Arabian Epigraphic Notes http://www.arabianepigraphicnotes.org ISSN: 2451-8875 E-mail alerts: To be notiied by e-mail when a new article is published, write “subscribe” to editor@arabianepigraphicnotes.org. Twitter: Subscribe to the Journal on Twitter for updates: @AENJournal. Terms of usage: This work is licensed under a Creative Commons AttributionNoDerivatives 4.0 International License. To view a copy of this license visit http://creativecommons.org/licenses/by-nd/4.0/. © the author. A Publication of the Leiden Center for the Study of Ancient Arabia http://www.hum.leiden.edu/leicensaa/ Prepositional Phrases in the Dadanitic Inscriptions Johan Lundberg Leiden University Arabian Epigraphic Notes 1 (2015): 123‒138. Published online: 26 November 2015. Link to this article: http://hdl.handle.net/1887/36473 Arabian Epigraphic Notes 1 (2015): 123‒138 Prepositional Phrases in the Dadanitic Inscriptions Johan Lundberg (Leiden University)* Abstract This article is concerned with the use and meaning of ten diferent prepositions attested in the corpus of Dadanitic inscriptions. Compared with previous overviews of the prepositional system, the article provides a more complete picture of the various semantic functions exhibited by these prepositions. It also discusses the impact of formulaic language on the semantic scope of individual preposition as well as instances where diferent prepositions have the same semantic function. It also compares the use of these prepositions with cognates in other ancient North-Arabian corpora. In addition to this, it contains some new interpretations and translations. Keywords: Ancient North Arabian; Dadanitic; Prepositions 1 Introduction This article concerns the prepositional system exhibited in the Ancient North Arabian inscriptions from the ancient oasis Dadan (modern-day ʾal-ʿUlāʾ in north-western Saudi Arabia). The term Dadanitic refers to inscriptions made in the local script of the oasis. These inscriptions were previously categorised as either Dedanite or Lihyanite. These terms could refer to successive dynasties in the oasis. However, the term Lihyanite is consistently used as an ethnonym. Dadanite, on the other hand, is also used with reference to the place. It is, therefore, possible that the terms have the same referent (Scagliarini 1995). Macdonald has argued in favour of the term Dadanitic because the texts in both groups belong to the same palaeographic and linguistic continuum and because of Sima's arguments in favour of the spelling Dadan instead of Dedan.1 For these same reasons the term Dadanitic will be used in this paper. * The writing of this article was greatly facilitated by access to the OCIANA database for which I am very grateful to Michael C. A. Macdonald. I also want to thank Ahmad Al-Jallad, Fokelien Kootstra, Chiara Della Puppa, Hekmat Dirbas, and Marijn van Putten for stimulating conversations about Dadanitic and their valuable comments during seminars and informal discussion. This paper has also beneitted from comments by Alessia Prioletta, Michael C. A. Macdonald, Peter Stein, and an anonymous reviewer. All errors remain my own. All ASA sigla can be found on the DASI (Digital Archive for the Study of pre-Islamic Arabian Inscriptions) website (http://dasi.humnet. unipi.it). 1 Macdonald 2004: 490‒492; for further discussions about the palaeographic development see Macdonald 2000: 33; for arguments in favour of the spelling Dadan see Sima 2000: 42‒46, and Macdonald 2000: n. 1. The vocalisation primarily rests on the spelling in two cuneiform texts from the neo-babylonian period related to Nabonid (URU da-da-nu and URU da-da-na; cf. Beaulieu 1989: 150‒151, 167) and the spelling Δαδαν in the Septuagint (cf. Gen 10:7). 123 PREPOSITIONAL PHRASES IN DADANITIC The Dadanitic corpus can be subdivided into several diferent categories based on content/formula or on the method of carving. As for the latter it is noteworthy that some inscriptions are written in relief with a formal hand, most of which have a commemorative function. The use of a stone mason makes these inscriptions unique within the ANA corpora. But it is also noteworthy that not all commemorative inscriptions were written in relief. The main part of the article contains a survey of ten diferent prepositions and the appendix contains a glossary of words with a new interpretation. 1.1 Aim and purpose The main aim of this article is to map the prepositional system as it is used in the corpus of the Dadanitic inscriptions. The following questions will be considered: 1. What prepositions are used in the Dadanitic corpus? (a) Do these inscriptions contain the same prepositions as other neighbouring ANA corpora (e.g. Safaitic)? 2. What was the semantic scope of these prepositions and what functions did they have? (a) How does the formulaic language afect their semantic scope? (b) Do these prepositions cover diferent and distinct semantic ields or are there overlaps so that some can be used interchangeably? 3. How many prepositional objects can a preposition govern? 