A New Approach to Prohibitive Constructions in the Rgveda and the Atharvaveda.

• Author: Hollenbaugh, Ian
• Position: Critical essay

1. INTRODUCTION

   In Vedic Sanskrit a negative command, or prohibition, is regularly expressed by ma plus the injunctive form of a verb (i.e., the augmentless verbal stem with secondary endings). These may be built to Present, Aorist, or (rarely) Perfect tense-aspect stems.1

   Hoffmann (1967) (hereinafter simply "Hoffmann") hypothesizes a semantic distribution for the three tense-aspect stems in prohibitions, with the Aor. inj. expressing one type of prohibition and the Pres. and Pf. injs. expressing another. Hoffmann's (pp. 44, 70, 91) primary semantic distinction is between what he terms "preventive" (Aor. inj.) and "inhibitive" (Pres./Pf. inj.). These are both types of interpretations or "readings" of a prohibitive verb. The Aor. inj. after ma is said to be preventive. An example of the preventive reading in English is given in (1).

   (1) ma + AOR. INJ. = "PREVENTIVE" Don't be alarmed by what I am about to tell you. This is interpreted as let it not be the case that you will be alarmed, with the presupposition that I'm afraid that you will or might be alarmed. The Pres. or Pf. inj. with ma is said to be inhibitive. English examples of the inhibitive reading are given in (2).

   (2) ma + PRES. INJ. OR PF. INJ. = "INHIBITIVE" a. Don 't leave me! (said to someone leaving = stop leaving). b. Don't cry! (said to someone crying = stop crying or don't keep crying). These are interpreted as let it not be the case that you continue leaving/crying, with the presupposition that you currently are leaving/crying.2

   I now turn to the Vedic data, beginning with examples that conform to Hoffmann's proposal. Example (3) shows five root-Aor. injs. in preventive use.

   (3) PREVENTIVE READING: AOR. INJ. ma no vadhir indra ma para du, ma nah priya bhojanani pro mosih anda ma no maghavah chakra nir bhen, ma nah patra bhet sahajanusani (RV 1.104.8) Don't smite us, Indra; don't hand us over. Don't steal our dear delights. Don't split apart our "eggs," o bounteous and powerful one; don't split our "cups" along with their contents. (3) Examples (4a) and (4b) show two Pres. injs. in inhibitive use, while (4c) shows a Pf. in inhibitive use.

   (4) INHIBITIVE READINGS: PRES. INJ. (a)-(b) AND PF. INJ. (c) a. srudhihavam indra ma risanyah (RV II. 11.1a) Hear (our) summons, Indra; stop doing damage (tr. mine). b. vi ucha duhitar divo, ma ciram tanutha apah (RV V.79.9ab) Shine forth, Daughter of Heaven; don't stretch out [= delay] your work any longer (tr. mine; cf. Hoffmann, p. 79). c. soma id vah suto astu, kalayo ma bibhitana aped esa dhvasmayati. svayam ghaiso apayati (RV VIII.66.15) Let just your soma be pressed. Kalis, stop fearing: this miasma will go away: by itself it will go away. Examples (3) and (4) are the "well behaved" examples for Hoffmann's account, though we shall see that these are in fact the exception rather than the rule.

   The structure of this paper is as follows: [section]2 demonstrates that Hoffmann's proposed distribution of the prohibitive Pres./Pf. and Aor. cannot be correct. [section]3 provides an alternative functional explanation for the inhibitive/preventive distinction, which shows how lexical semantics ([section]3.1), including actionality ([section]3.2), as well as pragmatics/context ([section]3.3) are responsible for the inhibitive or preventive character of any given prohibition. In [section]3.4, I give a quantitative overview of the distribution of prohibition types (preventive/inhibitive) with respect to stem selection (i.e., Aor. or Pres./Pf.), showing that Hoffmann's claims do not stand up against the data. [section]4 puts forth a formal analysis for the observed distribution of the two stem classes (Pres./Pf. and Aor.), noticing that simple root formations are preferred after ma in the earliest language and that this selectional restriction only gradually gives way to morphologically more complex stems (especially sibilant Aorists). The choice of Aor. or Pres./Pf. stem is thus shown to have nothing whatever to do with the "aspectual" meaning of these stem types and everything to do with the interaction between the selectional properties of ma and the formal limitations of the particular verbal bases involved. [section]5 summarizes and concludes.

2. PROBLEMATIZATION

   Hoffmann's theory has gone essentially unchallenged since its formulation and is not uncommonly presented as communis opinio (e.g., Willi 2018: 398, Clackson 2007: 162, Willmott 2007: 106). Such a semantic contrast between Aor. and Pres./Pf. injs. in prohibitions is said to provide our "clearest" evidence in support of the supposed perfective/imperfective aspectual contrast between Aor. and Pres. stems in Vedic (Kiparsky 1998: 46).4

   Yet a re-examination both of Hoffmann's treatment and of the relevant data in the RV and A V reveals that the distribution he reports is only rarely borne out in the texts. For one thing, there are many clear counterexamples in both directions. On the one hand, we find Pres. (5a) and Pf. (5b) injs. in preventive use (admitted by Hoffmann [pp. 88-90]). In (5a) the Pres. ma rarithah is immediately followed by an Aor. ma risama. Likewise, the Pf. ma vi mumucah in (5b) is most readily understood as preventive. It occurs in a typical "journey hymn," referring to a single, specific action in which Indra is asked, at the present moment, to drive to the sacrificers and drink soma without getting sidetracked at another sacrifice (cf. Jamison and Brereton 2014: 525).

