (14 total, 4 unique):

(a)sannidhikārake (6 occurrences: all in Pc 38)

(a)sannidhikāraka-saññī (4 occurrences: all in Pc 38)

sannidhikāraka-bhojane (1 occurrence: Cv.12)

sannidhikāraka-paribhog... (3 occurrences: AN 5:80 Thai, DN 1 Thai, 7)

Cases in which sannidhikāraka appears inflected in the form of the ṇamul absolutive sannidhikārakaṁ: (11 total, 2 unique)

sannidhikārakaṁ paribhuñjitabbāni (4 occurrences: NP 23, Parivāra)

sannidhikārakaṁ kāme paribhuñjituṁ (7 occurrences: DN 27, DN 29, DN 33, MN 76, AN 9:7, and AN 9:8)

Cases in which its inflection could indicate either8: (9 total, 3 unique)

sannidhikārakaṁ bhojanaṁ bhuñj... (5 occurrences, Pc 38, Parivāra)

sannidhikārakaṁ khādanīyaṁ vā bhojanīyaṁ (3 occurrences: Pc 38, Parivāra)

sannidhikārakaṁ sāliṁ (1 occurrence: DN 27)

An objection to interpreting sannidhikāraka as an adjective is the existence of the word sannidhikata, derived from sannidhi: “a store” + kata: “made” > “made into a store” or “stored-up.” The objection is that if the compilers had intended to refer to “stored-up” food, they would have used sannidhikata. First of all, it is presumptuous for a non-native speaker to assume to know the range of words from which a native speaker might choose in communicating a certain meaning. Second, Pāli has examples of past participle constructions that are nearly synonymous with the noun/adjective construction of the same verb. From parājeti: “to defeat” are derived both parājita: “defeated” (past participle) and pārājika: “of or pertaining to defeat” (noun/adjective), used to describe both a class of offense and a monk who has committed such an offense. The phrase in the relevant rule statements is: pārājiko (not parājito) hoti asaṁvāso: “he is defeated and no longer in affiliation.”

Thus there is clear evidence that sannidhikāraka can function grammatically as either a noun/adjective or an absolutive. That means that in this case, the context has to determine how the word should be interpreted grammatically.

Some other important points in the Vibhaṅga’s definition of sannidhikārakaṁ given above are:

1) The word is defined in terms of adjectives describing the food, rather than adverbs modifying the act of eating. Specifically, they are past participles (like the English word, “stored”). The definition doesn’t specify who receives the food, but the verb paṭiggaṇhāti is used in a technical sense for monks formally receiving things, so it must mean a monk.

2) There is no mention at all of actually storing the food—who is storing it or where, or whether or not the monk retains possession of it.

3) “Stored-up” here is defined as the combination of two factors: The food has been a) received today and b) eaten on a following day.

4) If it were argued that the Vibhaṅga here is not defining “stored-up,” but instead adding further qualifications on food that the monk is storing, that would mean that the part of this rule most in need of clear definition—what qualifies as “storing-up”—would be left undefined: Would it mean that if another monk stored it for him, the first monk could eat it on a later day without offense, as he was not the one who stored it? Similarly, if a novice or unordained person stored it for him, even though the monk expected it back, could he also receive it again and avoid an offense in eating it? Would he have to keep it in his hut or near him for it to count? If he put it in a communal place, like a monastery storeroom, intending it to be for all of the monks, would he count as storing it? If he were traveling, would he have to take it with him to count as storing it?

The compilers of the Vibhaṅga would surely have thought of these issues and passed judgments on them if they had intended sannidhikārakaṁ to imply that the monk has to be storing the food himself. As mentioned in point 4 in the Introduction, we should assume that the compilers of the Vibhaṅga considered carefully which aspects of the term needed to be defined in order to arrive at a clear, well-defined rule, and which were obvious enough to leave unstated. Treating this sentence as a definition avoids all of that confusion and prevents monks from getting around the rule by passing food back and forth among themselves. Thus, to stay on the side of scrupulousness, and out of respect for the compilers of the Vibhaṅga, we should assume that this is a comprehensive definition of what sannidhikārakaṁ means in the context of this rule.

All of these considerations support the interpretation that once a monk has received food and a night has passed, it is called “stored-up.” As we will see below, this is further confirmed by the absence9 in the non-offense clauses of the parallel passage in NP 23, which does allow a monk to re-receive tonics that have been relinquished without any sense of attachment:

From NP 23: anāpatti antosattāhaṃ adhiṭṭheti, vissajjeti, nassati, vinassati, ḍayhati, acchinditvā gaṇhanti, vissāsaṃ gaṇhanti, anupasampannassa cattena vantena muttena anapekkho datvā paṭilabhitvā paribhuñjati, ummattakassa, ādikammikassāti.

Non-offenses: Within seven days he determines it for use, throws it away, it gets lost, destroyed, burnt, snatched away, or taken on trust; or—having given it away to an unordained person, as if relinquishing it, vomiting it, releasing it—he receives it in return and makes use of it; for one who is insane; for the first offender.

5) The following passage from the Vibhaṅga to Pc 40 defines what it means to receive the food:

From Pc 40: adinnaṁ nāma appaṭiggahitakaṁ vuccati.

That which is not given is called “not received.”

dinnaṁ nāma kāyena vā kāyapaṭibaddhena vā nissaggiyena vā dente hatthapāse ṭhito kāyena vā kāyapaṭibaddhena vā paṭiggaṇhāti etaṁ dinnaṁ nāma.

