Printer Friendly

Evidence of an early metathesis among Akkadian piristum-stem nouns.


Discussions of Semitic grammatical issues frequently fail to draw a distinction between two discrete aspects of the investigation of morphology, viz. synchronic parsing and diachronic etymology. (1) The core of the typical Semitic word-stem is said to lie in the root, an ordered set of consonantal elements serving as the word's formal and semantic skeleton. The root serves as the "raw material" out of which the morphological machinery of the language "constructs" the stem in question. Identifying the root within a given stem is, in one sense, a fairly simple question of parsing. There is, however, an intrinsic duality to the role of the root within the methodology of grammatical analysis--while, on the one hand, the root serves as the origin of the word-stem, at the same time it is the product of the analyst's act of grammatical deconstruction. As long as we are dealing solely with the synchronic plane, the difference between these two aspects of the root--the root as a building block in the construction of words vs. the root as an artifact of grammatical analysis--is fairly trivial. The distinction becomes crucial, however, when the analysis shifts to the level of diachrony, where our principal concern lies in the explication of linguistic history. Once we allow for the possibility that language change might play a key role in determining the shape which any given word has assumed by the time of its documentation, it becomes apparent that it is impossible to draw a simple equation between that word's synchronic root--the set of consonants which the grammatical parsing of a given stem will extract--and its etymological root. Since linguistic formations are inherited from the past at least as much as they are generated in situ in the present, an etymological analysis will often demonstrate that, in historical terms, a given linguistic formation is ultimately the product of a set of developmental factors quite different from the grammatical processes of the language in which it is embedded.


The following pages are intended as a case study, drawn from the prehistory of Akkadian, of the significance of the distinction between the root accessible to a "parsing" and the root unearthed through an etymological analysis. The material under consideration here is the set of Akkadian nouns of the shape ([C.sub.1])i[r.sub.2]i[C.sub.3]tum, i.e., feminine nouns of the pattern "piristum" in which the second radical is -r- and the third radical is non-weak. An examination of the recent Concise Dictionary of Akkadian (2) provides us with twenty-eight instances of words of the shape (C)iriCtu(m) and (C)eriCtu(m). Of these, one at least ("miriqtu II") is probably best discarded, (3) and we should allow for the possibility that several of the remaining words actually reflect other formations--at least one of them (erimtu II "cover(ing)") is taken by von Soden as a reflection of the stem-shape *CariC-t-um, (4) two others (eristu | "cultivation" and giristu "loaf of bread") seem to contain an underlying long vowel, (5) and two further stems (pirindu and tirimtu) are liable to have yet other origins. (6) The remaining twenty-two words are listed in Table 1, along with any other terms with which the Akkadisches Handworterbuch (AHw) or the Chicago Assyrian Dictionary (CAD) associates them. Since there can be little doubt that the ongoing investigation of the Akkadian language will continue to unearth lexical items of this structure, we cannot treat this list as comprehensive. (7) Nevertheless, this list will serve as an adequate basis for the present investigation.

The claim will be made here that, in etymological terms, the members of this set of nouns may be traced back to two distinct root-types. While, for the majority of these nouns, the historical root proves to be the same as the root revealed by a synchronic parsing (viz. *[square root of ([C.sub.1][r.sub.2][C.sub.3])]), it is suggested that in a small but significant number of cases we find words which are better traced back to etymological roots in which the -r- was the third radical rather than the second (*[square root of ([C.sub.1][C.sub.2][r.sub.3])]), and that the forms in question seem to result from a metathesis which, at an early stage in the prehistory of Akkadian, reversed the order of the last two radicals. It is important to note that the hypothesized metathesis is envisioned as having operated specifically within the nominal stem-formation piristum (i.e., [C.sub.1]i[C.sub.2]i[C.sub.3]-t-um)--or, to be more precise, in the pre-Akkadian forerunner of this stem-type--rather than as operating at the more nebulous level of the abstract root. It is also worth noting that the potential relevance of the posited metathesis ultimately extends well beyond these specific forms.


The reference works on Akkadian grammar routinely describe the second -i- of the stemtype piristum as a secondary insertion, an epenthesis rendered necessary by the complexity of the cluster which would otherwise result from the addition of the feminine suffix -t- to a stem ending in a cluster (pirs-). The recent Structural Grammar of Babylonian of Buccellati (8) phrases this development in the following terms, and substantially similar analyses may be found in earlier studies: (9)
 ... Following morphophonemic rules ... when -t is added to the
 patterns pirs- and purs-, an auxiliary vowel -i is inserted between
 the second and third consonants of the pattern, e.g., {sikl-t-um}
 /sikiltum/ "gain" or {rutb-t-um} /rutibtum/ "swamp" ... (10)

While these analyses are framed in synchronic, structural terms (i.e., they are intended as characterizations of the derivational processes by which Akkadian generates a form of the shape piristum out of the underlying constituent elements *pirs- and *-t-), the notion that the shape piristum owes its second vowel to a secondary process seems to be equally valid on the diachronic, historical plane. There is a good likelihood that, at a prehistorical stage in the development of the Akkadian language, the ancestor of a stem of the piristum-type had no vowel in this position (*pirst-um or *[C.sub.1]i[C.sub.2][C.sub.3]tum), and that, in the course of the evolution of the early "pre-Akkadian" language into the documented, historical language, an epenthesis took place which had the effect of disrupting the original triconsonantal cluster through the insertion of an epenthetic vowel -i- (*pirst- > pirist-). This is made clear through a comparison of the corresponding structures evidenced by the related Semitic languages. Among the West Semitic languages we see no sign of a stem-shape *[C.sub.1]i[C.sub.2]i[C.sub.3]-t- as the counterpart to Akkadian pirist-: in Arabic we find rather the stem-shape [C.sub.1]i[C.sub.2][C.sub.3]at- (e.g., fikrat-un "idea," tiflat-un "young one (fem.)"), in Hebrew [C.sub.1]i[C.sub.2][C.sub.3]a (e.g., simha "joy," '[epsilon]gla "heifer"), and in [TEXT NOT REPRODUCIBLE IN ASCII] ([TEXT NOT REPRODUCIBLE IN ASCII] "drunkenness," [TEXT NOT REPRODUCIBLE IN ASCII] "filling up (n.)"). Such forms suggest that feminine-suffixed counterparts to the stem-shape *[C.sub.1]i[C.sub.2][C.sub.3]- in early West Semitic employed the vowel-bearing suffix *-at- rather than *-t-, and that consequently no epenthetic vowel was inserted between the second and third radicals. (11) Since the two principal branches of Semitic have adopted quite different strategies in shaping the feminine-suffixed counterpart to *[C.sub.1]i[C.sub.2][C.sub.3]-, there is every reason to suspect that Akkadian piristum represents an East Semitic innovation and thus that, at the time of the ancestral language, syllabic configurations of the shape *[C.sub.1]i[C.sub.2][C.sub.3]-t- did not yet pose a phonotactic problem (fig. 1).

To be sure, the reconstructed configuration *[C.sub.1]i[C.sub.2][C.sub.3]-t-um runs counter to certain preconceptions about the nature of Semitic syllables. Our oldest evidence for the Semitic languages comes from languages which display relatively impoverished sets of allowable syllable structures--the phonotactic system of historical Akkadian, for example, permits only single-consonant syllable onsets (i.e., syllables of the shape CV- but not, e.g., *CCV-) and only syllables ending in a short vowel (CV-), a long vowel (CVV-), or a short vowel followed by a single consonant (CVC-). Whether or not the Semitic ancestral language showed the same restrictions in its syllable system, however, is a question quite distinct from whether or not the descendant languages do. It is clear that, like any other linguistic feature, the set of allowable syllable structures for any given language is subject to change over the course of time--to take an example from Semitic, compare the wide range of syllable-types employed by the modern Maghribi Arabic dialects in contrast to the much more limited array of syllable-shapes displayed by standard Arabic (cf. Moroccan [TEXT NOT REPRODUCIBLE IN ASCII] "(I) wrote" vs. standard Arabic katabtu).