1.2 Prepositions within a uniform corpus: the scope and limitations of this study Before the main part of this study some remarks about the content of these inscriptions are in order. First, many inscriptions are irrelevant because they do not contain any prepositions. Secondly, many clauses lack prepositional phrases because they do not contain peripheral arguments. Thirdly, even those inscriptions that contain prepositions exhibit little variation of topics and syntactic constructions. Fourthly, many prepositions are primarily found in formulaic expressions (e.g. in ẓll-inscriptions). It is reasonable to assume that the formulaic character of these inscriptions has inluenced the semantic scope exhibited by diferent prepositions. The use of ʿly and bʿd in ẓll-inscriptions nicely illustrates this point. The latter is more often used to mark benefactive phrases. But ʿly is also used in similar clauses pointing to a semantic overlap. If ẓll-inscriptions had not been so numerous, the standardised use of bʿd could easily have obscured this polysemy. These observations are very important and serve as a reminder of the incompleteness of the material at hand. With this in mind we turn to the inscriptional evidence. 124 J. LUNDBERG 2 Dadanitic prepositions Both Macdonald and Farès-Drappeau have written brief surveys of the prepositions used in Dadanitic.2 But their treatments are limited in that they only give a few examples, list English and French counterparts, and occasionally provide etymologies. The present article attempts to ill this gap. 2.1 ʿly/ʿl This preposition is written either as ʿly or ʿl. The longer form occurs in most examples, making it diicult to determine if there is a distributional pattern. The shorter form ʿl is primarily found with clitics (e.g. ex. [5]) but it is also used once in the construction ʿl ḏ-kn (ex. [2]). It is therefore possible that ʿl and ʿly represent two distinct forms, the shorter being used with clitic pronouns and the long in most other constructions. However, more evidence would be needed to prove this. Moreover, the use of ʿly in similar phrases (e.g. ʿly ḏ-kn in AH 069 and 075) could indicate that ʿl ḏ-kn was a scribal error. If so, the diferent forms are probably due to orthographic conventions. y, whether it represented a diphthong or a long vowel, is never written word-internally in Dadanitic orthography. In terms of function ʿly was used as a marker of locative and benefactive phrases, with the meanings ‘on’ and ‘for the sake of’. The locative function is attested three times: [1]: l-ntnbʿl bn wny hn qbr ḏh ḥm ʿly ymn w ʿly šm mn ṯrq(h) (JSLih 081) ‘This tomb belongs to Ntnbʿl son of Wny. It is protected on the north and on the south against thieves.3 ’ Two observations are noteworthy about the phrases in ex. [1]. To begin with, the preposition is repeated before both ymn and šm. Secondly, ʿly could be interpreted as an adversative if the two phrases were not followed by mn ṯrq(h). However, together with the adversative phrase, a locative interpretation of ʿly seems more plausible. The two phrases emphasise that the tomb is thoroughly protected. Farès-Drappeau has suggested that ʿly primarily functioned as a locative preposition.4 The locative meaning is certainly closer to that of the root ʿly but the size of the sample makes it impossible to determine whether ʿly had any primary function in the local dialect. Moreover, the preposition is used as 2 Macdonald 2004: 519‒520; Farès-Drappeau 2005: 72‒74. & Reed 1970: 124, follow Jaussen & Savignac 1909-1922: 450‒451, Pl. LXXXV, and reconstruct šm[ʾl] at the end of line ive, i.e. one of the words meaning ‘left’ in CAr (Lane, 1601). For the present argument it is not necessary to determine whether the text was broken or not. But some comments about the state of the inscription are still in order. It is possible that the text is broken at the end of line ive because some of the previous lines are longer. But the last word on the previous line (ymn) ends at the same point as the m in šm[ʾl]. Moreover, the photograph does not contain a trace of more letters to the left of the m in šm. Lastly, it might not be not necessary to reconstruct [ʾl] at the end of line ive since CAr also contains the term šʾm, meaning either ‘left’ or ‘south’ (Lane, 1490). In comparison with other Central Semitic languages Hebrew and Aramaic has śmʾl (DNWSI, 1159–60) while šʾm is attested in Sabaic (SbD, 130) and Qatabanic (MuB 659, 9). 4 Farès-Drappeau: 74: “La premier sens de la préposition ʿly est «sur, au-dessus»” 3 Winnett 125 PREPOSITIONAL PHRASES IN DADANITIC much as a benefactive within the present corpus because of its occurrence in the formula of ẓll-inscriptions. This function suggests that there was a semantic overlap between ʿly and bʿd:5 [2]: llt ʾẓll h-ẓll l-ḏġbt ʿl ḏ-kn l-hm b-bdr f-rḍ-h (U 073) ‘Llt performed the ẓll-ceremony for Ḏġbt on behalf of that which belongs to them at Bdr and so favour him!’ [3]: mqḥ s¹lḥ ḏ-(ġ)bt ʾẓll b-khl ʿly m kn l-h b-ḏṯʿʿl mn dṯʾ w-ḫrf f-rḍ-h w-ʾṯb-h (U 059) ‘Mqḥ the priest of Ḏġbt performed (the ẓll-ceremony) at Khl on account of that which belongs to him in Ḏṯʿʿl from the spring harvest and autumn harvest, and so favour him and reward him.’ It is noteworthy that ʿly governs the same kind of objects as bʿd and that both are used together with the same kind of locative phrases. It is possible that a diachronic explanation could account for the use of ʿly and bʿd but the absence of dating formulae in many inscriptions and the lack of a ixed chronology makes it impossible to substantiate such a theory. Leaving this question to the side, we turn to syntactic matters: As for prepositional objects, ʿly governs both individual nouns and headless relative clauses. It mostly takes one object but there is one instance where it governs two coordinated objects: [4]: ---- ʿ ʿzy ---- ḏ ---- h-ẓll ʿly ḏṯʾ-h w nḫl-h f rḍ-h w ʾ{ḫ}rt-h (AH 107) ‘----ʿ ʾzy ---- ḏ ---- the ẓll-ceremony for his spring harvest and his palm garden and so favour him and his descendants.’ In addition to the benefactive use of ʿly, there is one example with a semantic function that is neither benefactive nor malefactive but somewhere inbetween the two: [5]: whblh bn zdqny w lmy bn nfyh wdyw nfs¹ mr bn ḥwt m{h} ʾḫḏ ʿl-hmy ḫrg (JSLih 077, 1–3)6 ‘Whblh son of Zdqny and Lmy son of Nfyh dedicated? the funeral chamber of Mr son of Ḥwt because of an obligation he had brought upon them.’ Lastly, the corpus does not contain any examples where ʿly means ‘against’ rather than ‘on’ or ‘on behalf of’.7 Neither does the corpus contain verbs of 5 Cf. Sima 1999: 101; Other examples of this type occur in U 050, U 071, U 059, U 087, U 125, U 126, AH 069, AH 071, AH 075, AH 079, AH 089, AH 107, AH 010. 6 The verb ḫrg usually means ‘to go’ or ‘to issue’ in CAr and there is also a noun from the same root with the meaning ‘disbursement’ or ‘expenditure’ (Lane, 718–19). The root is also attested twice in Central Middle Sabaic with the meaning to ‘sue’ or ‘bring a lawsuit against s.o’: w-ʾs¹d-hw ʾs¹d ḫrg-hw b-ʿbr mrʾ-hm ‘and his men are the men who sued him before their lord’ (CIH 398, 7); w-tws³ʿ-ḏ-ḫrg-hw b-ʿbr mrʾ-hw ‘and to take his due of the one who sued him before his lord.’ (Ja 646, 6–7). In the present context the combination of this verb with ʾḫḏ and ʿly probably means something along the lines of ‘to bring a claim/obligation upon s.o’. 7 Cf. the Safaitic use of ʿly in curse formulae (Al-Jallad 2015: 149). 126 J. LUNDBERG grief which normally have ʿl with their objects in Safaitic and Hismaic.8 However, the absence of these functions should not be taken as an indication that Dadanitic used other preposition for these functions. In light of the current evidence it is just as likely that the lack of comparable phrases and expressions accounts for the absence of examples. 2.2 bʿd The preposition bʿd is used frequently in ẓll-inscriptions, governing both nouns or headless relative clauses. In these constructions it has a benefactive function, like ʿly, indicating that it is a compound preposition consisting of bi- and ʿad.9 As was noted in the section above, it is unclear why both bʿd and ʿly were used but their relative frequency indicates that bʿd was part of the standard formula: [6]: ḥmyh bnt nẓrh ʾft h-ẓll ḏh l-ḏġbt b-khl bʿd m-l-h f-rḍ-h w-s¹ʿd-h (U 005) ‘Ḥmyh daughter of Nẓrh accomplished this ẓll-ceremony for Ḏġbt at Khl for the sake of that which was hers and so favour her and help her.’ [7]: ʿbdʾs¹ bn ʿgry ʾẓll h-ẓll b-khl l-ḏġb(t) bʿd nḫl-h b-bdr f-rḍ-h w-ʾḫrt-h (U 011) ‘ʿbdʾs¹ son of ʿgry performed the ẓll-ceremony at Khl for Ḏġbt for the sake of his palm garden in Bdr and so favour him and his descendants.’ As for the number of objects governed, bʿd is often followed by one noun phrase but in a number of inscriptions it also governs two phrases coordinated by w: [8]: ʿf bn ʿyḏh ʾẓll h-ẓll nḏr bʿd dṯʾ-h w-nfs¹-h f-rḍ-h w-ʾṯb-h w-s¹ʿd-h w-ʾṯb-h (U 021)10 ‘ʿf son of ʿyḏh performed the ẓll-ceremony as a vow for the sake of his spring harvest and for himself and so favour him and reward him and help him and reward him.’ [9]: ʿbdʾs¹ bn ws¹ṭ ʾẓll l-ḏġbt b-khl bʿd nḫl-h w-dṯʾ-h b-bdr f-rḍ-h w-ʾḫrt-h (U 009; cf. U 058, U079bis, AH100) ‘ʿbdʾs¹ son of Ws¹ṭ performed (the ẓll-ceremony) for Ḏġbt at Khl for the sake of his palm garden and his spring harvest at Bdr and so favour him and his descendants.’ 8 Al-Jallad 2015: 148; King 1990: 48 (C.5); According to Kootstra forthcoming.b ʿly is not attested in Taymanitic inscriptions. 9 Macdonald 2004: 519‒520; Cf. Al-Jallad 2015: 147; Farès-Drappeau 2005: 74, rightly notes that a temporal bʿd meaning ‘after’ does not occur in Dadanitic (cf. CAr baʿda). A temporal bʿd is attested in Safaitic, Aramaic, and Sabaic. It is possible that ḫlf (‘after’), which is not attested in these languages, was used in the local dialect of Dadan instead of bʿd. But this cannot be deduced from the evidence seeing that ḫlf is only attested twice in Dadanitic (see 2.6 below). 10 Note that the verb ʾṯb is used twice in the prayer creating a parallelism, an uncommon feature in other prayers. 127 PREPOSITIONAL PHRASES IN DADANITIC Example [8] is especially noteworthy because bʿd is used to govern two diferent kinds of objects, i.e. both ‘his spring harvest’ (dṯʾ-h)11 and ‘himself’ (nfs¹-h).12 By way of contrast some inscriptions contain chains of objects, all of which are preceded by bʿd: [10]: ʿ{y}ḏ bn ḥr ʾẓll h-ẓll l-ḏġbt b-khl bʿd-h w bʿd ʾ---- b-bdr frḍ -h w ʾḫrt-h (U 102bis) ‘ʿ{y}ḏ son of Ḥr performed the ẓll-ceremony for Ḏġbt at Khl for the sake of him and for the sake of ʾ---- at Bdr and so favour him and {his} descendants.’ [11]: ʿyḏ bn ḥr b-khl ʾẓll h-ẓlln bʿd-h w-bʿd ʾb-h w-bʿd nḫl-h <l>-ḏġbt f-rḍ-h w-ʾḫrt-h w-s¹ʿd-h (U 034) ‘ʿyḏ son of Ḥr performed two ẓll-ceremonies at Khl for his own sake and for the sake of his father and for the sake of his palm garden to Ḏġbt and so favour him and his descendants and help him.’ The repetition of the preposition could indicate that it was optional in constructions with two objects but mandatory with more than two. The sample is too small, however, to determine whether the use of multiple prepositions is coincidental or if the choice was regulated by syntactic rules. As for the order of clause constituents it is relevant to note that bʿd-phrases normally occur at the end of a clause (after b-khl or l-ḏġbt). The formulaic nature of these clauses explains the consistent placement at the end of the clause while ex. [11], could suggest that there was some lexibility. 