Structural and Formal Analysis

While all the treaties edited in the present volume differ somewhat in form and structure from one another, they also display considerable structural and formal similarities. It is easy to see that they consist of a limited number of structural elements, which may not all be obligatory or appear in the same order in every document, but have a well-defined function and are formulated according to well-established conventions. Chart 1 presents a survey of the attested elements in the order in which they appear in no. 6, the longest and best preserved text in the corpus.

Chart 1. Attested Structural Elements

Explanation: + indicates attested elements; - indicates omitted elements; * indicates omitted but structurally implicit elements; Lack of +/-/* indicates textual damage.

A detailed analysis of these elements is not within the scope of the present edition, and the following discussion will hence limit itself to the barest essentials.

Preamble

The Preamble, which can be taken as obligatory in all texts, opens the document by identifying the contracting parties. Its standard form is adê ša A issi B "Treaty of A with B", where A stands for the Assyrian king with his title and parentage and B for the other contracting party, usually another king (with his title) and a nation. This formula is attested in nos. 5, 6, 8, 10 and 11, and corresponds to the phrase adê ša A "treaty of A" by which the texts are referred to in the treaties themselves (no. 1 i 13, ii 17, v 8; 6:390, 555A, 612A, 666; 12:1f) and elsewhere.^[[15]]

adê ša "treaty of" can also mean "treaty which", depending on the context. This latter meaning occurs in the second paragraph of no. 6 (divine witnesses), where the words adê ša have been omitted as unnecessary but are implied by the subjunctive predicates concluding the paragraph, as well as further down in the same text (lines 41 and 283). It would, however, be a mistake to apply the meaning "treaty which" to the preamble also, as was done in earlier editions of the text.^[[16]] Both the preamble and the witness section are independent, though interconnected, structural elements, as shown by the rulings and seal impressions separating them in all manuscripts, and they have to be rendered accordingly. Note the unambiguous formulation of these sections in the Sefire treaties: "Treaty of A with B; this treaty which A concluded [before Aššur] and Mullissu..." (KAI 222 A).

Seal Impressions

Inserted between the preamble and the witness section in no. 6 are three cylinder-seal impressions; a caption describing the seals is inscribed at the head of the four columns of the obverse (i-iv) and separated from the text by a dividing line; see the copy on p. 28.

The middle seal, with the short legend "Of the God Aššur and the City Hall", dates from the time of the Old Assyrian city-state.

The impression on the left, from a Neo-Assyrian seal, shows the king of Assyria standing between the gods Aššur and Mullissu, and has the following legend:

"The Seal of Destinies, with which Aššur, king of the gods, seals the destinies of the Igigi and Anunnaki of heaven and earth, and of mankind. What he seals with it, he does not alter. He who should alter (it), may Assur, king of the gods, and Mullissu, together with their children kill him with their mighty weapons. I am Sennacherib, king of Assyria, a prince who fears you. Whoever erases my name and discards this Seal of Destinies of yours, erase his name and seed from the land!"

The impression on the right, showing the king kneeling between the gods Aššur and Ninurta, is from a Middle-Assyrian seal; both gods are mentioned in the largely illegible legend (lines 4f).

As recently pointed out by A. George, there can be little doubt that the document ratified by these sealings became, on the mythological plane, a 'Tablet of Destinies'.^[[17]] In impressing them, the king of Assyria was 'sealing destinies' as Aššur's earthly representative. These sealings thus had a dual function: to add to the eternal validity of the treaty terms, and to sanctify and protect the treaty tablet by portraying the god Aššur himself as it's ratifier and guardian. Compare lines 400ff of Text 6.

How many treaty tablets were actually sealed remains unclear. In the present corpus, the only text with a sealing is no. 6. It is definitely missing in nos. 1, 3, 8, and 10-12, all of which are one-column tablets. On the other hand, it could have been impressed on nos. 2, 5 and 13, all of which, like no. 6, are multi-column tablets.