“Given” means: As he/she is giving with the body or something connected to the body or by dropping, standing within a hatthapāsa, he receives it with the body or something connected to the body. That is called “given.”

In other words, “received” and “given,” in the sense defined here, are equivalent. In this definition there is no mention of the monk’s intention in receiving whatever is given, or even of whether he knows he is receiving something edible. The mere physical act of accepting an edible makes it count as “received.”10

The Commentary’s Allowance

The Commentary to this passage, however, contains this allowance:

From the Commentary to this rule: yaṁ bhikkhū nirapekkhā sāmaṇerānaṁ pariccajanti, tañce sāmaṇerā nidahitvā denti, sabbaṁ vaṭṭati. sayaṁ paṭiggahetvā apariccattameva hi dutiyadivase na vaṭṭati.

Whatever monks—without concern/desire—abandons to novices: If the novices, having kept it, give it, that is all allowable. If he has received it himself and has not abandoned it, it is not allowable on the second day.

No explanation is given in the Commentary of how this conclusion was reached.

Theories Explaining the Commentary’s Allowance

Several theories have been offered to account for the Commentary’s allowance here:

1) The authors of the Commentary interpreted sannidhikārakaṁ as a ṇamul absolutive, implying that the monk has to store the food himself.

2) Abandoning something is equivalent to not having received it in the first place. (This is the Sub-commentary’s explanation.)

3) The Commentary is importing the parallel non-offense clause from NP 23 to this rule.

4) The Commentary is making an allowance for what had become common practice at the time.

5) The authors of the Commentary added this allowance so that one could be absolutely certain of avoiding an offense under this rule.

If we look at these theories in detail, we see that none of them offers convincing support for the Commentary’s allowance:

1) The authors of the Commentary interpreted sannidhikārakaṁ as an absolutive happening continuously, i.e., the monk has to be storing the food himself, which presumably means from the time it is offered until the time he eats it. Thus, if he gives it away, he can re-receive it on a later day and eat it without offense. This theory then assumes that this chain of logic would be clear enough that the compilers of the Vibhaṅga felt no need to mention, in either the word-analysis or in the non-offense clauses, the condition that the monk has to store the food himself the whole time.

However, as we have already seen, sannidhikārakaṁ isn’t always an absolutive. That means that the “chain of logic” would not have been unambiguously clear to the compilers of the Vibhaṅga. The word’s grammatical ambiguity would have forced them to state clearly this chain of logic if that were what they intended.

Also, three commentarial passages show that the authors of the Commentary11 knew that sannidhikāraka could function as a noun/adjective, as they use it that way themselves:

A) In the following passage from the Commentary to DN 1, the word sannidhikārakaṁ makes sense only as an adjective, as the subject of the sentence is annaṁ: “food.” Interpreting it as a ṇamul absolutive is inconsitent with both the subject—an inanimate object—and verb “to be,” which makes no sense with the adverb, “while storing.” The passage is worth including in full because it gives the impression that giving food back “without desire” had already become an empty ritual in the time of the Commentary.

From the Commentary to DN 1: sannidhikāraparibhoganti sannidhikatassa paribhogaṁ. tattha duvidhā kathā, vinayavasena ca sallekhavasena ca. vinayavasena tāva yaṁ kiñci annaṁ ajja paṭiggahitaṁ aparajju sannidhikārakaṁ hoti, tassa paribhoge pācittiyaṁ. attanā laddhaṁ pana sāmaṇerānaṁ datvā, tehi laddhaṁ ṭhapāpetvā dutiyadivase bhuñjituṁ vaṭṭati, sallekho pana na hoti.

“Store-making consumption”: the consumption of what is stored-up. In that case, the discussion is two-fold: for the sake of of Vinaya, and for the sake of wearing down (the defilements). For the sake of Vinaya: Any food received today is, on a later day, stored-up. In its consumption there is a pācittiya offense. But, having given what was gained by oneself to novices, having had them set the gains aside, on the second day it is allowable to eat them. But there is no wearing-down.

B) The Commentary to Mv.VI.40.3, describing right-time (food) and night-watch-time (juice) as coming under this rule in the case of mixing edibles of different classes, contains the sentence idameva dvayaṁ antovutthakañceva sannidhikārakañca hoti: “Just this pair is (potentially) stored indoors and stored-up.” Here, sannidhikārakaṁ makes sense only as an adjective for the same reasons as those discussed above, in example (A).

C) Most importantly, the Commentary to Pc 38 itself makes it clear that it interprets sannidhikārakaṁ as an adjective modifying the food—derived from kāra: “making”—rather than an absolutive modifying the monk—derived from the action noun kāraka: “maker.”

From the Commentary to this rule: kāro karaṇaṁ kiriyāti atthato ekaṁ, sannidhikāro assāti sannidhikāraṁ; sannidhikārameva sannidhikārakaṁ

“Making/maker, making, action”: These are one in meaning. “There would be the making of a store”: a storing-up. Like13 a storing-up: stored-up.

First, the Commentary equates kāra, which can mean either “a maker” or “a making,” with karaṇaṁ and kiriyā, which have the meaning “a making,” but not “a maker.”14 In this case, then, kāra means “a making.”