2.3 l The preposition l is used frequently in Dadanitic. Semantically it covers the same areas as counterparts in other Central Semitic languages, indirect object (to), benefactive (for), possession (of), and possibly temporal duration (for/during). When it marks indirect objects, it is mostly used in connection with ḏġbt but it also occurs with other nouns outside of ẓll-inscriptions:13 [12]: ʾmtyṯʿn bnt dd nḏ[r]t bʿd bnt-h qn bnt ḥṯl l-s¹lmn hm-ḏ nḏrt ʿl-h ʾm-h f rḍ-h w s¹ʿd -h JSLih 073 ‘ʾmtyṯʿn daughter of Dd vowed on behalf of her daughter Qn daughter of Ḥṯl to S¹lmn according to that which her mother vowed on her behalf, and so favour her and help her.’ 11 The noun dṯʾ typically refers to the period of the later rains in Safaitic (Al-Jallad 2015: 311; ?macdonald1992). In Sabaic dṯʾ can refer to both spring (Ja 2848 ad) and spring harvest (CIH 2). In these instances, dṯʾ often occurs together with ḫrf meaning either autumn or autumn harvest. In Akkadian the noun dīšu refers to both spring and spring pasture (CAD D, 164). A noun dšʾ is also attested in Ammonite meaning ‘grass’ or ‘hay’ (DNWSI, 262). 12 The use of bʿd with pronominal clitics and nouns like nfs¹ suggests that the preposition is benefactive rather than directional. Example [8] and [11], especially, make the interpretation “in the direction of” unlikely (contra Sima 1999: 99–105). 13 An extension of this function occurs in existential clauses: ḏ-kn l-h b-bdr ‘that which belongs to him in Bdr.’ (AH 077, 3–4). Note that Farès-Drappeau 2005: 73, refers to this as possession. 128 J. LUNDBERG Farès-Drappeau argues that l- can be used to express time:14 [13]: s¹nt ṯtn l-tlmy (JSLih 045) ‘the third year of Tlmy ’ [14]: f-ḫbr h-l-gbl ḏ l-ṯlt s¹nn (JSLih 071) ‘… for three years’ As for ex. [13], it is diicult to argue that l expresses duration.15 The phrase as a whole refers to a period of time but this is not caused by the semantics of l. The phrase in JSLih 071, on the other hand, could be classiied as temporal if the above reading is correct. But there are some interpretive diiculties. First, the reading is somewhat uncertain. The last lines of the inscription are considerably shorter than the preceding ones and it is not clear (from the photograph) whether this was the result of damage on the rock before or after the carving of the inscription. If something is missing between ṯlt and s¹nn, it is far from certain that l has a temporal function. Secondly, Beeston has argued that the language of this inscription is closer to classical Arabic than Dadanitic.16 If valid, his conclusion would be suicient to exclude JSLih 071 from the present corpus even if l is temporal. Another feature of this inscription is the use of mixed letter forms, i.e. the carver used monumental script as well as less formal letter shapes.17 Because of these reasons it is diicult to argue with conidence that there is evidence for a temporal emphl in Dadanitic inscriptions. Lastly, it should be noted that the Dadanitic corpus only includes a few examples of a lam actoris.18 This diferentiates Dadanitic from Safaitic, Taymanitic, and Hismaic where l/lm is used frequently to mark authorship or possession.19 2.4 b The preposition b has three functions in Dadanitic: to mark locative phrases (spatial and temporal), instrumental phrases, and authors.20 When b is used as a spatial locative it primarily occurs in ẓll-inscriptions. As a locative, b often means ‘at’ but in some cases it could also mean ‘in’ (e.g. b-h-mṣd). One thing that diferentiates the Dadanitic inscriptions from the Safaitic ones is the ab14 Farès-Drappeau 2005: 73; Cf. Al-Jallad 2015: 145, for this function in Safaitic. 2004: 520 has a similar example of l in a dating formula: s¹nt ḫms¹ l-hnʾs¹ bn tlmy mlk lḥyn ‘year ive of Hnʾs¹ son of Tlmy king of Lḥyn’ 16 Beeston et al. 2005: 107; Beeston also argued that l-ṯlt s¹nn in early Arabic means ‘in the third year’ rather than ‘for three years’. 17 Beeston et al. 2005: 108; Macdonald 2000: 52, classiies this inscription as Dadano-Arabic. 18 l-gwr s²ms¹ ‘by gwr son of s²ms¹ (AH 265; cf. AH 295); Winnett and Reed, 1970, 123, 228–29 contains two possible examples: z l-dln l-rm, ‘This is for Dln by rm’ (nr. 3); and l-šnẖ ‘by Šnẖ’ (nr. 4). The irst inscription might not be Dadanitic because z is not used as a demonstrative in Dadanitc. The reading of the second one does not match the tracing or the photograph so it is uncertain. 19 Al-Jallad 2015: 145; Macdonald 2004: 518‒519; Kootstra forthcoming.b: 46. 20 Farès-Drappeau 2005: 72‒73, mentions three functions: (1) instrument or means; (2) place; (3) in dating formulas. The third corresponds to the temporal locatives. 15 Macdonald 129 PREPOSITIONAL PHRASES IN DADANITIC sence of unmarked locative nouns as well as the prepositions ʿnd and f.21 Given the formulaic nature of the inscriptions the absence could be circumstantial, especially since very few examples require the pragmatic meanings expressed by ʿnd and f.22 Most prepositional objects fall into one of two categories. The irst consists of phrases containing the two nouns khl (e.g. U 56) and mṣd (‘sanctuary’ e.g. AH 224, 244).23 For the interpretation of khl it is worth noting that these nouns never appear together and that they ill the same slot in the standard formula:24 [15]: w ʾẓlw b-h-mṣd ẓll h-[nq] l-ḏġbt (AH 197:6–7) ‘And they did the ẓll of the mountain for ḏġbt in the sanctuary.’ [16]: ns²l bn whblh ʾẓll h-ẓll b-khl l-ḏġbt bʿd ḏ-kn l-h b-bdr f-rḍ-h (AH 130) ‘Ns²l son of Whblh performed the ẓll-ceremony at Khl for Ḏġbt for the sake of that which belongs to him at Bdr, and so favour him.’ The distribution of khl and the lack of a preceding article indicates that it is a proper name, probably referring to a similar entity as mṣd. It is even possible that khl was the name of a sanctuary in Dadan where ẓll-ceremonies were performed.25 The second category of objects consist of the following terms: tḥfy, bnʾl, ṯr, bṯr, bdr, ḏʿmn, ḏʾm, blḥ, ḏtʿʿl, ms²hl, hmḏhb, ḏʾdn. All these words occur in phrases governed by bʿd and ʿly and they have so far been interpreted as place names. The absence of the article h- between b and these terms suggests that they referred to geographical locations. Another possibility is that they are calendrical terms. But the consistent placement directly after the noun/phrase governed by bʿd indicates that these locative phrases are part of the benefactive phrase, making a calendrical reading less likely. More importantly there are other calendrical terms found in dating formulae: [17]: s¹nt ḫms¹ b-rʾy ʿbdn hnʾs¹ (JSLih 072, 8–9) ‘year ive, at the rʾy of ʿbdn, Hnʾs¹’ The exact meaning of the phrase b-rʾy is at the present unknown and the meanings of most words that follow are uncertain (s¹lḥn, JSLih 068; ḏʾbs¹mwy, Nasif, 1988, 96; ḏʾs¹lʿn, AH 244; hrm, AH 219; ʿbdn, JSLih 072; [m]nʿy, JSLih 082; gltqs¹, JSLih 083; ḫmt, JSLih 085; hrʿ, as-Saʿīd 1420/2000, 3–14, no.1). The lack of an article before rʾy suggests that the noun is in the construct followed by a proper noun. It has been suggested that the following words are 21 Note, however, that the use of f to mark static location in Safaitic is rare and ʿnd is used only twice in the corpus of Safaitic texts included in An Outline of the Grammar of the Safaitic Inscriptions (Al-Jallad 2015: 150, 152‒153). 22 Cf. Al-Jallad 2015: 70–71, 245–26. The Safaitic corpus also uses the accusative to indicate the goal of travel. The Dadanitic texts neither mention travel nor journeys so it is impossible to determine whether the accusative, if it was still morphologically marked in Dadanitic, was used with this function. 23 The Dadanitic noun mṣd could be a cognate of Arm mṣd/mṣdʾ meaning ‘fortress’ or ‘stronghold’ (DTTML, 823). 24 The reading b-mṣd in AH 207 could be interpreted as b-[h]-mṣd or as elision of the article. 25 Note, however, that Robin 2003: 778, suggests that Khl was the ancient name of Al-ʿUdhayb. 130 J. LUNDBERG personal names. But this seems unlikely since most of them are not attested as names in the ANA and ASA corpora. Moreover, even if words such as ʿbdn, hrm and s¹lḥn are attested personal names, it is still unlikely that ḏʾbs¹mwy is a personal name. A more likely alternative is that they are calendrical terms referring to months, festivals, or astronomical phenomena.26 In either case, the use of b-rʾy X after s¹nt indicates that the preposition has a temporal function.27 The second function of b is to mark instrumental phrases:28 [18]: b-yd wt JSLih 106 ‘By the hand of Wt.’ Related to this function is the use of b to indicate the author of an inscription comparable to lam auctoris: [19]: b-ḏkrh wdd ḏ{h}k (AH 311)29 ‘By Ḏkrh son of Wdd son of Ḏ{h}k’ To summarise, the use of b in Dadanitic covers the same semantic spheres as other Central Semitic cognates even though the formulaic language of the Dadanitic material makes it diicult to determine the full semantic scope. 2.5 qbl The preposition qbl occurs three times in the Dadanitic corpus. Two attestations occur in dating formulae before the object rʾy and one is followed by ʾns¹ in a broken context: [20]: {s¹}nt ʿs²rn {w} tmn ṯlt ʾym qbl rʾy s¹lḥn (JSLih 068; cf. AH 244) ‘Year twenty-eight three days before the rʾy of s¹lḥn. ’ [21]: wʾl ʿbd s¹rmrʾ hʾ nṣb ---- h [l-]ʿtrġth qbl ʾns¹ ---- (AH 288) ‘Wʾl the servant of S¹rmrʾ, he set up a standing stone [to] ʿtrġth in presence of ʾns¹’ Contextual factors determine the meaning of the preposition. In the dating formula it is reasonable to assume that it has a temporal function because the formula has a similar structure as b-rʾy X, it is preceded by a temporal expression ({s¹}nt ʿs²rn {w} tmn ‘year twenty-eight’), and it modiies the temporal phrase ṯlt ʾym (‘three days’). The second example is more diicult to interpret because of the break at the end of the line. Two interpretations of ʾns¹ are possible. It could either 26 This interpretation was suggested by Kootstra and a more detailed discussion of this and other dating formulae will appear in her forthcoming article about Dadanitic dating formulae. 27 Moreover, rʾy is also preceded by qbl which is more narrow semantically. 28 Another possible example occurs in JSLih 70 (cf. n 29 below). 29 It is possible that bḏkrh is a personal name in which case the inscription would mean ‘Bḏkrh loves Ḏ{h}k’. But it is not uncommon that bn is left out in Dadanitic genealogies (cf. AH 157, U 038, U 078). 131 PREPOSITIONAL PHRASES IN DADANITIC mean ‘people’ or it could refer to someone called ʾns¹.30 In both instances the preposition would be spatial rather than temporal. However, it is also possible that the missing word(s) at the end of the line would lead to another interpretation. 