Divine Witnesses

All Neo-Assyrian treaties appear to have been sworn agreements, witnessed and sworn by the gods of both contracting parties. A list of divine witnesses (or an adjuration formula mentioning the same gods) is found in five texts of the corpus, and is certainly missing only in one (no. 8). The reasons for its omission in this treaty, which certainly was a sworn document, are unclear but possibly related to the small size of the text. As the gods by which the treaty was sworn were listed in the curse section, it may be that it was omitted as unnecessary in order to save space. Note that other stereotyped structural elements of the treaties, such as the adjuration or the treaty violating clause, could also be given in abbreviated form or even omitted altogether (see below).

Where given, the witness list immediately follows the preamble. In accordance with the order of the contracting parties, which reflects their mutual status, the gods of Assyria precede those of the other contracting party. This order is also followed in the Sefire Treaties (KAI 222 A), in the curse section of the treaties, and in references to treaty gods outside the treaty corpus itself (compare, e.g., ABL 918 cited above, p. XVII).

From the reign of Esarhaddon on, astral gods start heading the witness list prominently (cf. nos. 6, 8 and 11, but also references to sworn agreements such as Iraq 34 p. 23). This surely reflects increased interest in astrology, and it may be that having the treaties concluded "under the stars" was thought to add to their durability.

As it appears from no. 6 (lines 153ff and 494), and passages such as ABL 213:7ff ("let the gods come for the treaty"), the treaty gods were, at least in theory, physically present at the oath-taking ceremony in the form of their statues or otherwise. Whether this was always the case cannot of course be determined. But it should be noted that all the astral gods mentioned in the witness list of no. 6, with the exception of Mars, were indeed visible at the time this treaty was concluded.^[[18]]

Adjuration

An adjuration clause is attested in four texts (nos. 2, 6, 9 and 10); it is missing in no. 8, but may have been included in all the remaining texts. In nos. 6ff, it immediately follows the list of treaty gods; in no. 2, it is inserted at the end of the document.

The clause is slightly differently formulated in all the texts where it occurs. In no. 6, it is phrased in the imperative, "Swear by the god (...)!" In no. 10, the verb swear is for reasons of space omitted altogether, and only the gods sworn by are given. The relevant passage in no. 9 is broken, but it is likely that the adjuration there was expressed in the form of a 1st person plural vow, as the rest of the text. The formulation in no. 2 resembles that of no. 6, but the verbal form is different: "You are sworn by the god (...)!"^[[19]]

This last formula is psychologically interesting, because it reveals something of the Assyrian attitude towards the other contracting party. The same formula is quite well known from Mesopotamian exorcist literature, where it is used for conjuring evil demons (e.g., nīš ilāni rabûti ša šamê u erṣeti tummâtunu "you are sworn by the great gods of heaven and earth", KAR 227 ii 45). What is more, these conjuration formulae are followed by exorcisms formulated exactly like the stipulations of the present treaties! (In addition to KAR 227 ii 50, šumma tumaššarūšunūtimma, see also the Lamaštu passages cited by W. Farber, ZA 64 177f.) One gets the impression that the Assyrian kings were approaching their treaty partners - mostly strangers and actual or potential enemies - basically as evil demons, to be treated with the methods tested and found effective by domestic science.

Historical Introduction

A formal introduction to the contents of a treaty is attested in two texts of the corpus only (nos. 6 and 10), and in both cases it is very short. On the other hand, it is certainly missing only in two texts (nos. 8 and 9), and may have been included in the others. In any case, judging from the brevity of the attested examples, it seems that, possibly due to Assyrian superiority, the introduction was losing ground as a structurally significant element in Neo-Assyrian treaties, even though it was not omitted altogether.^[[20]]

Formulation of Treaty Stipulations

In most texts, individual stipulations are divided by rulings into separate sections corresponding to modern treaty articles. The order of the sections in no. 6 reveals a well thought-out logical scheme systematically covering all possible forms of threat against the ruling house, starting from the definition of loyal conduct, and gradually proceeding to open rebellion and murder of the king.