Next, the Commentary analyzes the etymology of the word:

sannidhi: “a store” + kāra: “making” > sannidhikāra “the making of a store” + -ka (suffix that makes noun/adjectives15) > sannidhikāraka “stored-up” accusative neuter singular: sannidhikārakaṁ

If sannidhikārakaṁ were interpreted as an absolutive, its etymology would be this:

kāra: “making/maker” + -ka (suffix that makes noun/adjectives) > kāraka “a maker” + sannidhi: “a store” > sannidhikāraka “one who stores up” ṇamul absolutive: sannidhikārakaṁ “while storing up.”

Every other explanation of sannidhikārakaṁ in Commentaries on other passages in the Canon (Vin. NP 23, DN 27, DN 29, MN 76, AN 9.7), glosses it as sannidhiṁ katvā. In fact, in every instance I have found in which the Commentary explains a ṇamul absolutive, it is glossed as an absolutive in -tvā, aside from one case where it is glossed as a present participle.12 This further supports the conclusion that the authors of the Commentary intend the present definition to mean that sannidhikārakaṁ should be interpreted here as an adjective.

In other words, the Commentary is saying here that sannidhikārakaṁ modifies the food, not the monk.

Because the Commentary specifically interprets sannidhikārakaṁ as an adjective here, theory (1) is baseless.

Theory (2) is stated in the Sub-commentary:

From the Sāratthadīpanī-ṭīkā Sub-commentary to this rule: Yadaggena hi paṭiggahaṇaṁ vijahati, tadaggena sannidhimpi na karoti vijahitapaṭiggahaṇassa appaṭiggahitasadisattā. Paṭiggahetvā nidahiteyeva ca sannidhipaccayā āpatti vuttā. Paṭiggahetvā ekarattaṁ vītināmitassetaṁ adhivacanan’ti hi vuttaṁ.

From the time he abandons the receipt (acceptance), he is also not storing up, because of the equivalence of that for which receipt has been abandoned with that which has not been received. Only concerning that which has been received and kept is there said to be an offense with storing-up as a condition. Thus it is said, “This is a term for what has been received and passed beyond one night.”

A modern defense of the Sub-commentary’s argument is that it is justified by the factor of intention in the derived offense for receiving food with the intention of eating it as stored-up food. It then paraphrases the Sub-commentary, saying that because giving away is the reverse of receiving, when a monk gives away food and abandons desire for it, that nullifies the receipt of the food. However, the factor of intention applies only to that derived offense, not to the full offense (as we will see below).

A possible explanation of the Sub-commentary’s reasoning here is that it is drawing a principle from the allowance in the non-offense clauses under NP 23 quoted above: If, within seven days, a monk gives—as if relinquishing it, vomiting it, releasing it—a tonic to an unordained person, he can re-receive it and consume it without offense. Thus, that tonic is equivalent to a tonic that has not yet been offered. If this is the Sub-commentary’s reasoning, then it may indicate that the authors of the Sāratthadīpanī-ṭīkā understood the Commentary in line with theory (3): The Commentary is importing the non-offense clause from NP 23 into Pc 38.

However, even under NP 23, a tonic that has been so given to an unordained person is not in every case equivalent to what has never been received: If a monk keeps a tonic for more than seven days, forfeits it to another monk, confesses the offense, and then gets it back from the other monk, he may use it only for external purposes. If the monk then gives it—as if relinquishing it, vomiting it, releasing it—to an unordained person who then offers it again, that tonic still may be used only externally. There is no mention in the Vibhaṅga to NP 23 of any allowance to consume the tonic in such a case. Thus it is not equivalent to a tonic that has never been received, and the Sub-commentary’s reasoning cannot be taken as a general principle even within NP 23.

Furthermore, there is nothing anywhere in the Vinaya that supports the Sub-commentary’s assertion that—as a general principle—abandoning something is equivalent to not having received it in the first place. If one adopted that principle, then one could retroactively nullify an offense for receiving money by abandoning it, and similarly with other offenses for receiving things like felt rugs, almsbowls, etc.

Thus theory (2) doesn’t offer convincing support for the allowance.

3) The Commentary is importing the parallel non-offense clause from NP 23 to this rule. As noted above, that rule contains an allowance, overriding the rule statement and word-analysis, to re-receive and consume tonics that have been abandoned. Thus, under that rule, a monk has to retain a sense of possession over the tonic—even if he isn’t actually storing it himself—for it to count as stored. That non-offense clause draws a clear line between storing and not storing: Giving the tonic to another monk does not count as abandoning—as it would under the rules dealing with robe-cloth—implying that when a monk gives a tonic to another monk, it still counts as being stored. If the authors of the Commentary were importing this non-offense clause into Pc 38, that would explain why they saw no need to address the issue in the Vibhaṅga to that rule.

In light of point (2) in the section on General Principles, we should assume that the difference between the non-offense clauses here and those in NP 23 was made intentionally by the compilers of the Vibhaṅga, in order to show that the rules function differently. If one followed Commentary’s interpretation of this rule—that if a monk gives away the food without expectation of getting it back, he can receive it on a later day and eat it without an offense—then this rule would function in the same way as NP 23, aside from the fact that under that rule, the offense is for storing it past seven days, with no mention of consuming it. We would expect to find the clause in the non-offenses in the Vibhaṅga to this rule, “having given it away to an unordained person, as if relinquishing it, vomiting it, releasing it—he receives it in return and makes use of it.” But it’s not there.