   (5) PREVENTIVE PRES. (a) AND PF. (b) INJ. a. majasvane vfsabha no rarith[a.sub.[PRES.]], ma te revatah sakhiye risam[a.sub.[AOR]] (RVVI.44.11 ab) Giv[e.sub.[PRES.]] us not to exhaustion, bull. Let us not come to har[m.sub.[AOR]] in our comradeship with you, the wealthy. b. ma are asmad vi mumucah (/iKIII.41.8a) Do not unharness at a distance from us. On the other hand, we find Aor. injs. in inhibitive use (6) (admitted by Hoffmann [pp. 72-73]).5 In (6a) "turn back" presupposes that the addressee is already in the process of going away at speech time. In (6b) the addressee is not currently present, and the speaker would like him to be.

   (6) INHIBITIVE AOR. INJ. a. ni vartadhvam majanu gata (RVXA9Aa) [NB: NOT 'ma(nu) jigata] Turn back; don't keep going. b. ma_are asman maphavan jyok kah (RV VII.22.6c) Don't spend/stop spending a long time at a distance from us, o bounteous one (tr. mine, following Hoffmann, p. 73). Examples (5a) and (7) show how prohibitions containing Pres./Pf. injs. often co-occur alongside prohibitions containing Aor. injs., without obvious difference in interpretation. In (7a) a prohibitive Pres. inj., ma veh, is surrounded by two prohibitive Aor. injs., all in preventive use (admitted also by Hoffmann [p. 88]). In (7b) a prohibitive Aor. inj. follows a prohibitive Pres. inj., ma... ava srjah (again admitted preventive by Hoffmann [p. 89]). In (7c) the Pres. inj. ma... pro madah is surrounded by Aor. injunctives. Hoffmann (p. 86) asserts, without providing a reason, that the lone Pres. inj. here is inhibitive while all the Aor. injs. are preventive.6

   (7) CO-OCCURRENCE OF PRES. INJ. AND AOR. INJ. a. ma kasya yaksam sadam id dhuro g[a.sub.[AOR]], ma vesasya praminato ma apeh ma bhratur agneanijorfnam ve[r.sub.[PRFS]], ma sakhyur daksam ripor bhujem[a.sub.[AOR]] (RVW.2A3)7 Don't ever chase afte[r.sub.[AOR.]] (us, as) the specter of a nobody, a crooked man--neither of a tricky neighbor, nor of a friend. Do not pursu[e.sub.[PRES.]] the debt of a dishonest brother (against us), Agni. May we not pay fo[r.sub.[AOR.]] the "skill" of a cheating partner. b. ma no agne ava srj[o.sub.[PRES.]] aghaya, avisyave ripave duchunayai ma datvate dasate madate no, ma rlsate sahasavan para da[h.sub.[AOR.]] (RVIA89.5) Do not releas[e.sub.[PRES.]] us, Agni, to the evil man, nor to the greedy one, the cheat, nor to misfortune. Do not hand us ove[r.sub.[AOR.]] to the toothed one who bites nor to the toothless, nor to one who does harm, o strong one. c. ma te manas tatra ga[n.sub.[AOR.]] ma tiro bhu[n.sub.[AOR.], ma jivebhyah pra mad[o.sub.[PRES.]] ma_anu ga[h.sub.[AOR.]] Pitrn visve deva abhi raksantu tveha (AVS VIII.1.7= AVP XVI. 1.7) Let your mind not g[o.sub.[AOR.]] thither; let it not becom[e.sub.[AOR.]] lost; do not neglec[t.sub.[PRES.]] (those) living, do not g[o.sub.[AOR.]] after the Fathers; let all the gods guard over you here. (8) Examples (6a) and (7a) demonstrate a further peculiarity of prohibitions in Vedic: One and the same form can have both inhibitive (6a) and preventive (7a) uses, in this case a root Aor. inj., ma ga- "don't go/stop going!" Pres./Pf. injs. to the same root may likewise attest both prohibitive uses, as in (8). Example (8a) is inhibitive, as "make manifest (your) forms" presupposes that the addressee has not yet appeared or stopped hiding; (8b) is preventive, the "clash" being something that has not happened yet.

   (8) PRES. INJ. TO ^Iguh 'HIDE' WITH BOTH READINGS (INHIBITIVE AND PREVENTIVE) a. avis kfnusva rupani. ma_atmanam ava guhathah atho sahasracakso tvam, pratipasyah kimidinah (AVS IV.20.5 -AVP VIII.6.11) Make manifest (Vour) forms: do not hide yourself away [any longer]; then may you, O thousand-eyed one, look upon the kimidins. b. kirn it te visno paricaksjyam bhut, pro yad vavakse sipivisto asmi ma varpo asmad apa guha etad, vad anyarupah samithe babhutha (R V VII. 100.6) Was (this speech) of yours to be disregarded, when you proclaimed of yourself: "I am Sipivista"? Do not hide away this shape from us, when you have appeared in another form in the clash. Hoffmann's proposal, then, becomes difficult to maintain. If either tense-aspect stem can have either interpretation, we must conclude that tense-aspect stem selection alone is insufficient to retrieve the reading intended by the speaker and, conversely, that the readings intended by the speaker do not determine tense-aspect stem selection. Thus, the semantic grounds for assuming an inhibitive/preventive distinction are shaky at best and, particularly in the often opaque...