Typologically, the attested stipulations fall into declarations, demands, injunctions, obligations, commands and prohibitions, and they are usually phrased in the form of sentences beginning with the particle šumma and ending in a subjunctive predicate. 'Normal' main clauses with indicative predicates are interspersed among the subjunctive ones, but they form a clear minority. The meaning of the subjunctive clauses has been the subject of much controversy and confusion and must hence be briefly discussed here.

The predominance of these clauses in the treaties is due to their character as sworn documents. Most of the stipulations are things that the other contracting party was pledged under oath to accept, do, or avoid doing; they thus carry the implication of a solemn, oath-bound pledge. This implication was expressed in Akkadian by a special linguistic formula used in all kinds of solemn, oath-bound statements. In 1st person verbal forms, it marks solemn promises or assertions, in 2nd and 3rd person forms, solemn pledges or assertions, depending on the tense of the verb.

A survey of the relevant formulae, which differed from dialect to dialect is given in Chart 3. They all involve a predicate in the subjunctive and are mostly preceded by an oath formula (e.g., "I swear by god so-and-so"), which could occasionally be omitted. The subjunctive verbal forms in the treaty stipulations are correspondingly connected with the preceding adjuration formula ("Swear/You are sworn by ...!").

Chart 2. Attested Types of Stipulations

Chart 3. Vow, Pledge and Assertion Formulae

FUTURE POSITIVE

FUTURE NEGATIVE

PAST NEGATIVE

Since the Neo-Assyrian formulae begin with the conjunction šumma, which usually means "if", they have in some editions been rendered as conditional sentences,^[[21]] but this is a mistake. These formulae have no more to do with conditional sentences than the corresponding formulae in other dialects of Akkadian; their origin may lie in oath-bound conditional clauses ("I'll be damned if..."), but in their actual usage they simply are the linguistic form of solemn statements in this dialect. As far as the treaties are concerned, note that the word šumma is missing in several subjunctive clauses in Text 6 where it is expected (e.g., in lines 49, 167, 283, 306, and 407) and hence must have been semantically redundant. The specific meaning of the formulae resided in their subjunctive predicates, which when negated indicated a positive vow or pledge, and when positive, a negative one.

Thus a subjunctive expression like (šumma) la taqabbûni "you shall say" was functionally more or less equivalent to indicative taqabbi "you will say", the only difference being - as in English - that the former was more solemn and binding than the latter. The equivalence of the two expressions is put beyond doubt by the fact that they occur as textual variants in different manuscripts of Text 6:

la tu-šá-aṣ-bat-a-ni "you shall install" (line 85, ms. G)

= tu-šá-aṣ-ba(t)-ta "you will install" (ibid. AId);

ta-šá-kan-a-ni "you shall not place" (line 301, B)

= la ta-šá-ka-na "do not place" (ibid. H).

Similar variation between other functionally equivalent expressions also occurs elsewhere in the same text:

li-ra-ah-ṣa "may they (the chariots) get drenched" (line 615, GH)

= lu-šar-hi-ṣu "may they drench (the chariots)" (ibid. AK);

lib-('i)-iš "may it stink" (605, HLTh)

= lu-ba-i-šu "may they make it stink" (ibid. frg.);

lu-ú-pal-li-šu "may they pierce (your women)" (line 598, h)

= lu pal-lu-šá "may (your women) be pierced" (ibid. T);

li-tah-li-qa-ku-nu "may it get lost for you" (line 445, A)

= lu tah-[li-qa-ku-nu] "may it disappear from you" (ibid. g).

That the difference between indicative verbs and subjunctive šumma clauses was relatively slight is further indicated by the fact that in Text 13 a single subjunctive šumma clause (iii 4) appears inserted between several indicative sentences and contrariwise indicative clauses are found among strings of subjunctive šumma clauses in Texts 4 and 6 (lines 83-91, 198-211, 380-384). The same phenomenon is attested in the Aramaic Sefire treaties, where paronomastic constructions with infinitive absolutus alternate with simple verbal forms (e.g. hskr thskrhm "you must surrender them" followed by simple yhskr in KAI 224:2f, rqh trqhm "you must conciliate them" followed by 'rqhm ibid. 6f; see also ibid. 12f and 18).