2.6 ḫlf The preposition ḫlf only occurs twice in dating formulae, both times with a temporal function: [22]: s¹nt ʿs²r w ṯlṯ ymn ḫlf ṭʿn ḏ ---- l{ʿ}{b} / [t]lmy / bn / [l]ḏ{n} / ml{k} / {l}{ḥ}yn (AH 197) ‘Year thirteen, two days after the ṭʿn of ḏ ---- l{ʿ}{b} [T]lmy son of [L]ḏ{n} king of {L}{ḥ}yn’ The damage on the second line of this year formula slightly obscures the meaning of the sentence. The term ṭʿn could be the opposite of rʾy.31 If so, then this formula would be similar to qbl rʾy and the missing word would be a calendrical term. 2.7 mʿ The preposition mʿ (Ar. maʿa/maʿ), which has a comitative function, is attested four times in Dadanitic: [23]: ḏbn ʿmr bn mr{d} ʾgw h-ẓll ḏh l-ḏġbt ʿl ---- mʿ hn-ʾfklt b-bnʾl f rḍ-h w ʾḫr[t]-h w ʾṯb-h hnʾ bn ʿmr (U 038) ‘Ḏbn ʿmr son of Mrd organised this ẓll-ceremony for Ḏġbt on behalf of ---- together with? the priestess in Bnʾl and so favour him and his descen[dants] and reward him. Hnʾ son of ʿmr’ [24]: wny bn fs¹y tqṭ mʿ ḏ mʿly f rḍy-h w s¹ʿd-h w ʾḫrt-h (WR 16) ‘Wny son of Fs¹y wrote together with the one of the family of Mʿly and so may (the deity) favour him and help him and his descendants’ The reading of mʿ in the irst example is slightly uncertain because of the gap in the text. As a comitative phrase, it would indicate some kind of involvement of a priestess. The ʿl immediately before the break could introduce 30 According to Harding 1971: 79, the name ʾns¹ is attested in Dadanitic, Safaitic, and Taymanitic. It should be noted that the ʾ of ʾns¹ has an unusual shape, perhaps because of the length of the word divider in the line above. 31 Cf. Kootstra (forthcoming.a) for a more detailed discussion of this and similar formulae; The term ṭʿn also occurs in JSLih 077 with b: b-{ṭ}ʿn ṣd ḏ —. It has previously been interpreted as ‘at the departure of Ṣd’. But in light of Kootstra’s work it seems more likely that it means: ‘at the ṭʿn of the turning of ḏ…’ The broken word at the end could be a divine epithet. The term ṣd could mean ‘turning’ or ‘return’ (Lane, 1658). An alternative meaning could be ‘sign’ if it is a cognate of Akk ṣaddu (ṣādu) which is sometimes used together with planets, e.g. Jupiter. (CAD Ṣ: 56). The second example with ḫlf occurs in JSLih 070: ḫls¹ zdḫrg bn bl ḫld s¹nt ʿs²rn wts¹ʿ ʿs²rʾym ḫlf fḍg wbmmʾ ʿly mgh mn hḫls¹ s¹h mʾlh w …. The meaning of the last term fḍg is unclear but it could be a ‘month’ name (cf. Kootstra forthcoming.a). 132 J. LUNDBERG a benefactive construction indicating that the priestess at Bnʾl (hn-ʾfklt b-bnʾl) was a beneiciary of the ẓll ceremony. If so, it would be a variant of the standard formula with several benefactive phrases introduced by ʿl and coordinated with mʿ. Alternatively, the comitative phrase could indicate that the priestess was involved in the performance of the ritual. The departure from the standard formula could be a way of emphasising that the ceremony had followed the proper procedures. But it could also be a way of highlighting that the ceremony was performed by a priestess rather than a priest. In ex. [24] the comitative function indicates that the inscription was a collaboration between Wny and someone of a family called Mʿly. 2.8 mn The Dadanitic preposition mn occurs ca. 15 times. The form is always mn, perhaps suggesting that the n was followed by a vowel (mina), and it usually has the meaning ‘from’, indicating origin or source: [25]: mqḥ s¹lḥ ḏ{ġ}bt ʾẓll b-khl ʿly m-kn l-h b-ḏṯʿʿl mn dṯʾ w ḫrf f rḍ-h w ʾṯb-h (U 059) ‘Mqḥ priest of Ḏ{ġ}bt performed [the ẓll-ceremony] at Khl on behalf of that which belongs to him in Ḏṯʿʿl from the spring harvest and the autumn harvest, and so favour him and reward him. ’ [26]: llt bn ʿbny ʾẓll h-ẓll bʿd ḏ-kn l-h b-bdr mn nḫl{-h} f rḍ-h l-ḏġbt AH 077 ‘Llt son of ʿbny performed the ẓll-ceremony for the sake of that which belongs to him in Bdr from (his) the palm garden – and so favour him – to Ḏġbt.’ It is probably that mn, in these examples, precedes products that were used to perform a ẓll. The irst example could then indicate that produce from both the spring and the autumn harvest was used in the ceremony. This means that Mqh either performed two separate ceremonies but only made one commemorative inscription or that he ofered the fruits of both harvests at one time. The second inscription has a similar pattern but contains another object (nḫl; palm garden). The use of these, therefore, objects suggests that various kinds of harvests could be used to perform a ẓll.32 In one instance it is possible that mn expresses manner meaning ‘according to’: [27]: tmlk bnt hd{l} ʾẓlt l-ḏġ[b]t b-khl s¹tt ʿs²r mn s¹nt mt ʿl-h f rḍ-h w ʾḫrt-h ʿdb s¹nt ʿs²rn tlmy [mlk l]ḥyn AH 064 ‘Tmlk daughter of Hdl performed [the ẓll-ceremony] for Ḏġ[b]t at Khl sixteen (times) according to the custom of reverence on her behalf, and so favour her and her descendants forever. Year twenty of Tlmy [king of L]ḥyn.’ 32 It is also possible that the preposition has a partitive meaning here. If so, the inscription emphasises that the ritual was only performed on behalf a speciic harvest or palm garden belonging to the person who performed the ritual. 