Following point (5) in the section on General Principles, if the compilers of the Commentary were invoking the Great Standards to justify importing the non-offense clause from NP 23, that would be a misapplication of those Standards. Had the compilers of the Vibhaṅga meant for the principle under NP 23 to be applied here, they could have done so themselves.

4) The Commentary is making an allowance for what had become common practice at the time16. However, it was determined by the elders of the Second Council that “The practice of what is habitual is sometimes allowable, sometimes not.” (Cv.XII.2.8) Thus common practice is no justification for the allowance.

5) The motivation for the allowance was to allay excessive scrupulousness: How can one ever know for sure that there is no trace of left-over food in one’s bowl, especially if it’s made of clay, and has a rough surface and tiny fissures? The discussion in the Commentary just prior to stating the allowance concerns just that question—perhaps implying that the allowance could be used to allay an overly scrupulous monk’s fear of eating stored-up food. However, the Commentary’s allowance opens up another avenue by which an overly scrupulous monk could suffer from uncertainty: He would never know whether the monk who abandoned the food to the novice on a previous day really abandoned all desire for it.

In any case, one can never be sure that the food one eats doesn’t contain trace amounts of unallowable foods, such as unallowable kinds of meat. Even if one asks the donors what is in the food, they might lie or be misinformed themselves about the ingredients. Even in simply eating, one can never be sure one in not committing an offense. There are many other rules in the Vinaya that one could easily break without knowing it, such as NP 20, prohibiting accepting money (as often happens, hidden in a bag of food a monk receives on almsround), Pc 6, against lying down under the same roof with a woman (even if she is in a distant room and one doesn’t know she’s there), and the many other rules where perception and intention are not factors for the offense. If one insisted on an interpretation of the Vinaya that allowed a monk to be certain he was never breaking a rule, that would involve significantly altering the rules.

Theories (4) and (5) explain possible motivations for making the allowance, but offer no evidence in support of accepting it.

None of the reasons mentioned above offers convincing grounds for accepting the Commentary’s allowance as valid. Thus we can answer the question posed above: In the context of this rule, sannidhikārakaṁ is an adjective.

Below is the continuation of the word-analysis from Pc 38:

khādanīyaṁ nāma pañca bhojanāni yāmakālikaṁ sattāhakālikaṁ yāvajīvikaṁ ṭhapetvā avasesaṁ khādanīyaṁ nāma.

‘Non-staple food’ means: Setting aside the five foods, wrong-time (juice-drinks), seven-day (tonics), and lifetime (medicines), what is left over is called non-staple food.

bhojanīyaṁ nāma pañca bhojanāni odano kummāso sattu maccho maṁsaṁ.

‘Staple food’ means the five foods: cooked grain, kummāsa, meal, fish, and meat.

khādissāmi bhuñjissāmīti paṭiggaṇhāti āpatti dukkaṭassa.

(Thinking,) “I will chew it; I will consume it,” one receives it: an offense of wrong-doing.

ajjhohāre ajjhohāre āpatti pācittiyassa.

For each mouthful (that one eats) there is a pācittiya offense.

The derived dukkaṭa that the Canon imposes here for receiving could be interpreted in two ways: 1) The monk receives the food intending to store it up and eat it on a following day, or 2) he re-receives food that has been received on a previous day, intending to eat it. (Pc 37 has exactly the same passage, not specifying whether the dukkaṭa is only for accepting the food in the wrong time, or also for accepting food in the right time, intending to eat it in the wrong time.) There is no mention of whether he knows that it has previously been stored, or what his intention was when he gave it away. If that factor made a difference, i.e., giving the food away without desire to get it back absolved the monk of an offense, then this is another place we would expect to see it mentioned. But it’s not.

The Commentary adopts the second interpretation: “It” here means food already stored up. In line with the Commentary’s above allowance, it must be interpreting this as food that one has received on a previous day and given away—as one is receiving it again—but without abandoning desire for it. However, the Vibhaṅga makes no mention of unabandoned/abandoned desire. Following the interpretation of the Vibhaṅga, “it” refers to food a monk has received on a previous day, regardless of what has happened to it in the meantime.

Also, it’s important to note that the factor of intention here, “(thinking,) ‘I will chew it; I will consume it,’” applies only to this derived offense, not to the full offense. As the definition of “receiving” in Pc 40 shows, neither intention nor knowledge are necessary factors in making food formally received, thus there is no basis for assuming that the factor of intention should apply to the full offense as well. That means that if a monk receives food, not intending to eat it, stores it overnight, and then changes his mind and eats it, the food counts as properly received under Pc 40, but he still incurs the full offense here.

[514] {255} sannidhikārake sannidhikāraka-saññī khādanīyaṁ vā bhojanīyaṁ vā khādati vā bhuñjati vā āpatti pācittiyassa.

In the case it is stored-up, one chews or consumes staple or non-staple food, perceiving it to be stored-up: a pācittiya offense.

sannidhikārake vematiko khādanīyaṁ vā bhojanīyaṁ vā khādati vā bhuñjati vā āpatti pācittiyassa.

In the case it is stored-up, one chews or consumes staple or non-staple food, doubtful (about whether it is stored-up): a pācittiya offense.

sannidhikārake asannidhikāraka-saññī khādanīyaṁ vā bhojanīyaṁ vā khādati vā bhuñjati vā āpatti pācittiyassa.