No subjunctive šumma clauses are found in Text 5, where all stipulations are phrased as indicative main clauses.

Of the contracting parties, the king of Assyria is basically identified by his name and title, occasionally by 1st person verbal forms and suffixes ("you shall send to me" 2 iii 10', "our life" ibid. v 1; "these cities which I destroyed" 5 iii 4', "I have appointed" ibid. 6; "if you do what is good to me 13 ii 7' etc.). The other party is basically addressed in the 2nd person, but may also be referred to by name (nos. 1, 2 and 5).

Violation Clauses

The stipulations are followed by a clause for the case of treaty violation, leading to the subsequent curse section. This clause is phrased slightly differently in every text of the corpus where it is extant, but it basically resembles similar clauses found in other types of texts involving curses or penalties (royal inscriptions, decrees, grants, legal documents, epitaphs, etc.). The best preserved versions of the clause in Texts 6 and 11 read:

"Whoever changes, disregards, transgresses or erases the oath of this treaty tablet, or disregards this treaty and transgresses its oath" (6:397ff, followed by curse);

"Whoever alters the wording of this tablet, or sins against the treaty of the great gods" (11 r. 5f, followed by curses).

In the former case, the clause cited is followed by injunctions against destroying the treaty documents, after which the curse section proper begins. The injunctions, formulated as subjunctive šumma clauses, have (by analogy to the violation clauses leading to the ceremonial curse section) been rendered as conditional sentences, but it is possible that they in fact represent treaty stipulations and have to be rendered accordingly ("You shall not remove it, consign it to fire, etc."). In that case, the curse section would lack an immediately preceding violation clause, like nos. 3, 4 and 10 (see below).

In no. 9, a treaty in the form of a 1st person vow, the clause is formulated "If we should transgress, break, erase or [ ... ] this treaty". In no. 1, the relevant passage is broken but seems to have read "Whoever sins against this treaty and does not carry out his duty" (line 15'). In no. 8, the clause is abbreviated into the form "Whoever (sins) against this treaty", and it is omitted altogether in nos. 3, 4, and 10, obviously as redundant. No data are available from nos. 5, 7 and 13.

In no. 12, the clause takes the form of a conditional sentence identical with the repetitive clause preceding the ceremonial curses (see below) in Texts 2 and 6: "If you should sin against this treaty". The predicate of this conditional sentence is abnormally in the subjunctive in all three texts, but appears at least once in the normal indicative mood in no. 2 (ihtiṭi, v 9). It seems likely that the abnormal subjunctives in this clause are due to a contamination with the clauses preceding traditional curses, which involve a relative pronoun (mannu ša) and hence a predicate in the subjunctive. Note that one of these clauses in text 2 actually reads " If PN, his sons, or his magnates who(!) sin against this treaty".

Curses

As already pointed out, curses formed the very backbone of Assyrian treaties in that they defined and sanctioned the punitive measures resorted to after a possible treaty violation. Accordingly, they were certainly included in every treaty. A curse section is actually attested in every text of the present corpus, except for two texts, both of which are very fragmentary, and it always follows right after the stipulations (and the violation clause).

Structurally, the curse section forms an independent unit within the treaty which has nothing to do with the preceding stipulation section. It is comparable to the penalty section of legal documents, which is likewise preceded by a violation clause and forms a self-contained unit within the document. Taking the curses as 'apodoses' to the stipulations, as done in some earlier editions, is out of the question even in cases where the violation clause has been omitted as unnecessary. Simply note the 3rd person object suffixes in the curse section of Text 4, and compare the 2nd person prefixes in the preceding šumma clauses.