Comparative Semitic data provide us with reason to suspect that the picture that has been fashioned for the Proto-Semitic syllable, based on the model provided by the languages of classical philology, is unduly simplistic. The juxtaposition of a set of forms such as the familiar Semitic terms for "name" (e.g., Akkadian sum-um, Arabic ('i)sm-un, Hebrew sem) is instructive in that, while each of these shapes conforms to the prevailing notions of allowable Semitic syllable configurations, each does so in a different manner--the Akkadian form avoids an initial cluster through the presence of an -u-, the Hebrew form by means of the reflex of an earlier *-i-, and the Arabic form by means of a preposed syllable ('i-). In the absence of a principled analysis capable of demonstrating that any one element of this set (Akkadian -u-, pre-Hebrew *-i-, or Arabic ('i-)) could have served as the historical starting point for all of these ultimate outcomes, it seems entirely likely that the original Semitic term for "name" contained no vowel at all between its *s- and its *-m- (i.e., *sm-um), and that the various ancillary vowels found among the descendant languages must be the result of secondary, and to one degree or another independent, phonological developments. (12)

It is therefore appropriate to conjecture that early Semitic feminine-suffixed counterparts to pirs-um might have contained a rather large sequence of contiguous consonants--viz. the second and third radicals of the stem plus the suffix *t--a situation which later, as the result of changes in the phonotactic regimens of the descendant languages, was to become impossible. While the suffixless stem itself (*pirs-), in conjunction with its case-marking endings, enjoyed a quite simple syllable configuration (viz. disyllabic *pir.sum), the syllabification of its feminine analogue (*pirs-t-) would have entailed either a rather complex coda for the first syllable (*pirs.tum) or a no less complex onset for the second syllable (*pir.stum). As we have seen, in the eastern branch of Semitic this situation was ultimately brought into conformity with the new, more restricted syllable scheme of historical Akkadian (CV, CVV, CVC) through the introduction of a third, epenthetic syllable (pi.ris.tum).

In the following pages, evidence will be presented which indicates that Akkadian provides us with further, indirect support for the reconstruction of the original shape of this stem as *[C.sub.1]i[C.sub.2][C.sub.3]t-. While, as we have seen, the general pattern used by pre-Akkadian to relieve itself of the cumbersome *-[C.sub.2][C.sub.3]t- clusters clearly involved epenthesis, one must also allow for the possibility that other developments may also have taken place within certain more narrowly defined phonological situations. Even before the point in prehistory at which Akkadian acquired its blanket phonotactic proscription against triconsonantal clusters, the possibility exists that the language may have found certain complex consonantsequences less tolerable than others--much as, for example, modern English allows for the existence of word-initial kr- and kl-clusters (creep, climb) but has abandoned what were originally *kn- clusters--hence forms such as knight and know (Old English cniht, cnawan), in which the modern spelling reflects a historical stage prior to the cluster-simplification *#kn- > #n-.


Among the piristum-nouns of documented Akkadian we find reason to suspect that sequences which contained an *r as the central element of a three-consonant cluster (*-CrC-) proved particularly bothersome for pre-Akkadian. This is understandable if one considers that studies in theoretical phonology have placed the various "r"-sounds very high on the scale of inherent sonority--Vennemann's "Universal Strength Hierarchy," for example, regards the "central liquid" elements such as r as the most sonorous of the consonants, assigning them a ranking between the l-sounds and the vowels. (13) In a given pre-Akkadian *-CrC- sequence, therefore, the *r would in all likelihood have been considerably more sonorous than either of the consonants flanking it (see Fig. 2-A). The first syllable of the word would thus have had a sonority profile markedly different from that of a theoretically "ideal" syllable (Fig. 2-B), which features a clear drop in sonority from its peak in the syllable nucleus to the low point at the conclusion of the coda. The sonority profile of a syllable closed by a cluster containing an -r- at its heart would thus deviate conspicuously from the theoretically optimal curve--a deviation which would be exacerbated when the preceding syllable-nucleus was composed of one of the high vowels (such as i), which have the lowest inherent sonority of the vowels. In other words, a form containing the lowsonority vowel -i- as its nucleus followed by a triconsonantal cluster containing the highsonority consonant -r- would have had a sonority profile which gave the impression of a pair of somewhat indeterminate sonority peaks rather than the neat parabolic curve of the theoretically "ideal" syllable. (14)

In the paragraphs below, I would like to examine certain of the nouns found in the list in Table 1 with the aim of demonstrating that we may identify simpler and more semantically satisfying etymologies for them if we entertain the notion that, at a certain point in their prehistory prior to the introduction of the epenthetic vowel -i-, a systematic metathesis took place which shifted the sequence from *-iCrt- to *-irCt- (Fig. 2-C). Such a metathesis would have had the effect of bringing these words more into conformity with the phonotactic regimen which we may presume obtained at that time. Of course, subsequent developments in the phonology of pre-Akkadian--viz. the drastic reduction in the range of allowable syllable-types to what we find attested in historical Akkadian--would later reshape these words (as they would all words of the shape *CiCCtum) by means of an epenthetic vowel (hence CiriCtum). Moreover, we may assume that, during the period between the metathesis and the earliest historical documentation, the majority of the words of this class would have been reconfigured by analogical forces (i.e., *Cir(i)Ctum > CiCirtum) under the influence of the various related stems (finite verbs such as *(y)i-[C.sub.1][C.sub.2][Vr.sub.3]-, deverbal nominals such as *[C.sub.1]a[C.sub.2]a[r.sub.3]-um, *[C.sub.1]a[C.sub.2]i[r.sub.3]-um, *[C.sub.1]a[C.sub.2][r.sub.3]-um, etc.) in which the underlying root ([square root of ([C.sub.1][C.sub.2][r.sub.3])]) had not undergone the metathesis. (15) Nevertheless, the effects of the early metathesis are still sufficiently visible, I suggest, to provide a fairly clear picture of the sequence of developments which has gone into the formation of this subsection of the Akkadian lexicon.


1. erimtum, erentum

The first of the forms in question, erimtum, occurs in texts from Elam (in both Babylonian and Elamite) as the characterization of a type of brick. (16)
 bit Te-ep-Hal-ki sarri tabik imurma sa libitti udappirma sa e-ri-im-ti
 ipus "he saw that the temple of Teptihalki was in ruins, and he
 removed the (work) of sun-dried brick (libitti) and made (it) the
 (work) of e.-brick." (17)

 []Hu-ut-ra-an-te-ip-ti si-ia-an [.sup.d]In-su-si-na-ak-me e-ri-
 en-t[u.sub.4]-um-im-ma ku-si-is a-ak mi-si-ir-ma-na u sar-ra-h hi-si-e
 e-ri-en-t[u.sub.4]-um pe-ip-si-ia-ma ta-al-lu-h "Hutran-Tepti had
 built the sanctuary of Insusinak of e.-brick, but it was in decay, I
 restored it and wrote his name on the new e.-brick structure." (18)

 e-ri-en-tu[m.sub.4] u-h-na ti-pi-ha "I (Hallutap-Insusnak) made
 'stone' e.-bricks." (19)

In light of the geographical limitations on its occurrence, the suggestion has been made that erimtum should be regarded as native Elamite rather than Akkadian, (20) but the final-tum gives it a highly Semitic appearance. In principle it is possible to relate erimtum to the root underlying the verbs aramu (haramu) and urrumu (hurrumu) "to cover," and to conclude therefrom that the erimtum brick must have been provided with a glaze or similar coating. However, Akkadian (h)aramu and (h)urrumu most typically were used to refer to organic membranes and clay envelopes, and there is no evidence that they ever referred to any process associated with the treatment of bricks. As the discussion found in the CAD observes, "... the word occurs too often, and mostly in contrast with libittu 'sun-dried brick' (Elam. halat), for it to be anything but a kiln-fired brick, nor are the bricks designated by this word actually glazed." (21) The erimtum of Elam evidently corresponded to the material known elsewhere in Akkadian as agurrum, and it is conceivable that erimtum represents a native Akkadian expression which in Mesopotamia was replaced through the introduction of the foreign agurrum.