133 PREPOSITIONAL PHRASES IN DADANITIC This formula is only attested in this inscription. An interesting feature is the phrase s¹tt ʿs²r (sixteen). It is possible that Tmlk performed the ritual 16 times before she commissioned the commemorative inscription, indicating that the inscription was made as a complement to the ceremony. The interpretation of mn rests on the meaning of s¹nt which could mean either ‘year’ or ‘custom’. If ‘year’, it would be desirable to connect mn s¹nt with ʿdb s¹nt. The word ʿdb could perhaps be a compound preposition consisting of ʿd and b meaning ‘until’. The main problem with this interpretation is the placement of the prayer between the two prepositional phrases. If the second phrase was not added as an afterthought it is simpler to view ʿdb as a part of the prayer. If s¹nt means ‘custom’, then mt could be understood as a term for a ritual/bond where one person seeks favour on someone else’s behalf.33 Lastly two additional functions should be highlighted, the use of mn in reason clauses and in adversative expressions, both of which are also attested in Safaitic:34 [28]: bʿls¹mn ʾḥrm h-qrt mn mh trq-h mrʾt ʾl-bhny hn-ʾfklt ḏ---- (JSLih 064)35 ‘Bʿls¹mn protected the village because the woman of the palm tree, the priestess of … cast a spell on it.’ [29]: l-ntnbʿl bn wny hn qbr ḏh ḥm ʿly ymn ʿly šm mn ṯrq(h) (JSLih 081) ‘This tomb belongs to Ntnbʿl son of Wny. It is protected on the north and on the south against thieves.’ Regarding ex. [28] it is possible that mn has an adversative function: ‘Bʿls¹mn protected the village against the woman…’. The exact function depends on the interpretation of trq. If casting a spell/charm is something positive, then mn is probably used to express the reason for Bʿls¹mn’s protection. On the other hand, if it is something negative, it is more likely that Bʿls¹mn is protecting the village against the sorceress, in which case mn has an adversative function. To summarise, mn, has a similar semantic range as its cognate in other Central Semitic languages. In addition to these examples mn also occurs in a spatial phrase together with the preposition ʿdky (‘until’), discussed in the following section. 33 Cf. Lane, 2688, for similar meanings of mt. [28] and [29] are discussed in Farès-Drappeau 2005: 73, but she does not place them in a separate category. Both of these occur in Safaitic (Al-Jallad 2015: 150–151) 35 This interpretation of JSLih 064 is the product of discussions during an ANA seminar in Leiden during the spring semester 2015 with Ahmad Al-Jallad, Fokelien Kootstra, and Hekmat Dirbas. The term qrt is taken as a cognate of Arm qrt (city/town; DNWSI, 1037). For bhn meaning palm tree see KAZ, vol 1, 174. 34 Ex. 134 J. LUNDBERG 2.9 ʿdky The Dadanitic corpus contains one attestation of the preposition ʿdky (to, until): [30]: ʾbʾlf b[n] ḥyw kbr h-dʿt s²ʿt hnṣ w rb-hm ḥrmnḥr bn wḫyn kbry s²ʿt hnṣ ʾḫḏw h-mkn w h-mqʿd ḏh kll-h mn mʿn h-gbl hnʾʿly ʿdky mʿd h-gbl hnʾs¹{f}l f rḍ-hm s¹nt ḫms¹ b-rʾy ʿbdn hnʾs¹ (JSLih 072) ‘ʾbʾlf son of Ḥyw the kabir of the adviser of the party of Hnṣ and their lord Ḥrmnḥr son of Wḫyn the two kabirs of the party of Hnṣ took possession of the place and also of this sitting-place, its entirety from the assembly place of the upper border until the sanctuary of the lower border and so favour them. Year ive at the rʾy of ʿbdn, Hnʾs¹.’ The preposition is found in the second part of a phrase that speciies the boundaries of h-mkn w h-mqʿd which ʾbʾlf and Ḥrmnḥr took as their possession. Semantically it has a terminative function meaning ‘until, as far as’.36 The area subdued ranged from the assembly place of the northern border ‘until’ the sanctuary of the southern border.37 2.10 ldy The preposition ldy (cf. Ar laday-, ladā) has only been identiied in JSLih 077:38 [31]: whblh bn zdqny w lmy bn nfyh wdyw nfs¹ mr bn ḥwt m{h} ʾḫḏ ʿl-hmy ḫrg w h-dṯʾ ldy dṯʾ ḥmm b-ḏʾfʿ w l-ḏġbt ---- (JSLih 077, 1–4) ‘Whblh son of Zdqny and Lmy son of Nfyh dedicated? the funeral chamber of Mr son of Ḥwt because of an obligation he had brought upon them and (they dedicated) the spring harvest on account of a harvest he had ofered at ḏʾfʿ. And for ḏġbt … ’ The translation of this inscription is somewhat uncertain. If it contains the preposition ldy it could mean ‘on account of’. Formally ldy could also be a G ininitive of wdy where the initial radical has been assimilated. More attestation of ldy or a similar formula are needed to determine the validity of the above interpretation. 36 Cf. Macdonald 2004: 519. 2005: 73, gives the following translation: “depuis mʿn en haut de la montagne, jusqu’à mʿd en bas de la montagne”. There are two problems with this interpretation. First, the deinite article before gbl indicates that it is not in the construct state, suggesting that the word that follows is an attributive adjective. Secondly, it is more likely that gbl means border than mountain. That leaves mʿn and mʿd. The irst could be a cognate of Hb. mʿwn which occurs in the Dead Sea Scrolls with the meaning ‘abode’ or as a reference to the temple or a dwelling place (TWQ II, 728–30). There is also an Akkadian noun māʾunnu with the meaning ‘dwelling’ which, according to von Soden, is derived from Canaanite māʿōn (AHW II, 637). Moreover, there is also an example of mʿwn meaning ‘temple’ in Punic (DNWSI, 668) and Arm has the word mʿwn (CAL). The noun mʿd could be a cognate of the Hb. mwʿd meaning “meeting place” (HALOT, 557–58) from the root yʿd which is waʿada in Arabic (cf. also mawʿid and miʿād in Lane, 2953). Another, less likely interpretation occurs in Zwettler 2000: 227‒239, where he cautiously suggests that mʿn could refer to Mineans and mʿd to the Arabic tribe Maʿaad. 38 Macdonald 2004: 520. 37 Farès-Drappeau 135 PREPOSITIONAL PHRASES IN DADANITIC 3 Conclusion This survey shows that prepositions in the Dadanitic inscriptions have similar functions as their counterparts in other Central Semitic languages. But the diference in formulae also makes it diicult to make comparisons with texts in other ANA scripts. The main diiculty concerns static location and goal of travel. Safaitic can use the accusative with the meaning ‘in’ or to indicate the goal of travel. Dadanitic locative phrases are preceded by b but in examples like b-h-mṣd it could mean either ‘at’ or ‘in’. In addition, there are no clauses in the Dadanitic corpus referring to the goal of travel. Appendix: Glossary of terms with a new interpretation For further discussion of these terms see the discussions in the footnotes notes in parentheses. ʾḫḏ: bhny: ḫrg: mʿd: mʿn: mṣd: mt: qrt: rqy: claim or obligation (n. 6) a kind of palm tree (n. 35) to sue, bring a lawsuit against s.o (n. 6) meeting place, assembly point (n. 37) sanctuary, temple, dwelling place (n. 37) sanctuary (n. 23) custom (n. 33) village (n. 35) to cast a spell (trq G preix conjugation 3fs in n. 35) Address for Correspondence: johan.m.v.lundberg@gmail.com 136 J. LUNDBERG Sigla Dictionaries CAD Gelb & Civil (1956-) CAL Comprehensive Aramic Lexicon. http://cal1.cn.huc.edu DNWSI Hoftijzer & Jongeling (1995) DTTML Jastrow (1886-1990) HALOT Koehler et al. (1995-2000) KAZ Kazimirski (1860) Lane Lane (1863-1893) SbD Beeston et al. (1982) TWQ II Fabry et al. (2013) Inscriptions AH Dadanitic inscriptions in Abū L-Ḥasan (1997; 2002) JSLih Dadanitic inscriptions in Jaussen & Savignac (1909-1922) U Dadanitic inscriptions in Sima (1999) References Abū L-Ḥasan, A. 1997. Alī: Qirā’a li-kitābāt liḥyānāiya min ǧabal ʿIkma biminṭaqat al-ʿUlā, Riyadh: Maktabat al-Malik Fahd al-Waṭaniyya. ——— 2002. Nuqūš Liḥyāniyyah min Minṭaqat al-ʿUlā, Riyadh: Ministry of information, Department of antiquity and museums. Al-Jallad, A. 2015. An Outline of the Grammar of the Safaitic Inscriptions, (Studies in Semitic Languages and Linguistics 80), Leiden & Boston: Brill. Beaulieu, P. 1989. The Reign of Nabonidus, King of Babylon, 556–539 B.C., Yale Yale Near Eastern researches 10, New Haven: Yale University Press. Beeston, A.F.L., Ghul, M.A., Müller, W.W., & Ryckmans, J. 1982. Sabaic Dictionary (English-French-Arabic), Louvain-la-Neuve: Peeters. Beeston, A.F.L., Winnett, F.V., al-Ghul, M.A., & Ryckmans, J. 2005. The inscription Jaussen-Savignac [Lihyanite] 71, in: Beeston at the Arabian Seminar and other Papers. (Supplement to volume 36 of the Proceedings of the Seminar for Arabian Studies), M.C.A. Macdonald & C.S. Phillips, eds., Oxford: Archaeopress, pp. 107‒110. Fabry, H., Dahmen, U., & Brooke, G.J. 2013. Theologisches Wörterbuch zu den Qumrantexten, vol. II, Stuttgart: Kohlhammer. Farès-Drappeau, S. 2005. Dédan et Lihyan: histoire des Arabes aux conins des pouvoirs perse et hellénistique, IVe–IIe s. avant l’ère chrétienne, (Travaux de la Maison de l’ Orient 42), Lyon: Maison de l’ Orient et de la Méditerranée. Gelb, I.J. & Civil, M. (eds.) 1956-. The Assyrian dictionary of the Oriental institute of the University of Chicago, Chicago: Oriental Institute. Harding, G.L. 1971. An Index and Concordance of Pre-Islamic Arabian Names and Inscriptions, (Near and Middle East 8), Toronto: Toronto University Press. 137 PREPOSITIONAL PHRASES IN DADANITIC Hoftijzer, J. & Jongeling, K. 1995. Dictionary of the North-West Semitic Inscriptions I-II, with Appendices by R.C. Stein, A. Mosak Moshave and B. Porten, (Handbuch der Orientalistik. I. Der Nahe und Mittlere Osten 21), Leiden: Brill. Jastrow, M. 1886-1990. A dictionary of the Targumim, the Talmud Babli and Yerushalmi, and the Midrashic literature, London: Trübner. Jaussen, A. & Savignac, M.R. 1909-1922. Mission archéologique en Arabie, vol. 1-5, Paris: Leroux/Geuthner. Kazimirski, A. 1860. Dictionnaire Arabe-Français, contenant toutes les racines de la langue arabe, leurs dérivés, tant dans l'idiome vulgaire que dans l'idiome littéral, ainsi que les dialectes d'Alger et de Maroc, 2 vols, Paris: Maissonneuve. King, G. 1990. Early North Arabian Ḥ ismaic. A Preliminary Description Based on a New Corpus of Inscriptions from the Ḥ ismā Desert of Southern Jordan and Published material, Ph.D. thesis, School of Oriental and African Studies. Koehler, L., Baumgartners, W., Richardson, M.E.J., & Stamm, J.J. 1995-2000. Hebrew and Aramaic Lexicon of the Old Testament I–V, Leiden: Brill. Kootstra, F. forthcoming.a. The Dadanitic dating formulae. ——— forthcoming.b. The language of the Taymanitic inscriptions and its classiication. Lane, E.W. 1863-1893. An Arabic-English Lexicon, London: Williams & Norgate. Macdonald, M.C.A. 2000. Relections on the linguistic map of pre-Islamic Arabia, AAE, 11: 28‒79. ——— 2004. Ancient North Arabian, in: The Cambridge Encyclopaedia of Ancient Languages, R.D. Woodard, ed., Cambridge: Cambridge University Press, pp. 488‒533. Robin, C. 2003. Review of: Sima, A. 1999. Die lihyanischen Inschriften von alʿUdayb (Saudi-Arabien), Bibliotheca Orientalis: Col. 773‒778. Scagliarini, F. 1995. La Chronologie Dédanite et Liḥyanite: Mise Au Point, in: Présence arabe dans le Croissant fertile avant l'Hégire. Actes de la Table ronde internationale (Paris, 13, novembre 1993, H. Lozachmeur, ed., Paris: Editions Recherche sur les Civilisations, pp. 199‒132. Sima, A. 1999. Die lihyanischen Inschriften von al-Udayb (Saudi-Arabien), (Epigraphische Forschungen auf der Arabischen Halbinsel 1), Leidorf: Rahden/Westf. ——— 2000. Zum antiken Namen Dedan, Biblische Notizen, 104: 42‒47. Winnett, F.V. & Reed, W.L. 1970. Ancient Records from North Arabia, Toronto: University of Toronto Press. Zwettler, M. 2000. Maʿaad in Late Ancient-Arabian Epigraphy and other pre-Islamic Sources, Wiener Zeitschrift für die Kunde des Morgenlandes, 90: 223‒309. 138