An analysis of the curses occurring in the texts falls out of the scope of the present edition. It will suffice to call attention to the fact that they fall into two classes, 1) curses involving individual witness gods as agents of punishment and destruction, and 2) curses effected collectively by all the treaty gods. The former are referred to in this book as 'traditional curses': because they largely consist of curses taken over from the Mesopotamian literary tradition and attested in an identical or nearly identical form in several other texts and text genres. The latter are referred to as 'ceremonial curses', because they involve parables or references to symbolic acts actually carried out during the conclusion of the treaties.

Ceremonial curses are attested only in two texts (nos. 2 and 6). In the latter, they make up a long section concluding the treaty; in the former, they occupy a position right at the beginning, between the divine witnesses and the treaty articles. In either case, they occupy a very prominent position and are of such a nature as to have a clearly calculated effect: the other party was to witness and to have vividly and undeletably impressed on his mind the consequences a violation of the treaty would entail.

Vow

Inserted between the two long curse sections in Text 6 (lines 414- 493 and 518-663) is a short section containing a solemn vow in the first person plural to abide by the terms of the treaty. It is phrased according to the classic Assyrian formula for a promissory oath, being headed by an invocation of gods already attested in the Old Assyrian period, and was certainly to be pronounced by the other contracting party. The individual clauses of the vow are couched as the subjunctive šumma sentences (see above, p. XL) recapitulating the central points of the treaty. An affirmatory sentence resembling ones found in contemporary legal documents closes the vow.

A similar section is attested in two other texts of the present corpus only (nos. 4 and 9), and in both of these it seems to have constituted the entire text of the treaty proper. On the other hand, it seems to have been missing in at least nos. 1, 5 and 8. It would thus seem that the vow was an alternative way of phrasing treaties unilaterally accepted by the other party, and that the presence of a vow and treaty stipulations in the same text was an exception rather than a rule.

Colophon and Date

Only two texts of the corpus end in a colophon specifying their purpose and date. However, it is quite possible that many more texts originally contained this entry, since it is certainly missing in only one text (the Zakutu treaty), which also contains many other atypical features.

15 E.g., adê ša šarri bēlīja "the treaty of the king, my lord", ABL 350 r.4; adê ša abika u ... adêka "the treaty of your father and your own treaty", ABL 1217:4f // CT 53 17:4 // CT 53 938:5; adê ša šarri "the king's treaty" LAS 247:22 and often in contemporary legal documents.

16 E. Reiner, ANET³ (1969 ) p. 544a, correctly translates "(the treaty) which he has made..."

17 See A. George, "Sennacherib and the Tablet of Destinies," Iraq 48 (1986) 133ff, especially p. 141.

18 See LAS II (1983), p. 5.

19 Instead of the D-stem stative tummātunu occurring in Text 2, line 384 of Text 6 has the corresponding Neo-Assyrian G-stem form ta"ākunu (from *tam'ākunu) "you are sworn". The forms are semantically free variants, and corresponding variation is also attested in exorcistic adjuration formulae, e.g. nīš DN lu tāmāta // [nīš] DN tāmāta, W. Farber, Beschwörungsrituale an Ištar und Dumuzi (Wiesbaden 1977). pp. 231 and 232.

20 A historical introduction appears to have been an integral element of most 2nd millennium treaties even though it was occasionally omissible in these too, see e.g. J. Nougayrol, Le palais royal d'Ugarit IV (1956), p. 153f (no. 61).

21 E. Reiner, ANET³ (1969) 535ff; R. Borger, TUAT 1/2 (1983), 161ff; K. Watanabe, BaM Bh 3 (1987), 147ff; correctly D.J. Wiseman, VTE (1958), 34ff and Grayson, JCS 39 (1987) 134ff.

Simo Parpola

Simo Parpola, 'Structural and Formal Analysis', Neo-Assyrian Treaties and Loyalty Oaths, SAA 2. Original publication: Helsinki, Helsinki University Press, 1988; online contents: SAAo/SAA02 Project, a sub-project of MOCCI, 2020 [http://oracc.org/saao/saa02/thetreatycorpus/structuralandformalanalysis/]