The form erimtum presupposes an earlier *(H)irm-t-um (where "*H" represents an original laryngeal, pharyngeal, or voiced uvular which has been lost). If, as was suggested above, we allow for the possibility that in this shape the sequence *-rmt- has arisen secondarily as the result of metathesis from an earlier *Himrt-, it becomes possible to compare the name of the Elamite erimtum with Akkadian amaru (emeru) "pile of bricks (often of standard dimensions)" and the related (denominal?) verb amaru "to pile up bricks."
 libittam ina dasim ustalbinma e-me-ra-am e-ti-mi-ir "I had bricks made
 in the spring, and I stacked (them) in a pile." (22)
 a-ma-ra ina kasesu libitti uqni ina subalkutisu "when he (Enlil)
 arranged the stack of bricks, when he turned over the blue-glazed
 brick" (23)

We may therefore hazard an explanation for erimtum by positing a pre-Akkadian noun *Himr-t- "*a (type of) brick" alongside *Hamar- "a pile of bricks," with the former stem subsequently undergoing metathesis to yield *(H)irmt- and, ultimately, the historical erimtum. (24)

2. miriktum

There has been little hesitation in relating the Old Babylonian term written as mi-ri-IK-tum (etc.) to the verb maraqu "to crush, grind." (25) The term in question was used to characterize areas of walls and dams which had suffered deterioration, and the assumption has been made that the deterioration in question--i.e., the crumbling of eroded brick or earthwork--must have resembled the results of the "crushing" of seeds, grains, and similar materials.
 igartam sa durim istu pan mi-ri-IK-ti-su aqqurma kima sa beli
 unahhidanni mi-ri-IK-<<ri->>-tam sati epus udannin u ana asrisu
 uttersi "I tore down the fortress wall starting with its m. and, as
 my lord had cautioned me, I made and strengthened that m. and
 restored it to its place." (26)
 mimmuka sa nuhhusata kima ana mi-ri-IK-tim annitim la ikassadu tammar
 "you shall see that all of your wealth will not be sufficient for
 (the repair of) this damaged area." (27)

There are very clear differences, however, between the natural forces of weather and water which are responsible for the appearance of a miriKtum and the deliberate action exerted in the act of crushing something by means of teeth or a tool. It is therefore possible in principle that an etymology might be sought for miriKtum which is preferable to the standard one in that it entails less of a semantic leap. In light of the posited metathesis to which the present study is devoted, I would like to suggest the possibility that the pre-Akkadian forerunner of this Old Babylonian word was built upon a root *[square root of (mgr)], with an underlying *-g- (thus miriktum < *mir(i)gtum by voicing assimilation). (28) This would enable us to relate the Akkadian miriktum to the Northwest Semitic verbal root [square root of (mgr)] (G-Stem "collapse," D-Stem "throw down"), the semantics of which are conspicuously closer to those of miriktum.
 Syriac [TEXT NOT REPRODUCIBLE IN ASCII], glossed by mat wa-npal (=
 Arabic tamayala fasaqata waqa'a) "it tottered and fell." (29)
 Targumic Aramaic mgr "to drag down, to throw over ... to diminish,
 destroy" (30)
 Egyptian Aramaic wkzy knbwzy 'l lms[r.sup.yn] 'gwr' zk bnh hskhh w'
 gwry 'lhy msryn kl mgrw w'ys mnd'm b' gwr' zk l' hbl "when Cambyses
 came to Egypt he found that temple built, and they overthrew all the
 temples of the gods of Egypt but no one damaged anything in that
 temple." (31)
 Hebrew [TEXT NOT REPRODUCIBLE IN ASCII] "you have cast his throne
 down to the earth." (32)

As in the case of erimtum above, I suspect that Old Babylonian miriktum is no longer visibly connected to the root from which it originated (viz., the lost *[square root of (mgr)]), and that we are once again in the presence of a product of the pre-Akkadian metathesis posited above (*migrtum > *mirgtum > miriktum).

3. piristu

Relating the terms erimtum and miriktum to historically underlying roots having -r- as the third radical enables us to identify a new set of quite plausible etymological starting points for these words. The same is true, I suggest, in the case of the noun piristu "secret, plot, esoteric knowledge."
 wasib mahar sarrim pi-ri-is-ti sarrim ana mat nakrim ustenessi
 "someone in the presence of the king will continuously betray his
 secrets to an enemy country." (33)
 lupteka Gilgames amat nisirti u pi-ris-ta(var.-ti) sa ili kasa luqbika
 "I shall reveal to you a hidden matter, Gilgames, and a secret of
 the gods I shall tell you." (34)

Piristu has routinely been taken to be related to the verb parasu "divide, separate, decide," (35) but the assumption of an early metathesis, I suggest, allows for a historical analysis of piristu which is a good deal simpler. Although a connection with parasu is irreproachable from the phonological point of view, the semantic displacement which it entails is not negligible--note that the verb parasu itself is seldom if ever used to refer to situations involving concealment. I would like to suggest that, if we allow for the possibility that the medial -r- of piristu was originally the third radical rather than the second radical, it becomes possible to achieve a more semantically transparent etymology by relating this word to the root [square root of (pzr)], well known in connection with secrets and clandestine matters--cf. pazaru "to travel in secrecy," puzzuru "to conceal" (Old Assyrian pazzurum "to smuggle"), pazru "hidden," etc.
 pa-zu-ur-tu-su mera Irra asser Pusu-ken useriamma pa-zu-ur-tu-su
 issibitma Pusu-ken ekallum isbatma "the son of Irra sent his
 smuggled goods to Pusu-ken, but his smuggled goods were seized and
 the palace caught Pusu-ken." (36)
 qisatisunu pa-az-ra-a-ti sa mamma ahu la usarru in libbi la ikabbasu
 itasin sabi tahazija qerebsin erubu emuru pu-uz-ra-sin "my troops
 penetrated their secret groves, into which no outsider is admitted
 and within whose boundaries no outsider may walk, and they saw
 their mystery." (37)

If we assume as the original starting point a stem *piZrt-um (i.e., earlier *pizrt-um or *pi[delta]rt-um), the metathesis proposed here provides a simple means of accounting for the attested shape of piristu.
 *piZr-t-um > *pirZtum > *piriZtum > piristum

It is quite possible that the regular shift of *-Zt-> -st- was the key to the survival of piristum in the face of the analogical pressures which might otherwise have replaced *pir(i)ztum through the creation of a new, morphologically transparent *pizirtum (with restoration of the original root-sequence): the combination of the metathesis with the shift to *-st- seems to have provided piristum with enough formal distance from its source root to allow it to survive as an independent lexical item rather than being reincorporated into the derivational network dependent upon the root [square root of (pzr)].

4. sirimtu

In the preceding cases (erimtum, miriktum and piristu), we have encountered instances of noun-stems which, as a result of the metathesis which was posited above as pre-Akkadian's systematic response to sequences of the type *-iCrt-, have ended up as lexical "isolates." Relating these stems to their etymological origins obliges us to draw a sharp line between the search for a historical etymology and the search for a grammatical parsing. Since the metathesis-process which we have posited above had long been obsolete by the time of the earliest Akkadian documents, the grammar of Akkadian was no longer capable of drawing any systematic connection between these stems and the roots to which they ultimately owed their existence. In the absence of any grammatical means of linking forms such as erimtum, miriktum, and piristum to the remaining, unmetathesized manifestations of their roots (amarum, *magarum, puzrum, etc.), it proved necessary simply to store the altered shapes as frozen lexical material.

Survival as an isolated fossil is only one of several possible outcomes to which an extinct sound change may lead, however. Even after an altered form has lost its ties to its etymological root, it is possible for it to continue to play a part in the language's derivational processes. Just as the root system provides the Semitic languages with the potential to "undo" a great deal of the effects of historical sound change, it also permits whatever products of change might survive to be used as the input for further derivations. In terms of the present investigation, I suggest that certain of the metathesized stems of the CiriCt-um type (< *CiCrt-um) became the models by which new roots of the shape [square root of ([C.sub.1][r.sub.2][C.sub.3])] developed, either in coexistence with the original etymological root [square root of ([C.sub.1][C.sub.2][r.sub.3])] or in competition with it.

It is in this light that we should consider the curious pair of Akkadian verbal roots [square root of (smr)] (attested as a D-Stem summuru and as various residual nominals reflecting earlier G-, D-, and Gt-Stems) and [square root of (srm)] (attested in the G- and D-Stems, as well as in at least one nominal form). These roots have strikingly close meanings.
 samaru saramu
 "to strive for (something), "to endeavor, to strive (for something),
 to pursue, to plot" to apply oneself (to something), to exert
 one's influence (upon somebody or on
 behalf of somebody), to be concerned"

 si-ri-im-ma kipidma hantis supra "apply yourself and think hard, and
 write me quickly." (38) ahka la tanaddi si-ri-im-ma susiiassu "do not
 delay, apply yourself and get him released." (39) u-sa-am-mar umisam
 ana hulluq mat [.sup.d]Assur "every day he endeavors to ruin Assyria."