In the case it is stored-up, one chews or consumes staple or non-staple food, perceiving it to be not stored-up: a pācittiya offense.

This shows that if a monk receives food, not knowing that it had been received on a previous day, he would still incur the full offense in eating it. This also would cover the case where a monk gets mixed up about which food he accepted today and which he is storing up without intending to eat it.

yāmakālikaṁ sattāhakālikaṁ yāvajīvikaṁ āhāratthāya paṭiggaṇhāti āpatti dukkaṭassa.

One accepts wrong-time (juice-drinks), seven-day (tonics), and lifetime (medicines) for the sake of food: an offense of wrong-doing.

ajjhohāre ajjhohāre āpatti dukkaṭassa.

For each mouthful (that one eats) there is an offense of wrong-doing.

asannidhikārake sannidhikāraka-saññī āpatti dukkaṭassa.

In the case it is not stored-up, (one chews or consumes it) perceiving it to be stored-up: an offense of wrong-doing.

asannidhikārake vematiko āpatti dukkaṭassa.

In the case it is not stored-up, (one chews or consumes it) doubtful (about whether it is stored-up): an offense of wrong-doing.

asannidhikārake asannidhikāraka-saññī anāpatti.

In the case it is not stored-up, (one chews or consumes it) perceiving it to be not stored-up: no offense.


[515] {256} anāpatti

Non offenses:

yāvakālikaṁ yāvakāle nidahitvā bhuñjati

Having kept right-time (food) during the right time, one consumes it;

yāmakālikaṁ yāme nidahitvā bhuñjati

having kept wrong-time (juice-drinks) during the wrong time, one consumes them;

sattāhakālikaṁ sattāhaṁ nidahitvā bhuñjati

having kept a seven-day (tonic) for (up to) seven days, one consumes it;

In the non-offense clauses here, as in the origin story, the verb nidahati, rather than sannidhiṁ karoti, is used to describe storing the edibles. This means that, in the same way, if one ‘keeps’ (nidahati) tonics, they count as stored-up (sannidhikāraka).

yāvajīvikaṁ sati paccaye paribhuñjati

one consumes lifetime (medicines) when there is a reason;

ummattakassa ādikammikassāti.

for one who is insane; for the first offender.

As noted above, there is no parallel passage in these non-offense clauses to that in NP 23, making an exception for food that a monk has relinquished to unordained people. And it is in keeping with the style of the Vinaya that no mention is made of that passage to explain the discrepancy. The non-offense clauses—here as everywhere else in the Vibhaṅga—do not address the issue of what are not non-offenses. So it is not surprising that, although the non-offenses under NP 23 include the allowance to re-receive tonics, the question of whether a similar allowance would hold here is not discussed. As mentioned above, this implies that there is no allowance: Staple and non-staple food may not be re-received in the same way.

There is no non-offense mentioned for simply storing food. That is because the only offenses under this rule are for eating stored-up food and, as a derived offense, for receiving it with the intention to eat it as stored-up food. There is no offense for simply storing food mentioned in the rule in the first place. That means if a monk receives food, not intending to eat it, changes his mind and stores it overnight, intending to eat it on a later day, then changes his mind again and doesn’t eat it, he incurs no offense.

aṭṭhamasikkhāpadaṁ niṭṭhitaṁ.

The eighth training rule is finished.


In the case of this rule, as in others in the Vinaya, the explanation in the Vibhaṅga takes the statement of the rule, made by the Buddha, and puts it into a clearly defined, practical form. The interpretation the compilers arrived at is both strict and simple. However, some object to this interpretation on the grounds that it is impractical, and that it goes against the values of the Dhamma, specifically in two areas: contentment and unburdensomeness. The argument is that monks should not be above eating left-over food, and that allowing the lay people to re-offer food makes their job easier and prevents waste.

In response to the first objection, the rule doesn’t prohibit eating lay people’s left-overs. MN 82 tells the story of Ven. Raṭṭhapāla returning home after attaining arahantship. He goes on almsround to his parents’ house and—after his father drives him away—asks a slave woman to pour some left-over rice into his bowl rather than throw it away.

As to the second point, if someone brings more food than can be used in one day, it is not difficult to divide the dish, serve just enough for that day, and serve the rest on a later day.

In fact, if it were allowable to re-offer food to monks, it would be a cause for anxiety for the lay people who live in monasteries:

There is a story in the Petavatthu (KN 7.5: tirokuṭṭapetavatthu) of some lay people who go to help out with the daily meal-offerings at the monastery of a past Buddha for three months. Some of them are there out of strong conviction; some just come along. Some of the latter surreptitiously take home food set out to offer the monks for their children to eat, or for themselves. After death, they are born in hell for 92 eons. After they are released from hell, they get reborn as ghosts tormented by hunger for another 92 eons, until eventually King Bimbisāra makes merit and dedicates it to them, allowing them to be freed from that state.

Because offerings of food and other requisites offered to a monastery are, unless otherwise specified, assumed to be intended for the monks, lay people who live in monasteries tend to be very careful about making use of those offerings. If lay people assume that it is allowable to re-offer food to monks, then even when the monks hand back the food they don’t need, the lay people may still feel uneasy about eating it. They might wonder whether the original donors would have wanted them to set it aside and re-offer it to the monks, and they might be anxious that they are making bad kamma by not doing so. This problem would be even more acute in times of famine, when there might not be any food offered at all some days. But if the food that the monks give back cannot ever be re-offered, then it is unquestionably for the lay people alone. They will have no qualms about keeping it and eating it themselves (especially if it’s cheesecake).