Despite the discrepancy in the order of the radicals, [square root of (srm)] seems to belong to the same semantic complex as [square root of (smr)]. As the CAD observes, the two verbs saramu and samaru must surely be connected, since their meanings are so nearly synonymous and since they enter into the same constructions. (41) The D-stem manifestations of these roots stand in particularly close semantic proximity.
 [x] u-sa-am-ma-ru "may my hands do (whatever) I endeavor." (42)
 lipusa qata'a
 asar u-sar-ra-mu "may my hands achieve whatever I strive for." (43)
 liksuda qata'a

It is hardly surprising that this pair of verbal roots has been described as the result of a root metathesis by which the order of the final two radicals has been reversed. Lambert adduces a fairly plausible Arabic cognate ("... Meanings of the Arabic damara are equally valid for Akkadian: 'think of', 'determine upon a thing', 'apply oneself to' ..." (44)), which suggests that, of the Akkadian pair [square root of (smr)]/[square root of (srm)], it is [square root of (smr)] that predates the metathesis. A simple characterization of this sort, however, gives the spurious impression that the metathesis in question was a random, sporadic event, an arbitrary act of lexical whimsy--that there were simply times at which the radicals of roots ended up reversed. It is considerably more probable that any such metathetical development would have been grounded within a specific set of phonological circumstances. Let us consider the various attested nominal stems associated with each of the verbs derived from these two roots. Since the non-productive formations are liable to be of the most use to us in interpreting the history of these forms, we shall draw a distinction in the list in Table 2 between those derived forms which are the results of regular, productive morphology and those which are attested as frozen lexical items.

Leaving aside the productively generated forms linked with living verbs (i.e., the infinitives of the G- and D-Stems and the D-Stem verbal adjective) and the marginally documented sirm-um, (45) we find that the root containing -r- as the medial radical is associated specifically with the stem-pattern which is the topic of the present paper (sirimtum). Given the discussion above regarding erimtum, miriktum and piristum, I would like to suggest that we regard the stem sirimtum and its root [square root of (srm)] as the products of the same metathesis that we saw above, which has once again reversed the order of the last two radicals (*simrtum> *sirmtum> sirimtum). Once the form sir(i)mtum had come into being, Akkadian seems to have employed it to create a new set of verbs (viz. saramum and surrumum) by back-formation.

In principle, there thus seems to have been nothing "sporadic" about the metathesis which gave rise to the root [square root of (srm)] out of [square root of (smr)]: the reversal of the order of *m and *r here seems to reflect a regular sound change which took place under clearly identifiable phonological circumstances (viz. *-iCrtV-> *-irCtV- > -iriCtV-), just as we have seen in the other forms discussed above. The "sportive" impression which the phenomenon gives derives not from any indeterminacy in the sound change itself but from the fact that sound change is not the sole factor involved in linguistic history. Once the metathesized shape sirimtum had appeared, the language was able to make use of its productive morphological processes to produce the new verbs saramum and surrumum. The first of these evidently ultimately occupied the place in the lexicon left empty by the fading of the original G-Stem verb (*)samarum, while the new D-Stem verb surrumum, as we have seen, came in time to serve a function conspicuously close to that of the original summurum. The lexical redundancy notwithstanding, Akkadian was evidently able to tolerate the semantic overlap displayed by summurum and surrumum to the extent that neither of the two verbs had to be eliminated.

From the perspective of analytical method, the creation of the new verbs saramum and surrumum had the effect of obscuring the circumstances under which the metathesis of *m and *r originally took place--i.e., we now find words featuring the "r-m" radical-order outside of the sequence -ir(i)mt- with which it was originally associated. If we were relying solely upon an examination of the roots [square root of (smr)] and [square root of (srm)], we would be at a loss to pinpoint the details of the connections linking these two roots. The fact that the stem-configuration of the derived noun sirimtum matches precisely the shape which we have elsewhere found to be a locus of the shift *-Cr- > -rC- provides us with a non-arbitrary way to approach the question of the relation between [square root of (smr)] and [square root of (srm)]. Etymological puzzles of the [square root of (smr)] / [square root of (srm)] sort can seldom be satisfactorily resolved unless they can be placed into the context of a broader, general phenomenon. What makes this possible is the fact that phonological changes are the reflections, not of the idiosyncrasies of individual words, but rather of the general evolution of a given language's sound system.

5. siriktum

In the case of sirimtum, we found reason to suspect that the process of metathesis which was posited above resulted in a proliferation of roots, in that the effects of the metathesis on the original stem *simrtum ultimately led to the appearance of two distinct roots, the original [square root of (smr)] and the innovation [square root of (srm)]. The retention of both the metathesized and non-metathesized shapes of the root as discrete lexical forms seems to have been a rather unusual approach to the problem posed by the diversification of the shapes in which a given root was realized. When all is said and done, the historical outcome of such a situation must depend upon which of the various possible approaches pre-Akkadian adopted in order to deal with the differing manifestations of a given lexical entity. Most typically, we may presume, the force of analogy would have intervened to create a new form in which the "correct" order of the radical elements was reestablished--i.e., a new *[C.sub.1]i[C.sub.2](i)[r.sub.3]-t-um was formed to replace the *[C.sub.1]i[r.sub.3](i)[C.sub.2]-t-um which had arisen from *[C.sub.1]i[C.sub.2][r.sub.3]-t-um. Such neologisms, of course, erased all indication that any metathesis had ever taken place. From the point of view of the investigation of the language's prehistory, they are of little probative value, since, unless they can provide us with reason to believe that they have not undergone analogical reshaping, they provide no evidence of relevance to the question of the pre-Akkadian metathesis.

Another possible outcome from an ancient root diversification was for the altered shape to be maintained at the expense of the original, so that in time the new shape entirely displaced the root that served as the starting point. A set of developments of this sort may well have been responsible, I suspect, for the existence of the Akkadian verbal root [square root of (srk)] "to dedicate, grant, bestow." It is significant that chief among the substantival manifestations of this root is the noun siriktum (or seriktum) "marriage prestation, grant, gift, offering."

Akkadian saraku and its derivatives show no obvious sign of a cognate root--i.e., we find no reflex of a Proto-Semitic *[square root of (srk)], *[square root of (srk)] or *[square root of ([theta]rk)] with this sense elsewhere among the Semitic languages. If, in light of the existence of the new root [square root of (srm)] which Akkadian seems to have acquired on the basis of *sir(i)mtum, it is possible for new, independent roots to develop out of metathesized stems, we find it possible to take Akkadian [square root of (srk)] likewise to be the product of the same early metathesis, in this case operating upon an earlier underlying *sikr-t-um, the pre-Akkadian precursor of the noun siriktum.

Positing that the pre-Akkadian forerunner of siriktum had the shape *sikr-t-um allows us to compare the West Semitic verbal root *[square root of (skr)], which is well known as a verb covering situations of exchange and compensation. Compare the following examples:
 Biblical Hebrew sakar "(he) hired," s[epsilon]k[epsilon]r "wages,"
 mistakker "wage-earner"
 [TEXT NOT REPRODUCIBLE IN ASCII] sakara "(he) hired,"
 Arabic sakara "(he) thanked," sukr-un "gratitude," tasakkara "(he) was
 Palmyrene 'lh' tb wskr "the good and rewarding god" (presumably <

If we hypothesize that early pre-Akkadian had a noun *sikrtum (< *sikr-t-um) we arrive at the attested shape siriktum automatically via the processes of metathesis and epenthesis posited above. The problems of what the original meaning of this *sikrtum might have been, and of what (if any) semantic shifts it might have undergone in pre-Akkadian, remain matters meriting careful reflection, but they do not seem inherently insurmountable. (46) The verb saraku (and ultimately the other reflections of the Akkadian root [square root of (skr)]) may thus be taken to be the result of a back-formation from the metathesized *sir(i)ktum, either replacing a now lost verb *sakarum or reconfiguring it so that the order of its radicals matched that of siriktum. The change from *sakarum to sarakum may well have received its impetus from syntagms in which this verb directly governed *sir(i)ktum in its predicate, constructions which are well documented in historical Akkadian (cf. CAD siriktu mng. 2).
*[s.sub.1]i[r.sub.3](i)[k.sub.2]tam a bestowal."
 [r.sub.3] "*I bestowed
*'ana [s.sub.1]i[r.sub.3](i)
 [k.sub.2]tim as a bestowal ..."

 ana suati Zababa u Innana ana Samsuiluna rubem talimisunu sulmam u
 balatam sa kima Sin u Samas darium ana qistim liqisusum ana se-ri-
 ik-tim li-is-ru-ku-sum "on account of this may Zababa and Istar
 grant as a present well-being and an eternal life forever like Sin
 and Samas to Samsuiluna the prince, their beloved brother, may they
 bestow (them) on him as a gift." (47)
 summa ugbabtum naditum ulu sekretum sa abusa se-ri-ik-tam is-ru-ku-si-
 im "if there is an ugbabtum, a naditum, or a sekretum whose father
 bestows upon her a dowry ..." (48)

6. Provisional Notes on Two Inadequately Documented Words

a. eriptu. Though the term eriptu is only marginally attested through a few lexical sources, the gloss that it receives there renders it fairly clear that the term refers to a certain type of coat. While the conclusions drawn from the investigation of such minimally documented words must of course remain tentative, it is interesting to note that eriptu is found listed alongside a form epartu, which has a closely related meaning but in which the second and third radicals display the opposite order ([square root of ('pr)]).
 e-pa-ar-tu = na-ah-lap-tu "a coat"
 ka-su-ri-tu = MIN bur-um-tu "a multicolored (coat)"
 e-rip-tu = MIN (a multicolored coat)"
 (An VII 193ff.)
 e-pa-[ar-tu] = [...]
 ka-su-[ri-tu] = [...]
 e-ri-[ip-tu] = [...] (Malku VI 102ff.)

Positing that the location of the -r- in eriptu is due to a metathesis opens up the possibility of relating eriptu to the epartu which precedes it in these lists. The relation between them would thus be roughly comparable to what we have posited for the link between erimtum and amaru--in the case of eriptu, the absence of an original vowel separating the *-p- and the *-r- (*Hipr-t-um) led to a metathesis of the second and third radicals (*Hirptum), whereas in the case of epartu we may assume that a vowel separated these two segments (*Hapar-t-um or *Hipar-t-um), and the metathesis thus did not occur. As a plausible potential etymological root underlying these stems we may posit *[square root of ([gamma]pr)], manifested in such terms as Arabic [gamma]afara "(he) concealed," [gamma]ufrat-un "a cover," [gamma]ifarat-un "a headcloth," Syriac [TEXT NOT REPRODUCIBLE IN ASCII] "(he) covered," 'et'appar "(he) was covered," [TEXT NOT REPRODUCIBLE IN ASCII] 'afara "(he) covered," ma'fart "headband." (49)

b. piriKtum. A lexical item with a similarly restricted attestation is piriKtum, a substance associated with the preparation of beer. In two lines in the list Har-ra = hubullu we find the term piriKtum listed among ingredients employed by the brewer.
 SUN.AL.GAZ.ZA = has-lu-tu "crushed (mash)"
 SUN.AL.KUM.MA = MIN "(crushed mash)"
 SUN.SU.AK.A = mar-su "soaked (mash)"
 SUN.KA.LAL.MUN[U.sub.4] = sa pi-rik-ti "(mash) of p." (Hh. XXIII
 iii 22f.)
 MUN[U.sub.4].AL.GAZ.ZA = has-l[u-ti] "crushed (malt)"
 MUN[U.sub.4].AL.KUM.MA = MIN "(crushed malt)"
 KA.LAL.MUN[U.sub.4] = pi-ri[k-tum] "p." (Hh. XXIII
 iv 14f.)

In TIM 9, 51: 11 we likewise find piriKtum juxtaposed to a beer ingredient:
 pi-ri-ik-tum ti-ta-p[i] "p. (and) mash"

Unlike the piriktu meaning "opposition," there is no clear reason that the brewers' piriKtum should be connected with the verb parakum 'obstruct.' Faute de mieux, I tentatively suggest that we adduce the obscure verb paKaru. (50) Although the meaning of this verb remains unclear, the fact that it routinely occurs in the context of cereal products suggests the possibility that it is of relevance to the question of piriKtum. (51)
 ZID.SE.[GIG ZID].SE.MU[S.sub.5] ZID.SE.IN.NU.HA eli napsalti 9 [x x]-
 ri ta-pak-KIR "nine [...] you tapaKKir wheat flour, segussu flour,
 and inninu-barley flour over the ointment." (52)
 naptanu iqqarrub naru ussab [... d]u ina panisunu innamdima [...] u
 ZID.MAD.GA i-pak-KIR-su-nu-ti "a meal is served, a singer is
 present, [...] is set before them and the [x] ipaKKir [...] and
 scented flour for them." (53)
 [x]-ma nis-qi ZID.MAD.GA i-pa-KIR [x] "... and he ipaKKir fine scented

7. A "Northern Peripheral" Dialectal Development?

We have seen above (IV.4) that, out of the several derived nominals associated with the "[square root of (smr)]/[square root of (srm)]" complex, the radical-order [square root of (srm)] is virtually restricted to the stem-pattern [C.sub.1]i[C.sub.2]i[C.sub.3]tum. The sole counterexample to this generalization is the isolate sirmum, documented as a hapax legomenon in line 20 of ARM 2, 118, a letter in which the writer is assuring his lord of the high morale of his troops. (55)
 15. ia-ga-tum u mi-im-[ma] II n'y a ni mauvaise humeur, ni
 16. u-ul i-ba-as-[si] n'importe quoi (de semblable).
 17. su-hu-um-ma me-lu-lu-um-ma il n'y a que rire et plaisir;
 18. ki-ma bitati(ta-ti)-su-nu comme s'ils etaient chez eux,
 19. li-ib-ba-su-nu ta-ab leur coeur est satisfait.
 20. sa si-ir-mi-im-ma e-pe-es Pour toute pensee, (ils) n'(ont) que
 21. ka-ak-ki-i u da-ak na-ak- des armes et la defaite de l'ennemi.
 22. li-ib-bi wardi(mes) be-li- (Voila ce que) dit le coeur des
 ia i-da-ab-bu-ub serviteurs de mon seigneur.

Jean's reading assumes a connection between sirmum and the verbal root [square root of (srm)], as do the CAD (56) and Durand. (57) A reading based upon the sense of the verbal complex [square root of (srm)] is indeed fairly consistent with the context (lit. "... that which is of the intent(?) (is) the applying of weapons and the killing of the enemy ...").

It remains to be determined to what degree this form is of relevance to the discussion above. Sirmum shows no sign of the suffix -t- which elsewhere seems to have played a crucial role in motivating the metathesis. As one possibility, it might be speculated that the sirmum of Mari arose by back-formation from the suffixed sir(i)m-t-um. Another possibility is that, among the remote northern forms of the Akkadian language, the metathesis *-rC- < *-Cr- encompassed a broader scope than what we find in Mesopotamia, perhaps affecting not only the feminine stem *[C.sub.1]i[C.sub.2][r.sub.3]tum but the simplex *[C.sub.1]i[C.sub.2][r.sub.3]um, either as a general rule or within certain phonological situations earmarked by a high degree of syllabic complexity--one may imagine, e.g., a metathesis of this sort occurring in the construct state (*[s.sub.1]i[m.sub.2][r.sub.3] sarrim "*the intent of the king" > *sirm sarrim > *sirim sarrim, whence sirmum by analogy). A third possibility--less likely, perhaps, but theoretically not impossible--is that sirmum is actually the northern reflex of the pre-Akkadian *sirmtum, and that, instead of simplifying the elaborate cluster of *sirmtum by means of an inserted vowel (*sirmtum > sirimtum) as has happened in southern Mesopotamia, the dialect of the Mari area achieved the same end through the dropping of the final consonant of the cluster (*sirmtum > sirmum).