If they don’t want to keep it, they can give it to a charity or feed it to animals. The Buddha taught that even animals are worthy of compassion and worthy to receive generosity. In AN 3:58, the Buddha says, “I tell you, Vaccha, even if a person throws the rinsings of a bowl or a cup into a village pool or pond, thinking, ‘May whatever animals live here feed on this,’ that would be a source of merit, to say nothing of what is given to human beings.” Giving food to animals is a form of merit, not a waste, which means that it is not difficult at all to ensure that no food goes to waste.

Another pragmatic issue is that if monks regularly give left-over food to lay people—theoretically abandoning all desire for it—but the lay people regularly re-offer it, the monks will naturally start to expect to get their leftovers back, and eventually the “abandoning of all desire” won’t be honest. The passage from the Commentary to DN 1 quoted above shows that this was already a problem in the Commentary’s time, as monks were “having novices set aside” their leftovers for them, and that in not storing it themselves, there was “no wearing-down” of the defilements, which probably means that they expected to get it re-offered.

The Commentary’s allowance would create further causes for doubt such as these: If the intention of the monk giving away the food (often the junior-most monk at the end of the line in the meal hall) is what determines whether it may be re-offered, then only he will know whether the food can be re-received. The other monks would have to be in doubt unless they ask him about the matter. The lay people won’t know for sure either. If the monk returning the food still hadn’t completely given up desire for it, would they be encouraging the monks to unwittingly commit an offense if they re-offered it? What should the monk giving back the food do if he knew that he still had desire for it when he gave it? Should he tell the lay people not to re-offer it? Or should he say nothing, but—if he saw what he knew was the same food being offered on a later day—announce to the monks that he still had desire for the food when he gave it away? In times of plenty this might seem like a minor point, but in times of famine, it would not be minor.

The simple and strict interpretation of Pc 38 stated in the first paragraph of this essay avoids all of these problems. It also simplifies and clarifies issues around offerings of food, for both the monks and for the lay community:

Once staple or non-staple food has been received hand-to-hand by any bhikkhu—regardless of his intention, whether or not the food has been cooked, even if he doesn’t know that it is food—no bhikkhu may eat it on any following day, regardless of what happens to the food in the meantime.


There is a more fundamental problem with the argument based on interpreting sannidhikārakaṁ as a ṇamul absolutive: its general approach. It focuses on one detail—that sannidhikārakaṁ can sometimes function as an absolutive—and then forces the definition in the Vibhaṅga to fit in line with that detail, ignores the evidence in the non-offense clauses, ignores the instances where sannidhikārakaṁ is clearly an adjective, ignores the fact that the authors of the Commentary and Sub-commentary knew it could be an adjective, ignores the unresolved issues such an interpretation would introduce into the rule, and ignores the practical problems it would create in the relationship between the monks and the lay community.

The kind argument that focuses on a single grammatical detail and proceeds logically from there, regardless of what conclusions it leads to, is based on a narrow and naive understanding of human language. Human languages are not perfect logical systems—even modern languages that have been standardized under the influence of political power and writing, to say nothing of unstandardized languages in mostly illiterate societies. The Pāli of the Canon was a language in the process of being standardized, and shows many signs that the process was not complete.

Grammatical analysis of the language can often yield important insights into the meaning of various points of Dhamma and Vinaya, but if that analysis is blind to other considerations in the texts, then it can misrepresent the Dhamma and Vinaya instead. For example, another product of such narrowly logical grammatical analysis is the assertion that because the passage expounding the four noble truths doesn’t follow the standard rules of canonical Pāli grammar, it must be a later addition to the Canon. (For a detailed and fascinating description of language change and standardization, see John McWhorter, The Power of Babel (New York: Perennial, 2003), especially chapter 6.)

If such grammatical analysis were being used to argue for a stricter interpretation of Pācittiya 38, then it might be worth giving it the benefit of the doubt, as in many cases the Buddha assigned a dukkaṭa for acting in doubt that one is breaking a rule. But here it is being used to support a looser interpretation. This attitude of focusing on one detail and blocking out the larger context is one way the mind can justify all kinds of unskillful behavior, not only creating loopholes in the Vinaya, but also creating distortions in the Dhamma.


1. The word used to describe Ven. Veḷaṭṭhasīsa keeping the rice, nikkhipati, which is what makes it qualify as “stored-up” (sannidhikāraka) below, can also mean (quoting from Margaret Cone, A Dictionary of Pāli, Oxford: The Pali Text Society, 2001): “1. throws or puts or lays down; puts into; places, sets down; sets out; lays (eggs); 2. lays aside; sets aside; puts away, stores; deposits; entrusts.” As there are no more details given here, the most intuitive explanation is that he himself is keeping it in his hut or somewhere. The word is used elsewhere in the Vinaya to describe setting a bowl or robe in the sun and to describe the storehouse guardian keeping robes, so it is rendered here as “keeps.” It is not defined formally anywhere in the Canon.

2. The word sannidhikārakaṁ is derived from sannidhiṁ karoti: “he stores up,” or more literally, “he makes a store/stockpile/stash.” Here, as we will see in the definition below, it is used in a technical sense—in other words, precisely defined in the word-analysis for use specifically in the context of this rule—meaning (food) accepted by any monk on one day and eaten on a later day, regardless of what happened to it in the meantime. Because of the way it is defined in the word-analysis, taking it to be an adjective, “stored-up,” gives a more accurate translation than taking it to be an absolutive: “that he is storing” or “while storing it.”

In the Vinaya, there are many other such technical terms whose meaning differs from common usage. For example: 1) In the context of Pc 35, pavārita, literally, “invited,” means “having turned down an offer of further food (from a donor standing within hatthapāsa).” 2) In the rules dealing with the Kaṭhina [BMC], kaṭhinaṃ attharituṃ, literally, “to spread the robe-sewing frame,” means “to accept the Kaṭhina privileges, (relaxing some of the rules dealing with robe-cloth).” 3) saddhivihārika, literally, “one who lives together with,” means “a student, specifically in the context of his relationship with his preceptor.” Even in these cases, where the technical meaning of the term is very different from the common meaning, the Vinaya rules in question should be interpreted in light of that technical meaning.

3. In describing the absolutive, Steven Collins (A Pali Grammar for Students. Chiang Mai: Silkworm Books, 2005: 114-117) says, “Pali grammarians call words in this class tvādiyantapadāni, words with the endings -tvā, etc. Western scholars have called them indeclinable (past) participles, (past) gerunds, and absolutives. None of these three terms seems entirely appropriate. Although tvādiyanta-s are indeclinable, they are not participles and do not always refer to the past. ... Pali tvādiyanta-s are not inflected, since the -tvā ending is the instrumental case of an old action noun in -tu, and so, like every instrumental case, they have already been declined. ... In most cases the absolutive is used to express a previous action by the subject of the sentence, and is to be understood as a verb in the same tense and mood an the main verb. But this is not always the case.”

4. Wilhelm Geiger (A Pāli Grammar. Oxford: The Pali Text Society, 1994: 199) on the ṇamul absolutive: “There are also examples of an absol. in -aṁ and, with a -ka extension, -akaṁ, which are developments of the Skt. ṇamul absol.” He refers to this passage in William Dwight Whitney’s Sanskrit Grammar (New York: Dover, 2003: 359), “995. The accusative of a derivative nomen actionis in -a, used adverbially, assumes sometimes a value and construction so accordant with that of the usual gerund that it cannot well be called by a different name.”

In other words, the so-called ṇamul absolutive is formed from an action noun. Keeping in mind that nouns and adjectives are classed together, it should be no surprise that there would be cases of formally identical noun/adjectives and ṇamul absolutives. The following passage from Whitney gives a sense of how rare the ṇamul absolutive is: “No example of a peculiar gerundial construction with such a form occurs in either RV. or AV., although a dozen adverbial accusatives are to be classed as representing the formation...”

5. Also from Whitney (p. 359 h.), “The gerund is in the later language sometimes found in composition [i.e., compounds], as if a noun-stem: e.g. prasahyaharaha taking with violence ...” An example in Pāli is niggayha-vādiṁ: “one who, rebuking, speaks,” i.e., “one who rebukes.”

6. In many languages, a single word can have multiple grammatical functions, depending on the context. An example in English is “Watch your speed (noun). If you speed (verb), you will get a ticket for breaking the speed (adjective) limit.” In some languages, such as Thai, there are cases of a single word functioning as a noun, verb, adjective, and adverb. And even though romance languages, such as Spanish, French, and Portuguese, are inflected and have national academies devoted to maintaining “correct” usage, in the normal speech of all three languages, the distinction between adjectives and adverbs is blurred. (Spanish “va rapido,” which translates as “go quickly,” should theoretically be “va rapidamente.” “Habla claro”: “speak clearly” is similar.)

If we look in Pāli for adverbs that can also function as an adjective or that are formally identical to an adjective, they are extremely common, because, quoting Warder (Introduction to Pali. Bristol: The Pali Text Society, 2010: 116), “the accusative singular neuter of a noun or adjective (i.e., a masculine noun is made neuter, etc.) may be used as an ‘adverb’ or indeclinable.” The singular accusative masculine and feminine forms of noun/adjectives end in -aṁ as well, meaning that there will be cases where the ‘adverb’ and the ‘adjective’ are identical. For example, rassa “short”: rassaṁ passasāmi “I breathe in short” and yo ca dīghaṁ va rassaṁ vā ... “(he takes nothing not-given)–long, short ...”

7. The fact that the Thai recension differs from the others doesn’t prove that it is in error. Throughout the history of the language, there has been the tendency to “correct” irregularities in line with what contemporary scholars considered correct. Thus it is possible that the editors of other recensions may have changed sannidhikārakaṁ to sannidhikāraṁ in their editions. See O. V. Hinuber, “Pāli as an Artificial Language” in Indiologica Taurinesia, Vol. X, 1982.

8. The first passage in which sannidhikārakaṁ could be read either as an adjective or as a ṇamul absolutive occurs in the origin story to the rule under discussion, Pc 38, and the Parivāra passage explaining it. It occurs in variations (in the mood or tense of the verb) on the sentence kiṁ pana tvaṁ, āvuso, sannidhikārakaṁ bhojanaṁ bhuñjasi: “But friend, are you consuming stored-up food?” If sannidhikārakaṁ were interpreted as an absolutive, the result would be “But friend, are you consuming food from store / while storing it?” Both translations are plausible.

The second phrase occurs in the rule statement of Pc 38, as well as twice in the Parivāra explaining that rule. The phrases in the Parivāra are two versions of sannidhikārakaṁ khādanīyaṁ vā bhojanīyaṁ vā bhuñjanta: “one consuming stored-up staple or non-staple food.” Similarly, following the interpretation that sannidhikārakaṁ is an absolutive would give, “one consuming staple or non-staple food from store / while storing it.” Again, both translations are plausible.

The third example is taken from DN 27. The complete phrase is sattā sannidhikārakaṁ sāliṁ upakkamiṁsu paribhuñjituṁ: “beings began to consume the stored-up rice.” Alternately, taking sannidhikārakaṁ as an absolutive, “beings, having stored it, began to consume the rice.” Again, both are plausible.

9. This is not absence of evidence implying evidence of absence. It is clearly evident that the allowance is in fact absent in the non-offense clauses.

10. This definition limits the scope of the word ‘stored-up’ to food that a monk has received. That means that if lay people store up food prior to offering it to monks, the monks are allowed to accept and make use of it. It makes practical sense that the monks aren’t responsible for what the lay people do with the food before offering it. But once a monk has received food, he is responsible for ensuring that he doesn’t eat it on a subsequent day, which would involve teaching the lay community not to re-offer monks’ left-overs to monks.

11. The authors of the Sub-commentaries also knew that sannidhikāraka could function as an adjective: The Vimativinodanī-ṭīkā to this rule uses the word in the compound sannidhikāraka-kataṁ: “(night-watch-time (juice) that is) made stored-up,” where it is clearly an adjective. Other passages from the Sub-commentaries that either use the word as an adjective or quote it from the Canon in a way showing that they read it as an adjective are: Vinayasaṅghe-aṭṭhakathā 100, Vajirabuddhi-ṭikā 455, Vimativinodanī-ṭīkā 457.

12. Apparently the ṇamul absolutive had passed out of common usage by the time of the Commentary. Here is a list of Commentarial glosses of other ṇamul absolutives:

Sk 44: piṇḍukkhepakanti piṇḍaṁ ukkhipitvā ukkhipitvā

Sk 45: kabaḷāvacchedakanti kabaḷaṁ avacchinditvā avacchinditvā

Sk 46: avagaṇḍakārakanti makkaṭo viya gaṇḍe katvā katvā

Sk 47: hatthaniddhunakanti hatthaṁ niddhunitvā niddhunitvā

Sk 48: sitthāvakārakanti sitthāni avakiritvā avakiritvā

Sk 49: jivhānicchārakanti jivhaṁ nicchāretvā nicchāretvā

Sk 50: capucapukārakanti capu capūti evaṁ saddaṁ katvā katvā

Sk 51: surusurukārakanti surusurūti evaṁ saddaṁ katvā katvā

Sk 52: hatthanillehakanti hatthaṁ nillehitvā nillehitvā

Mv.I.25.23: samparivattakaṁ samparivattakanti samparivattetvā samparivattetvā

MN 22: anvesanti anvesantā gavesantā; urattāḷinti uraṁ tāḷetvā; dantullehakanti dantehi ullehitvā

DN 2: paripphosakaṁ paripphosakanti siñcitvā siñcitvā

DN 27: āluppakārakaṁ ... ālopaṁ katvā; pheṇuddehakanti pheṇaṁ uddehitvā

13. In glossing sannidhikārakaṁ as sannidhikārameva, the Commentary is making the point that here, the suffix -ka is equivalent to eva, defined in Cone, 2001: “eva2, like, as; as it were ...” Thus, here sannidhikāraka is equivalent to the adjective “like a storing-up,” rather than the action noun “one who stores up.”

14. Cone, 2001:

kāra 1. doing, making; a maker, a worker; 2. making; acting; act; an act of homage or service ...

karaṇa 1. (i) the act of making, doing, effecting; enacting; (ii) instrument, means of action; organ of sense or speech; 2. doing; making; effecting;

kiriyā 1. (i) doing; performance; action; activity; what one does or has done; way of acting ...

15. The suffix -ka in Pāli makes noun/adjectives. Sometimes the sense is of an action noun, “one who ...” Sometimes it is used as an adjective, “of or pertaining to ...” Examples include, dovārika: “a gate-keeper” (action noun), and Sandiṭṭhika akālika ehi-passika: “to be seen here & now, timeless, inviting all to come & see.” Sometimes a single word can take on both meanings, as in ārāmika, which can mean either a monastery attendant, or function as an adjective, meaning “having to do with a monastery.” (English uses the suffix ‘-ic’, a cognate of -ka, in the same way: “ascetic,” “romantic,” and “eccentric” can be used either as a noun or an adjectives.)

16. This is consistent with the following passage from AN 5:80, lending further support to the interpretation that it makes no difference who is storing the food. Monks living without monastery attendants and novices would probably give left-over food to animals, so there would be no chance of it being re-offered, whereas in a monastery with monastery attendants and novices, it would be natural to give left-over food to them, and easy for them to offer it again to monks out of confusion, carelessness, a desire to be frugal, or fear of taking food that was intended as an offering to monks.

“And further, in the course of the future there will be monks who will live in close association with monastery attendants and novices. As they interact with monastery attendants and novices, they can be expected to live committed to the consumption of many kinds of stored-up possessions, and to making large boundary posts for fields and crops.”