For the moment the question of sirmum must remain unresolved, pending further investigation of the details of the phonological and morphological development of the northern Akkadian dialects. It is intriguing to note, however, that whatever phenomenon has given rise to sirmum seems also to have left a trace in the Akkadian of Nuzi, for we find the relation between sirmum and its original root ([square root of (smr)]) recapitulated quite precisely in the Nuzi form pirqu "claim," which cannot be separated from the verb paqaru "to claim, contest" and related manifestations of the root [square root of (pqr)]. (58)
 summa A.SA.MES u E.MES pi-ir-qa irtasu Puhisenni u Hurazzi uzakkuma
 ana Unaptae inandinu summa GIS.SAR pi-ir-qa irtasi Unaptae uzakkama
 ana Puhisenni inandin "if the fields and houses become subject to a
 p., Puhisenni and Hurazzi will clear (them) and give (them) to
 Unaptae: if the orchard becomes subject to a p., Unaptae will clear
 (it) and give (it) to Puhisenni." (59)

The existence of these two curious forms (the sirmum of Mari and the pirqu of Nuzi) raises the possibility that northern pre-Akkadian underwent a metathesis of its own, whether as an extension of the southern metathesis discussed above or independently of it. It is to be hoped that continued study of the fragmentary information available on the Akkadian of this area will ultimately enable us to draw a clearer picture of the details of its prehistorical development.


In the preceding pages the suggestion has been made that the evidence available warrants positing that a systematic metathesis took place at one point in the prehistory of the Akkadian language which had the effect of turning sequences of the type *-iCrt- into *-irCt-. A total of seven individual words of the shape [C.sub.1]i[r.sub.2]i[C.sub.3]tum were identified which, with one degree of certainty or another, may be analyzed as reflections of the hypothesized metathesis. While seven examples might appear to be a rather modest number, it should be borne in mind that this number constitutes a rather large subset out of the total number of nouns of the structure [C.sub.1]i[r.sub.2]i[C.sub.3]tum listed in the Concise Dictionary (viz., twenty-two).

The existence of such words can best be explained if we assume that the pre-Akkadian ancestor of this stem-type had only a single stem-vowel (*[C.sub.1]i[C.sub.2][C.sub.3]t-um). This reconstruction parallels the long-standing approach toward the synchronic analysis of this stem-pattern, which holds that the second vowel of stems of the piristum-type should be ascribed to a secondary epenthesis. The metathesis posited here seems to have had the effect of ameliorating the awkward pre-Akkadian consonant sequence *-Crt- by drawing the *-r- leftward into direct contact with the preceding vowel *-i-, thereby yielding the much less marked syllable configuration *CirC- instead of the cumbersome original *CiCr-.

The various surviving manifestations of the effects of this prehistorical metathesis show differing fates within historical Akkadian, depending upon the connections which they show to their new roots ([square root of ([C.sub.1][r.sub.2][C.sub.3])]). In certain cases (erimtum, miriktum, piristum), the metathesized shapes appear as frozen lexical isolates, while elsewhere (siriktum) the metathesized stem has produced a new root ([square root of (srk)]) which has ousted the original root from the language, and in yet another case (saramum, samarum) the two manifestations of the etymological root--viz. the continuation of the original root ([square root of (smr])) and the new root ([square root of (srm)]) extracted from the metathesized stem sirimtum--coexist side by side. In all of these cases, it is incumbent upon us to draw a clear distinction between the synchronic root which a grammatical parsing reveals to us--e.g., the [square root of (prs)] of piristum--and the etymological root (in this case, *square root ofpzr) toward which a historical analysis points. Recognizing that these two aspects of the root need not in principle be the same leads us to a keener appreciation for the importance of the historical dimension for the investigation of Akkadian and the languages to which it is akin.
Table 1. Akkadian nouns of the shape (C)iriCtu(m)

 1. *birimtu, birimdu "sealing" baramu "to seal"
 2. erimtu I, erentu "baked brick" *
 3. eriptu (a colorful garment) *
 4. hiristu "deduction" harasu "to cut off, incise"
 5. eristu II "demand, request, need" eresu II "to desire"
 6. kiribtu "blessing" karabu "to bless"
 7. kiriktu "blocking of water flow" karaku "to dam, immerse"
 8. miriqtu I "damaged part of a building" maraqu "to crush, grind"
 9. piriktu "opposition, violent act"? paraku "to lie across"
10. piriktu (a technical term in brewing) (Not distinguished from No.
 9 above)
11. piristu "falsity, falsehood" parasu "to breach"
12. piristu "secret" parasu "to block, divide"
13. pirittu "terror" paradu "to fear"
14. *sirimtu, sirendu (a utensil or tool) saramu "cut, incise"
15. sirihtu I "inflammation" sarahu "heat," serhu "fever"
16. sirihtu(m) II "lamentation, wailing" sarahu "cry out," serhu
17. sirimtu(m) "effort, intention" saramu "strive," sirmum
18. siriptu "reddening, discoloration (of sarapu "burn"
19. seriktu(m), siriktu(m) "present, saraku I "send"
 grant, offering; dowry"
20. sirimtu, serimtu, sirindu "cutting, saramu "knock out, cut
 weeding" out"
21. teriktu, tiriktu (kind of reed or a taraku "strike," terku
 corner)? "blow"
22. tiriptu "discoloration" (a) turrupum "discolored"

a. Not listed in AHw, but proposed by von Soden ("Krankheit und Unfall
eines Kindes nach dem Karana-Brief OBTR No. 124," Nouvelles
assyriologiques breves et utilitaires 1991, no. 54) on the basis of an
Old Assyrian hapax (summa zakuma {ma} pusum u ti-ri-ip-tum la isu
simsu gumra "wenn (der Hematit) sich dann als rein erweist, indem er
einen weiss (grau)en Fleck oder sonst eine Verfarbung(?) nicht
aufweist, dann zahlt seinen vollen Preis," Ankara Museum, Nr. 9-245-
87: 6ff.). Given that ti-ri-ip-tum is conjoined here with the color
term pusum, von Soden's interpretation on the basis of the adjective
turrupu "discolored(?)" and the obscure verb tarapu appears at least
provisionally justified.

Table 2. Substantival stems reflecting the Akkadian roots
[square root of (smr)] and [square root of (srm)]

 Productive Lexical

[square root of (srm)] D-Stem infinitive tasmirtum "goal"
 summurum summiratum, sum(mu)ratum
 tismuru "intent, mindful"
 [Hapax: G-stem infinitive
 samarum] (a)
[square root of (smr)] G-Stem infinitive sirimtum "striving"
 D-Stem inf./adj. [Hapax: sirmum "endeavor(?)"
 surrumum (Maril)]

a. In Lugale III 28; see CAD samaru, lex. section.

1. The present study was made possible by the greatly appreciated support of a research fellowship from the American Council of Learned Societies. I am also grateful to the University of Chicago's Oriental Institute and its director, Professor Gene Gragg, for the opportunity to be a Visiting Scholar at the Institute. The roots of this investigation lie in work I did as a research associate for the Chicago Assyrian Dictionary project, and I express my sincere thanks to the editor in charge, Professor Martha T. Roth, and the staff of the CAD for the encouragement that they have provided and the resources to which I had access. Erica Reiner and John Huehnergard have kindly provided numerous helpful comments on earlier drafts of this study, and Nancy L. Dray has read and commented upon the manuscript.

2. J. Black, A. George, and N. Postgate, eds., A Concise Dictionary of Akkadian (Wiesbaden: Harrassowitz Verlag, 1999).

3. This hapax, found in an Old Babylonian marriage contract (1 DUG mi-ri-iq-tum, CT 48 50: 10), is probably better read as 1 DUG SAGAN! ri-iq-tum "one empty sikkatum-vessel."

4. See AHw s.v. erimtu II.

5. This is evident from the plural shapes (e.g., gi-ri-sa-a-te ADD 995: 8, e-ri-pe-tim VAB 6, 188, etc.), which show no sign of the syncope which one would expect if both stem-vowels were short (*[C.sub.1]i[C.sub.2][C.sub.3]at < *[C.sub.1]i[C.sub.2][C.sub.3]-at-). Von Soden describes giristu as a loanword from West Semitic (cf. Syriac [TEXT NOT REPRODUCIBLE IN ASCII] "cake, bread"). Moreover, one cannot rule out the possibility that a form like the obscure te-rik-tum (van Lerberghe, OB Texte 19, CBS 340: 8, as well as the lexical references cited in AHw sub te/iriktu), listed below as no. 21, actually represents the pattern tapristum (*ta(X)riKtum or *tar(X)iKtum); it is evidently distinct from the teriktum "uncultivated land" (read as teriqtum by the AHw) discussed in note 7 below.

6. All that is clear from the context of the few attestations of pirindu is that it seems to be a fruit of some sort, and that it is often associated with pomegranates.

U pi-ri-in-di GIS.NU.UR.MA "p. (of) the pomegranate tree" (STT 94: 12)

lu pi-ri-in-du l[u ...] lurindu " ... or p.'s, or [x], or pomegranates ..." (KAR 373: 8)

x GIS pi-ri-in-du x munziqu "x p.'s and x raisins" (as poultry feed) (VAS 20 72:6)

On the basis of the West Semitic verbal root [square root of (prm)] (Hebrew param "(he) chopped," Syriac [TEXT NOT REPRODUCIBLE IN ASCII] "(he) cut into small particles"), pirindu has been traced back to an earlier *pirimtum and given such tentative glosses as "Fruchtschnitze?" (AHw) and "a slice" (Concise Dictionary). A considerably more compelling comparandum, however, is to be seen in Arabic firind-un (pl. faranid-u) "grain of the pomegranate." While it is clear that there must be a connection of some sort between Akkadian pirindu and Arabic firindun, the details of that connection remain to be resolved. Are these two forms cognate to one another, reflecting an otherwise lost quadriliteral root *[square root of (prm/nd)]? Or is firind-un to be traced back to Akkadian pirind-u, presumably having entered Arabic via Aramaic or Iranian? Or might the term have originated outside of Semitic altogether and entered both languages as loanwords?

Tirimtu, the name of a type of beverage container, should probably be regarded as a by-form of tilimtu, a Sumerian loanword referring to a type of jar.

7. For example, the word read by the AHw and the Concise Dictionary as "teriqtu(m)" "uncultivated land," which is held to be related to raqu "to be empty," is probably better read as teriktum (see W. H. van Soldt and M. Stol, "The Old Babylonian Texts in the Allard Pierson Museum," Jaarbericht ex Oriente Lux 25 [1977-78]: 52), judging by the plural te-er-ke-ti-si-na found in OECT 3 33 (= AbB 4 111): 26; one may presume that this is distinct from the teriktum listed as no. 21 below). Rather than treat teriktum "uncultivated land" as a derivative of taraku "to strike," it might be better to regard it as a derivative from a lost Akkadian cognate of the Arabic verb taraka "(he) left behind, left aside" (hence originally "*(land) left unworked"). For the obscure te-rik-tum (van Lerberghe, OB Texte 19, CBS 340: 8, etc.), possibly to be read as teriktum as the AHw has done, see note 5 above.

8. G. Buccellati, A Structural Grammar of Babylonian (Wiesbaden: Otto Harrassowitz, 1996).

9. See E. Reiner, A Linguistic Analysis of Akkadian (The Hague: Mouton, 1966), 117f., and W. von Soden, Grundriss der akkadischen Grammatik (Rome: Pontificium Institutum Biblicum, 1969), [section]60 (lb).

10. Buccellati, 145.

11. To be sure, a limited number of stems of the shape [C.sub.1]i[C.sub.2][C.sub.3]at- are also found in Akkadian--see D. O. Edzard, "Zu den akkadischen Nominalformen parsat-, pirsat-, and pursat-," Zeitschrift fur Assyriologie 72 (1982): 68-88, esp. 77ff. There is a clear preference for the shape pirsat- over pirist- in the case of "geminate" ([square root of ([C.sub.1][C.sub.2][C.sub.2])]) roots (e.g., simmatum "paralysis" rather than *simimtum, zibbatum "tail" rather than *zinibtum), but certain of the details of the circumstances under which the stem-shape containing -at- appears rather than the shape with a second -i- remain to be determined--cf. the minimal pair simdatum "royal decree" alongside simittum "crosspiece of a yoke" (see Reiner, 122; Buccellati, 145). The ultimate resolution of this complex of questions must await a comprehensive historical assessment of the relation between the two shapes of the feminine marker (*-at- vs. *-t-).

12. For the reconstruction of Semitic *sm- and related issues in early Semitic phonology see D. Testen, "Secondary Vowels in Semitic and the Plural Pronominal Endings," in Kurytowicz Memorial Volume, Part One, ed. Wojciech Smoczynski (Krakow: Universitas, 1995), 543-51.

13. T. Vennemann, Preference Laws for Syllable Structure and the Explanation of Sound Change (Berlin: Mouton de Gruyter, 1988), 9.

14. For further discussion involving the theoretical ramifications of syllabic structures for Semitic linguistics, see L. Edzard, "Semitic Phonology and Preference Laws for Syllable Structure," in Semitic Studies in Honor of Wolf Leslau on the Occasion of his Eighty-Fifth Birthday, ed. A. S. Kaye (Wiesbaden: Otto Harrassowitz, 1991), 396-410.

15. This is presumably the source of many of the nouns of the shape (C)iCirtum documented in historical Akkadian. Of course, words of the shape "piristum" which were coined later (i.e., after the point at which the pre-Akkadian metathesis rule ceased to be in effect) would likewise show no sign of the metathesis *-Crt- > *-r(i)Ct-, and it would be difficult to determine which words have reversed the effects of the metathesis and which words simply postdate it. K. Hecker's reverse dictionary (Rucklaufiges Worterbuch des Akkadischen [Wiesbaden: Otto Harrassowitz, 1990]) lists the following instances of this stem-shape in Akkadian: sibirt-, idirt-, midirt-, sidirt-, tidirt-, bihirt-, sihirt-, gimirt-, kimirt-, simirt-, sipirt-, sipirt-, niqirt-, isirt-, kisirt-, kisirt-, nisirt-, kisirt-, misirt-, itirt-, mitirt-, sitirt-, izirt-, kizirt-, and nizirt-.

16. See W. Hinz and H. Koch, Elamisches Worterbuch (Berlin: Verlag von Dietrich Reimer, 1987), 399f.

17. V. Scheil, Textes elamites-semitiques (Paris: Ernest Leroux, 1900), 120, II. 2ff.; see F. W. Konig, Die elamischen Konigsinschriften, Archiv fur Orientforschung, Beiheft 16 (1965), 72 no. 9.

18. Konig, 91, text no. 39, ll. 2ff.

19. Konig, 169, text no. 77, 1, 2. For a discussion of what "stone" bricks might have been, see the literature cited by Hinz and Koch, 1202 (sub u-h-na).

20. "De ce que erimti apparait dans un texte semitique il nous semble difficile de concluire necessairement que ce mot soit lui-meme emprunte au babylonien, comme le veut Scheil; en dehors des textes de Suse, bien difficile serait de trouver un mot assyro-babylonien erimtu ayant le sens de brique. Les textes semitiques des rois d'Elam contiennent d'ailleurs, souvent des mots elamites ..." (M. Pezard, Mission a Bender-Bouchir: Documents archeologiques et epigraphiques [Paris: Ernest Leroux, 1914], 70 n. 2).

21. CAD s.v. erimtu, Cf. similarly Scheil, 120, and Konig, 72f. n. 9.

22. T. G. Pinches, "The Cappadocian Tablets Belonging to the Liverpool Institute of Archaeology," Annals of Archaeology and Anthropology (Univ. of Liverpool) 1 (1908): 53, II. 6ff. (No. 1, pl. 19).

23. CT 3 pl. 38: 62f.

24. I find the relation between the reconstructed pair *Himr-t- "brick" and *Hamar- "stack of bricks" to be curiously reminiscent of Arabic and Southwest Semitic pairs such as Arabic qit at-/qita- "piece/pieces" and Tigre [TEXT NOT REPRODUCIBLE IN ASCII] "hole/holes." The formation of broken plurals from feminine singular nouns of the shape CVCC-at- through the dropping of the feminine suffix and the insertion of -a- between the second and third radicals (CVCaC-) is a widespread phenomenon throughout those Semitic languages which employ broken-stem patterns to express plurality--see R. Ratcliffe, The 'Broken' Plural Problem in Arabic and Comparative Semitic: Allomorphy and Analogy in Non-Concatenative Morphology (Amsterdam: John Benjamins. 1998). Is it possible that Akkadian amar-/emer-"pile of bricks" is a remnant of such an ancient broken plural?

One might also wonder whether there is any connection between Akkadian amar-um "pile of bricks" and the Arabic term 'amarat-un (pl. 'amar-, 'amar-, 'amarat-) "(a heap of) stones ... set up in order that one may be directed thereby to the right way ... a structure like a [manarat- or mosque tower], upon a mountain, wide like a house or tent, and larger, of the height of forty times the stature of a man, made in the time of [the pre-Islamic peoples of] Ad and Irem ... it consists only of stones piled up, one upon another, cemented together with mud, appearing as though it were a natural formation ..." (E. W. Lane, An Arabic-English Lexicon [New York: Frederic Ungar, 1955], 97). It is curious--and quite possibly of relevance to the present discussion--that Arabic 'amar-un shows what appears to be an etymological doublet in 'iram-un ("stones set up as a sign, or mark, to show the way in the desert ... like a man in a standing posture upon the head of a hill, whereby one is directed to the right way ..." [Lane, 51]), with what appears to be a reversal of the two final radicals quite similar to what we find in the Akkadian forms amarum and erimtum. If further instances of such doublets were to be identified within Arabic or elsewhere in West Semitic, it might become appropriate to ask whether the metathesis rule which we have posited above as "pre-Akkadian" or "early East Semitic" might actually have been a feature of the Semitic ancestral language.

25. The use of maraqu with the meaning "shatter" (used in reference to tablets or containers) is, as the CAD notes (s.v. maraqu mng. 2), restricted to the Neo-Assyrian period and thus is unlikely to be of relevance for the early history of Akkadian.

26. ARM 2, 88: 18.

27. R. Frankena, Tabulae cuneiformes a F. M. Th. de Liagre Bohl collectae Leidae conservatae IV (Leiden: Nederlands Instituut voor het Nabije Oosten, 1965), no. 52: 31f.

28. Although the spelling -IK- of miriktum does not make clear the nature of the velar which this word contains, the occasional attestation of the construct form (me-er-KI-IT ARMT 14, 13: 19) and the plural form (mi-ir-KI-ti-su ARM 3, 42: 16) shows that synchronically this word contained either a -k- (miriktum, mirket, mirketu) or a -q-. We may presume that the forms of this word containing a vowel after the velar have been reshaped under the analogical influence of the basic form miriktum; we would otherwise expect a paradigm miriktum, construct *migret, plural *migretu).

29. R. Payne Smith, ed., Thesaurus syriacus (Hildesheim: Georg Olms Verlag, 1981), 2007.

30. M. Jastrow, A Dictionary of the Targumim, the Talmud Babli and Yerushalmi, and the Midrashic Literature (New York: Title Publishing, 1943), 730.

31. A. Cowley, Aramaic Papyri of the Fifth Century B. C. (Oxford: Clarendon Press, 1923), 112 (Text 30, ll, 13f.); see B. Porten and A. Yardeni, Textbook of Aramaic Documents from Ancient Egypt | (Jerusalem: Hebrew University, 1986), 68.

32. Psalm 89:45.

33. YOS 10, 25: 31.

34. Gilgames Tablet XI, 9f.

35. See AHw s.v. piristu(m).

36. B. Kienast, ed., Die altassyrischen Texte des Orientalischen Seminars der Universitat Heidelberg und der Sammlung Erlenmeyer-Basel (Berlin: Walter de Gruyter, 1960), text no. 62, ll, 28ff.

37. Rassam Cylinder vi 65ff.; see R. Borger, Beitrage zum Inschriftenwerk Assurbanipals (Wiesbaden: Otto Harrassowitz, 1996), 55.

38. PBS 1/2, 67: 20f.

39. TCL 17, 59: 28f.

40. Tukulti-Ninurta Epic "ii" 17; see E. Ebeling, "Bruchstucke eines politischen Propagandagedichtes aus einer assyrischen Kanzlei," Mitteilungen der Altorientalischen Gesellschaft 12/2 (Wiesbaden: Otto Harrassowitz, 1938), 7.

41. See the CAD's discussion sub saramu.

42. CT 37, pl. 20 (No. 85965 iii) 57.

43. R. C. Thompson, The Prisms of Esarhaddon and Ashurbanipal Found at Nineveh, 1927-8 (London: British Museum, 1931), pl. 18 vi 26f. (Prism T); see Borger, 171.

44. W. G. Lambert, Babylonian Wisdom Literature (Oxford: Clarendon Press, 1960), 231.

45. For discussion of Mari sirmum, see section IV.7 below.

46. Of the three general semantic areas manifested in the forms which we have adduced here--viz. "bestowing" (Akkadian), "gratitude" (Arabic), and "compensation" (elsewhere in West Semitic)--the last two have been related to one another already by J. L. Palache (Semantic Notes on the Hebrew Lexicon [Leiden: E. J. Brill, 1959], 38), who compares Hebrew hodu "(they) gave thanks/praise" and Arabic wada "(he) paid wergild" as semantic analogues. The senses "bestowing" and "hiring" share an intrinsic asymmetry of status, in that the individual who is the source of the object exchanged (whether that individual be a god, a monarch, a family head, or an employer) stands in an elevated position with respect to the individual who is the recipient of that object. One might conjecture that the Arabic sense "gratitude" originated under similarly asymmetrical conditions (perhaps "*an exchange carried out in recognition of services rendered," comparable to what is found elsewhere in West Semitic), but by the time of documented Arabic the term's semantics seem to have become rather more "egalitarian."

47. YOS 9, 35: 144ff.; see D. R. Frayne, The Royal Inscriptions of Mesopotamia--Early Periods, volume 4: Old Babylonian Period (2003-1595 BC) (Toronto: Univ. of Toronto Press, 1990), 388, II. 128ff.

48. CH [section]178 (xxxvii 61ff.); see M. T. Roth, Law Collections from Mesopotamia and Asia Minor (Atlanta: Scholars Press, 1995), 117.

49. See W. Leslau, "Southeast Semitic Cognates to the Akkadian Vocabulary I," JAOS 82 (1962): 1-4. It is unclear how best to interpret the connection between the West Semitic words built from *[square root of ([gamma]pr)] and the Akkadian terms upru "a type of headdress" and aparu "to provide a headdress." The voweling of eriptum is consistent with an earlier *[gamma]iprtum, whereas that of aparu (rather than *eperu) is difficult to reconcile with the presence of an original *[gamma]. The lexical form epartu listed above seems to be an Assyrianism from either *[gamma]apartum or *[gamma]apartum.

50. I am assuming that this verb is distinct from the verb pakaru "to tie," although it is tentatively listed under the latter in the AHw. Since the known instances of the verb labeled here "paKaru" are all written with PIS as the final sign, there is no way to know whether this is to be read as -kir, -gir, or -qir.

51. Lambert renders the verb as "offer" in his readings of OECT 11, 47: 11 and RA 91, 68: 5, but this translation is less felicitous in the ritual described in Or NS 39 (1970): 120. Perhaps a meaning on the order of "sprinkle" or "scatter" would fit better in these passages. Piriktum (if it should prove to be related to this paKaru) might thus have originated as "*a sprinkling (of something added during the brewing process)."

52. R. Caplice, "Namburbi Texts in the British Museum IV," Orientalia n.s. 39 (1970): 120, II. 62f.

53. OECT 11, 47: 9ff.

54. W. G. Lambert, "Processions to the akitu House," Revue d'Assyriologie 91 (1997): 68, 1. 5.

55. C.-F. Jean, Lettres diverses, Archives royales de Mari II (Paris: Imprimerie Nationale, 1950), 198.

56. "[T]he endeavor of fighting battles and defeating the enemy" (CAD s.v. sirmu).

57. "L'unique objet de desir, c'est livrer combat et battre I'ennemi"--J.-M. Durand, Les documents epistolaires du palais de Mari (Paris: Editions du CERF, 1998), vol. II, 200 = text no. 577.

58. If the interpretation given here is correct, we must assume that the Nuzi form pariqanu, which is attested alongside the regular and expected shape paqiranu (found both at Nuzi and elsewhere) owes its shape to the influence of the metathesized pirqu.

59. A. Shaffer, "Kitru/Kiterru: New Documentation for a Nuzi Legal Term," in Studies Presented to A. Leo Oppenheim (Chicago: Oriental Institute, 1964), 182, 11. 29ff.
COPYRIGHT 2003 American Oriental Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2003 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Testen, David; Minnesota, St. Paul
Publication:The Journal of the American Oriental Society
Date:Jul 1, 2003
Previous Article:"Arabia Without Spices": an alternate hypothesis: the issue of "Makkan trade and the rise of Islam".
Next Article:The origins of separation of the sexes in China.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |