1 Animal Sacrifice
Animal sacrifice is the most common religious rite commemorated in the inscriptions and appears to have occupied a central role in the ritual life of the nomads.1 G. Ryckmans2 produced an article-length study of the phenomenon in Safaitic, but it is now quite outdated due to the discovery of many new texts. Macdonald also treats sacrifice in detail, covering themes and motivations for the performance of the ritual.3 The following section hopes to build on the progress made by Macdonald and Ryckmans by synthesizing the information from the available inscriptions to reconstruct the details and motivations of this practice.
There are 91 texts that mention sacrifice with the verb ḏbḥ, a large enough number for basic statistical analysis. Charts 4–7 give the breakdown of this data according to whether ḏbḥ texts occur with a prayer (Chart 4); sacrifices mentioning explicitly to which gods they are dedicated (Chart 5); gods in prayers following sacrifices (Chart 6); narrative themes associated with sacrifices (Chart 7).4
Animal sacrifice was practiced widely in the Ancient Near East and in Classical Antiquity, and seems to have continued into Late Antiquity in Arabia, eventually being incorporated into Islamic practice.5 In the Safaitic context, the rite was intended to motivate the gods to respond to their worshippers.6
The primary verb signifying this rite was ḏbḥ, which finds cognates across the Semitic language family.7 Inscriptions occasionally employ other verbs, perhaps referring to different types of sacrifices: ṣmy (KRS 68), hrq (AAEK 9).8 None of the inscriptions that attest this rite are accompanied by any information as to what was done with the carcass following the animal’s destruction. Occasional attestations of terms for “burnt offerings” (see § 6.2) suggest that immolation of the animal, or at least part of its remains, sometimes followed.
Chart 4
The occurrence of prayers with ḏbḥ-inscriptions
Chart 5
Gods to whom sacrifices are explicitly dedicated
The most basic commemoration of the sacrifice is accomplished with the verb ḏbḥ in the narrative section of the inscription, often, but not necessarily, followed by a prayer. Sometimes authors will mention to which specific deity an animal was slaughtered and others might note its purpose.
C 4410, 4409 = LP 894 = Mr.A 5
w ḏbḥ l-bʿlsmn
‘and he performed an animal sacrifice to Baʿal-Samīn’9
Whenever an animal is specified, it is a camel. If this is not simply an accident of attestation, it could suggest that the sacrifice of such a valuable and culturally important beast merited explicit mention in the texts.10 None of the inscriptions so far discovered indicate that the sacrifice required the officiation of a priest—if we take the texts at face value, the rite was performed by the worshipper directly as an individual act of devotion or involving a small number of people (see § 1.3).11
Chart 6
Gods mentioned in prayers following the sacrifice
Chart 7
Themes in narratives associated with sacrifice
Texts commemorating sacrifice are often followed by prayers: requests for security, successful raiding, the return of lost loved ones, or simply to improve the condition of the author.
AbWS 8
ḏbḥ l-gdḍf wqyt m-bʾs
‘he performed an animal sacrifice to Gadd-Ḍayf to be protected from misfortune’
MA.1
ḏbḥ {l-}rḍy w ġnm nqt
‘he performed an animal sacrifice to Roḍay so may he obtain a she-camel as spoil’
KRS 756
ḏbḥ f h śʿhqm slm [m] ʿl-h-ʾbl mtʿt l-mdbr
‘he performed an animal sacrifice so, O Shayʿhaqqawm, keep secure what is upon the camels, provisions for the inner desert’
AHS 10
ḏbḥ w ḫrṣ dd-h ʾsr f h gdḍf rwḥ
‘he performed an animal sacrifice and kept watch for his paternal uncle, who was taken captive so, O Gadd-Ḍayf, send relief’
RWQ 315
ḏbḥ w ḥḏr
‘he performed an animal sacrifice because he was weary’
1.1 Sacrifice before/during Dangerous Activities
Several inscriptions record the performance of sacrifice before embarking on a raid or journey.12 It seems that in these cases, the sacrifice was meant to ensure divine protection during activities involving danger and uncertainty. HH 1, for example, is by a man charged with keeping watch while his companions set off to raid. He takes his position and performs a sacrifice for the successful outcome of his raiding party’s enterprise.
HH 113
n{ṣ}b w ḏbḥ w ḥll w ḫrṣ {ʾ}śyʿ-h ḍbʾn f h lt w dśr [s][l]m w qb{l}{l} {f} {h} {l}t {r}w[ḥ] w {ġ}nmt
‘he erected a cult stone and performed an animal sacrifice, then encamped and kept watch for his companions who were on a raid, and so, O Allāt and Dusares, [grant] {security}, and {a reunion of loved ones} {and then} {O} {Allāt} [grant] {relief} and {booty}’
The author of AbSWS 11 makes a sacrifice while taking part in a rebellion, which has separated him from his companions. He invokes the god Shayʿhaqqawm to grant a safe reunion with loved ones.
AbSWS 11
ḏbḥ w mrd f tśwq ʾl-ʾśyʿ-h f h śʿhqm qbll ʾslm
‘he performed an animal sacrifice and took part in the rebellion and longed for his companions so, O Shayʿhaqqawm, may there be a safe reunion’
1.2 Seasonal Sacrifices
The performance of a sacrifice is sometimes connected with the transition of seasons and migration. In these cases, it seems the rite was intended to influence the gods to provide favorable meteorological and environmental conditions, in addition to protection during journeys. Some inscriptions, such as C 860, suggest that a sacrifice could be made to improve seasonal circumstances as well.
C 860
ḏbḥ f h gdʿwḏ slm w trd f rmd bqr snt ʾty ʾ-ṣf qr
‘he performed an animal sacrifice so, O Gadd-ʿAwīḏ, may he be secure and have bounty as the cattle froze the year the cold came during the early summer’
Al-Namārah.M 5814
ḏbḥ ⟨⟨⟩⟩ f ṣdy {l-}śʿhqm {w}-drbt mʿ-h f slm w dṯʾ snt {ʾ}{m}{r}{t} ṣbrṣ h-mdnt
‘he performed an animal sacrifice and called out to? Shayʿhaqqawm while Drbt was with him, so may he be secure while he spends the season of the later rains the year Severus took control of the province’
RWQ 307
ḏbḥ w dṯʾ
‘he performed an animal sacrifice and spent the season of the later rains (here)’
AWS 279
ḏbḥ w ʾśrq f h gdʿwḏ w h dśr slm w mgdt
‘he performed an animal sacrifice and then set off for the inner desert so, O Gadd-ʿAwīḏ and Dusares, may there be security and abundance’
1.3 Location of Sacrifice: The ṣamd
The inscriptions discussed above demonstrate that sacrifices did not need to be performed at temples.15 Rather, animals were slaughtered in the Ḥarrah itself. Yet, not all places were equal. Ababneh and Harahsheh document a fascinating site called Tell al-Rāhib, in northeastern Jordan (Figures 3 and 4).16 On the summit of this imposing hill was a stone installation, the eastern wall of which contained fifty-three inscriptions. Twenty-three of these commemorate animal sacrifices. Four inscriptions (1, 6, 7, 9) record that the sacrifice was performed for Gadd-Ḍayf, and the same god is mentioned alone in the prayer component of five further texts (2, 4, 5, 8, 10). The remaining sacrifice inscriptions record prayers to Allāt (9, 14, 15, 16, 17, 18, 24) or the pair, Allāt and Dusares (13, 19, 21, 23, 27). It is impossible to know if these were all produced at the same time, but such a high concentration of sacrifice texts in one place suggests that the site was of ritual importance. None of the inscriptions mention tribal affiliation so we cannot know if the site was used by more than one group. All of the texts except one (#27, which is carved in the Safaito-Hismaic variety) were inscribed in the fine variant of the Safaitic script. Nevertheless, the diversity of deities attested here suggests that the site was not associated with a single god—worshippers were free to sacrifice to the deity of their choice at this sacred place.
Ababneh and Harahsheh published a rough drawing of the site.17 The installation consists of a tailed stone circle, facing east. The cleared-out area in the center may have been the place where sacrifices were performed, but no altar or other cultic objects were identified by the authors.
This phenomenon of the “high place” finds parallels across Near Eastern traditions, most clearly the bāmâ “high place” of the Hebrew Bible, which was also associated with animal sacrifice and other religious rites.18 The large open-air Nabataean ritual site Jebel al-Maḏbaḥ “the mountain of the altar” provides an important, near-contemporary parallel. The sacrificial ritual site sits atop a 200 m ridge. It boasts a rock-cut altar on which sacrifices were likely performed. Basins near the site point towards an associated ablution or libation ritual.19 While all of these elements are present in the Safaitic inscriptions, and are associated with high places, we must await the discovery of another sacrificial site and the complete archaeological investigation of it to draw secure parallels. Unfortunately, shortly after Ababneh and Harahsheh’s visit to Tell al-Rāhib, the top of the hill was cleared to make room for a Jordanian military installation. Any hopes of determining what ritualistic role the stone circle and cairn described in their article may have played are now lost.
Figure 3
Tell al-Rāhib
Ababneh and Harahshe (2015)
Another site provides us with the name of the high place in Safaitic. Two authors—a father and son—record sacrificing a camel at a ṣmd, which corresponds to Classical Arabic ṣamdun “high ground.”20
KRS 818
l wdmʾl bn grmʾl bn nḫr bn ġrb h-ṣmd w ḏbḥ gml ʿl-h f slm yṯʿ m-śnʾ w ʿwr m ʿwr h-ʾsfr
‘By Wdmʾl son of Grmʾl son of Nḫr son of Ġrb, at the high place, and he sacrificed a camel upon it so, O Yayṯeʿ, grant security from enemies and blind whosoever effaces these writings.’
KRS 824
l qdmʾl bn wdmʾl bn grmʾl bn nḫr bn ġrb bn slm h-ṣmd ḏbḥ gml
‘By Qdmʾl son of Wdmʾl son of Grmʾl son of Nḫr son of Ġrb son of Slm, who sacrificed a camel at the high place.’
Figure 4
Summit of Tell al-Rāhib
Courtesy: Ali al-Manaser
“High” is relative and not all ṣmd’s were at such imposing places in the landscape as Tell al-Rāhib. Figure 5 is of a site in Wādī al-Khuḍarī, northeastern Jordan, that is identified as a ṣmd by an inscription. It is not on a very tall hill, but simply at an elevated place in the landscape relative to the wādī before it.
A vague memory of the sacrificial ṣmd may be found in a tale about the destruction of the ancient tribe of ʿĀd related in the Akhbār al-Yaman, a 9th c. work attributed to ʿAbīd b. Sharyah, who according to legend was the tutor of Muʿāwiyah I.21 In this story, Muʿāwiyah asks about poetry relating to the idols of ʿĀd. ʿAbīd relates the following line:22
ʾAḫbār, 326
la-nā ṣanamun yuqālu la-hū ṣamūdun
wa-yuqābilu-hū ṣudāʾu wa-l-buġāʾu
‘We have an idol called Ṣamūd, facing it is Ṣudāʾ and al-Buġāʾ.’
A poem ʿAbīd attributes to a Muslim man (qāla raǧulun mina l-muslimīn) mentions al-ṣamūd in a sacrificial context, although the exact sense of the line is difficult to ascertain on account of its difficult syntax.
ʾAḫbār, 330
fa-ttabaʿat mina l-maḥārībi ǧamḥun
mina ṣ-ṣamūdi ḏibḥatan limā ḏubiḥ
‘and they cleaved to the altars performing sacrifices, a wayward course of al-Ṣamūd’
The idol ṣamūd could reflect some tradition that preserves knowledge of high places associated with sacrifice, reimagined as the name of a god. The Muʿallaqah attributed to Ṭarafah may also preserve a sacral meaning of ṣmd, there applied to a noble temple.
Ṭarafah, Muʿallaqah
48 wa-ʾin yaltaqi l-ḥayyu l-ǧamīʿu tulāqi-nīʾilā ḏirwati l-bayti l-karīmi l-muṣammadi
‘and if the entire tribe assembles,23 you will find me at the summit of the noble temple on high ground (muṣammad)’
1.4 Thanksgiving?
While it appears that sacrifice was intended to increase the effectiveness of an invocation, that may not have been its only purpose. Several inscriptions record a sacrifice to a deity followed by a prayer to another, implying that the ritual was performed out of obligation, perhaps in thanksgiving, rather than to obtain some future favor.24 Indeed, such request may reflect the expectation of the author to be rewarded by the second deity for dutifully performing his religious obligations.
AH 9
ḏbḥ l-gdḍf f h lt slm w mgdt
‘he performed an animal sacrifice to Gadd-Ḍayf so, O Allāt, may he be secure and have abundance’
2 Erection of the nṣb Stone
At least one installation is associated with the performance of the animal sacrifice: the nṣb (and other variations thereof), conventionally translated as a “cult stone.” This corresponds to the standing stones and altars of the Bible, the maṣṣēbôt, and the aniconic betyls that were widely used as representations of the divine throughout the ancient Near East.25 Nabataean attests several forms of this word, nṣybʾ, nṣbtʾ and mṣbʾ, which all signify a cult stele of the deity, a betyl.26 Reflexes of this tradition stretch back to the Bronze Age, in the form of the sikkānum/skn “standing stones” of the Syrian city states Mari, Ugarit, and Emar.27 While clearly from a different root, Mettinger establishes a semantic connection between skn, which derives from the root “to dwell,” “to abide” and the Greek term for such cult stones, βαίτυλοι, which renders the Semitic phrase byt ʾl “dwelling of (the) god.”28 The various sources converge on the idea that these stone stelae were originally regarded as vessels that divinities could occupy for the duration of a ritual.
In the Bronze- and Iron-Age Syrian context, sacrifices made before the nṣb/skn stones belonged to a ritual of divine feasting. As S. Sanders argues, a supernatural presence in such stele was signified by appetite—sacrifices before them were part of a mortuary ritual where humans and the divine shared a meal.29
The Quran associates the nuṣub and ʾanṣāb with sacrifice explicitly, and prohibits the consumption of animals slaughtered before them. This appears to echo the dietary codes of the Pentateuch, which seek to curb participation in the rituals of other gods.30
Q 5:3
wa-mā ʾakala s-sabuʿu ʾillā mā ḏakkaytum wa-mā ḏubiḥa ʿalā n-nuṣub
‘(forbidden is) what the predator has eaten save for that which you have purified and (forbidden is) what was slaughtered upon the stone altars’
Hoyland connects this Quranic reference to sacrifices before stone idols mentioned in pre-Islamic poetry and in the South Arabian inscriptions.31
Ibn al-Kalbī describes the naṣb as an act of erecting an idol or stone representation of a divinity.32 Indeed, one frequently encounters references to blood libations before altar stones in the pre-Islamic poems, but nothing—as far as I know—suggests the existence of a ritual feast involving gods and men. Rather, the “blood at the nṣḅ” theme appears to be a stereotyped topos connected to oath making in this corpus.
Ṭarafah, °10
ʾinnī wa-ǧaddi-ka mā haǧawtu-ka
wa-l-ʾanṣābi yusfaḥu bayna-hunna damu
‘By your Gadd and the cult stones among which blood is poured, I have not disparaged you.’
The nṣb of the Safaitic inscriptions resembles Biblical and Arabian traditions as mentioned in the Quran and later sources rather than functioning as a funerary stele.33 The notion of a divine presence embodied in the nṣb is implied by the association of these stele with particular gods; for example, the author of SESP.S 1 erects the cult stone of the deity ʾṯʿ (nṣb ʾṯʿ) while RQ.A 9 records the erection of the cult stone of the Goddess of Dṯn (nṣb ʾlt dṯn).
When nṣb and ḏbḥ co-occur, nṣb comes first, suggesting that the structure had to be in place to fulfill the sacrificial rite, as illustrated in HH1 above, nṣb w ḏbḥ “he set up a cult stone and made a sacrifice.”34
JaS 100.1
l bgt bn ʿdy bn lśms w nṣb w ḏbḥ
‘By Bgt son of ʿdy son of Lśms and he set up a cult stone and performed an animal sacrifice.’
Several inscriptions record the erection of nṣb stones without mentioning a sacrifice or any other associated ritual or deity. It is possible that the simple mentioning the installation implied a sacrifice. In other words, nṣb and ḏbḥ could have been components of a single sacrifice ritual, which could be referred to in its entirety nṣb w ḏbḥ or perhaps by synecdoche, either w nṣb or w ḏbḥ. On the other hand, the comparative evidence provides several other possibilities, such as libations, the taking of oaths, or simply to commemorate an interaction with a divinity.35 Like the ḏbḥ inscriptions, nṣb ones can be followed by a prayer.
C 527
l mfny bn mśʿr h-nṣb f yṯʿ flṭ mn-sqm
‘This cult stone (was erected) by Mfny son of Mśʿr so, O Yayṯeʿ, deliver from illness.’
C 2019
l mrʾ bn mfn{y} bn mśʿr h-nṣb f h ʾ{ṯ}ʿ slm w rwḥ
‘This cult stone (was erected) by Mrʾ son of Mfny so, O ʾAyṯeʿ, [grant] security and relief.’
While most nṣb inscriptions are followed by an invocation to a deity, for whom it is natural to assume the cult stone was erected, some texts directly associate the nṣb with a particular deity yet invoke another in the prayer. Like the ḏbḥ texts, these types of inscriptions could be understood as commemorating, or be in thanksgiving of, a previous interaction with a deity or a divine boon.36 In RQ.A 9, for example, the author erects the cult stone for the Goddess of Dṯn but then calls upon Gadd-ʿAwīḏ to protect him and the writing. LP 237, which records the erection of the cult stone of hlt and then invokes lt, could also reflect the same phenomenon, but it is my opinion that both hlt and lt are phonetic variants of the same goddess.
RQ.A 9
w dṯʾ h-wrd w nṣb ʾlt dṯn f h gdʿwḏ slm w l-h h-ḫṭṭ
‘and he spent the season of the later rains in the lowland/watering place and erected a cult stone for the goddess of Dṯn so, O Gadd-ʿAwīḏ, may he be secure, and these carvings are his’
LP 23737
w nṣb h-lt f lt slm snt {g}lḥ h-ḏʾb ʾns w nqʾt l-ḏ yʿwr h-sfr
‘and he erected the cult stone for (h)Allāt so, O Allāt, may he be secure the year the wolf attacked mankind and may he who would efface this writing be thrown out (of the grave)’
2.1 The nṣb, nfs, and Mortuary Installations
There is a small amount of evidence to suggest that the nṣb played a role in funerary rituals, as it did nearly a thousand years earlier across ancient Syria. In 2009, P. Bikai published a collection of inscriptions from a cairn located on Jabal al-Muqallah in Wādī Rāǧil in eastern Jordan. He recorded 125 inscriptions, but not all seem to be associated with the structure or to have been produced at the same time or by the same social group. Among these, however, eight attest to a funerary ritual of a deceased man named ʾs (ʾAws). In BRCM 37.3, a man called Ġlb records his grief for several relatives and then terminates his inscription with w bny ʿl-ʾs “and he built upon ʾs.” This phrase is associated with other burials, most famously the burial cairn of Hāniʾ, and seems to signify the placing of stones upon the dead by the bereaved.38 Other members of this procession recorded their participation using the same phrase or by employing the typical grieving formula, wgm ʿl-ʾs “he grived for Aws.” The author of BRCM 14, however, inscribed the following:
BRCM 14
nṣb w l-h [h-] frs w bny ʿl-ʾs
‘he erected a cult stone and this (image) of the horseman is by him and he built upon ʾAws’
The mentioning of the cult stone could imply that the object played some role in the funerary ritual, although the laconic language of the texts prevents us from knowing what that entailed. Another collection of inscriptions further supports the possible mortuary role of the nṣb.
G. King documented a number of texts from a single site during her 1989 survey in north eastern Jordan that employ the verb ʾḫḏ in a funerary context. I have suggested that ʾḫḏ refers in some contexts to the reuse of funerary installations, that is, “taking” them for the dead.39 This is supported by the phrases such as KRS 1429: ʾḫḏ h-rgm l-ʾb-h, literally “he took the funerary cairn for his father,” but probably meaning “he interred his father in the funerary cairn.”
In King’s collection of texts, there can be no doubt that ʾḫḏ signifies some funerary activity. KRS 945 states: l gdy bn tmhm h-ḍfy w ʾḫḏ l-ṣʿd “By Gdy son of Tmhm the Ḍayfite and he took (it) for Ṣaʿd” (i.e. interred him in the cairn/funerary installation); KRS 939 states the same thing: l śḥtr bn sʿd bn sʿd bn skrn w ʾḫḏ l-ṣʿd “by Śḥtr son of Sʿd son of Sʿd son of Skrn and he took (it) for Ṣʿd.” A third associated inscription, however, mentions the nṣb.
KRS 929
l ʾnʿm bn hnʾ w ʾḫḏ w nṣb f h dśr slm
‘By ʾnʿm son of Hnʾ and he took (it for Ṣʿd) and erected a cult stone so, O Dusares, may he be secure.’
A fourth inscription, carved at a later point, confirms the hypothesis that this collection is funerary, as its author provides an elaborate description of his grief after finding the ʾṯr “trace” of a dead man named Ṣʿd.
KRS 941
l nʿmn bn ṣʿd bn ysmʿl w wgd ʾṯr ṣʿd f ngʿ w bʾs m ẓll w rġm m{n}{y} {ʿ}{n}{y} {w} {q}l ḫbl-h trḥ w h lt ʿwr ḏ yʿwr h-s{f}r
‘By Nʿmn son of Ṣʿd son of Ysmʿl and he found the trace of Ṣʿd and then grieved in pain; and those who remain (alive) despair and Fate strikes down sufferers (or: he was struck down by Fate, suffering); and he said aloud: “grief drove him mad,” so, O Allāt, blind him who would efface this writing.’
What further merits consideration is the presence of a drawing of a horse on the stone of KRS 929, similar to BCRM 14, which explicitly references the image. We shall return to this issue in § 6.
So then, these texts suggest that a nṣb could be a component of at least some mortuary complexes and the rituals involved with the burial and commemoration of the dead. Despite this, its function remains unknown—does the association with animal sacrifice elsewhere imply a mortuary feasting ritual, as in the Bronze Age Syrian context? While one cannot rule out such a possibility, there is yet no direct evidence that sacrifice and feasting were part of Safaitic funerary rituals. Only one Safaitic text mentions eating, and there is no explicit funerary or ritual context.40
The funerary dimension provides a new interpretive lens through which to view nṣb texts of the type: l PN + h-nṣb. The dative preposition may refer to the deceased rather than the agent. In other words, such texts could translate as “this nṣb is for PN” rather than “by PN is this nṣb,” bearing the name of the dead.
A variant of this installation is known, maṣṣeb and maṣṣebat. These could refer to the cult stone itself or the sacral/mortuary installation in its entirety. This form is directly compatible with the Biblical maṣṣēbâ < *manṣibat and Nabataean mṣbʾ.41
ZEGA 1
l gfft bn kn ḏ-ʾl ʾty w l-h mnṣb f h ʾlh slm
‘By Gfft son of Kn of the lineage of ʾty and this cult-stone was [set up] by him so, O Allāh, may he be secure.’
C 3097
l śqr bn hggt h-mṣbt w ngb
‘This cult-stone was (set up) by Śqr son of Ḥggt so may he be rewarded.’
KRS 3250
l zmhr bn kbr h-mṣbn
‘these two cult-stones were (set up) by Zmhr son of Kbr’
HNSD 196
l ʿbn bn wsmt h-mṣb
‘This cult-stone was (set up) by ʿbn son of Wsmt.’
I would, however, stop short of seeing the nṣb as a funerary stele as such. Rather, Safaitic uses the term nfs(t) = nepheš-stele, a widespread monument type in Levant and North Arabia, for the purpose of commemorating a dead person.42 We can be sure that the two are not synonymous as they have a different distribution in the corpus. While nṣb stones can be erected for specific deities, there are no examples of such in the case of the nfs. And while inscriptions of the type l PN h-nṣb f + prayer exist, this is a rather rare construction for the nfs(t). One clear example of this is attested which requests the protection of the nfs(t) itself from vandals rather than containing more general petitions to the gods for favor and bounty.43 Finally, unlike the nṣb, the erection of which is signified by its own denominal verb, the nfs is built, bny, just like the funerary cairn. Multiple nfs(t) (pl. ʾ(n)fs) can be constructed and associated with a single individual.44
Inscriptions commemorating the construction of the nfs(t) do not appear to be inscribed on the funerary monument itself but are rather carved on a nearby surface. A new set of Safaitic inscriptions discovered during the 2019 summer campaign of the Badia Survey project illustrate this clearly.45 These texts are inscribed upon a single stone (Figure 6) positioned on the perimeter of a stone enclosure in the area of Wādī al-Khuḍarī, northern Jordan.46 The first text contains the name of the dead man and indicates that the ʾfs “funerary monuments” belong to him. Two of his sons are present along with a friend—together they build the ʾfs and grieve for him. The final text contains a two-generation name; this text may have been carved at a later point and have nothing to do with the original mortuary procession.
Figure 6
The ʾfs-inscriptions, BESS19 3a–e
Photo: A. Al-Jallad
a) l śʿ bn ʿḏr bn wdm h-ʾfs‘These funerary monuments are for Śʿ son of ʿḏr son of Wdm.’b) l mġny bn śʿ bn ʿḏr w bny h-ʾfs w nʿś m-ḥrn‘By Mġny son of Śʿ son of ʿḏr and he built these funerary monuments, and he carried (him) upon the bier (to this place) from the Ḥawrān.’c) l ʾs bn śʿ bn ʿḏr w wgm ʿl-śʿ‘By ʾs son of Śʿ son of ʿḏr and he grieved for Śʿ.’d) l śmt bn śnʾ bn grm w wgm ʿl-ḥbb w bny h-ʾfs‘By Śmt son of Śnʾ son of Grm and he grieved for a beloved and built the funerary monuments.’e) l kḥsmn bn sʿd‘By Kḥsmn son of Ṣʿd.’
What these ʾfs may have been is unclear.47 The site itself where this stone was discovered contains a large cleared-out area and to its west a straight line of four small cairns running northeast to southwest (Figures 7 and 8).48
While we cannot be sure as to the antiquity of these structures, if they are coeval with the Safaitic inscription, then I would very cautiously hypothesize that these small tumuli are what the term nfs(t)/ʾ(n)fs signifies in the desert context. There appears to be nothing else in the area that could be regarded as “built,” but this may only mean that the site had been reused by later groups who had dismantled whatever structure the loved ones of Śʿ had constructed in his memory.49
In settled areas, the Safaitic nfs seems to resemble more the funerary stele as known from Northwest Semitic cultures. In 2017, H. Hayajneh published an interesting example of a Safaitic nepheš from the vicinity of Jerash, Jordan. It is an inscribed stone slab stating simply l-nhb h-nfs “this funerary monument is for Nhb” (Figure 9), resembling the inscribed Aramaic nepheš stones from the Ḥawrān.50
Figure 7
nfs-site in Wādī Khuḍarī, Northern Jordan
Photo: A. Al-Jallad
Figure 8
A possible nfs in the background with nfs-inscriptions in the foreground
So then, to round up our discussion: the Safaitic nṣb appears to be a stone medium that serves as a channel for the presence of deities during ritual acts. The evidence we have suggests that it was erected before sacrifices and as a part of certain mortuary installations, although ambiguous inscriptions could indicate that other ritual functions required it as well. The nṣb contrasts with the nfs(t), which seems to be purely funerary and associated with humans rather than divinities. The Safaitic nfs(t)—at least in the desert context—does not seem to be an inscribed stone or slab but rather some type of memorial installation, very possibly a small tumulus. Multiple nfs-monuments can be dedicated to a single deceased person.
Figure 9
North Arabian inscribed Nefesh Stele from Jerash (H. Hayajneh 2017)
Photo: @ArScAn, F. Villeneuve
3 The Ritual Shelter
The term str, and a variant, possibly a diminutive strt, are translated in the Safaitic dictionary neutrally as “shelter.”51 Like the nṣb texts, the str-inscriptions often contain the name of a dedicant or the person who constructed the installation. Besides the word’s etymology, there is nothing that clearly indicates what the function of the str was. While it is certainly possible that many of these str-texts simply record the building of temporary structures to shield from the elements, similar perhaps to the ẓlt or the mʿmr,52 some inscriptions describe the construction of the str before the performance of sacral activities, such as sacrifice and mourning. These would suggest that it may not necessarily refer to a habitation but rather to a ritual installation, perhaps comparable to the biblical Tabernacle.53 Before discussing these cases, let us look at how the inscriptions mentioning the str(t) break down.
The corpus presently contains 46 str texts. 44 of these deal with the construction of the str or name its owner (Chart 8). In two bny + str inscriptions, the act is followed by the performance of an animal sacrifice. And in BES15 871, this coincides with the seasonal encampment by water.55
AHS 19
bny h-str w ḏbḥ f h lt w dśr slm
‘He built the str-shelter and performed an animal sacrifice so, O Allāt and Dusares, may he be secure.’
BES15 871
ḥḍr f bny ʾ-str w ḏbḥ
‘He encamped by water and then built the str-shelter and performed an animal sacrifice.’
Indeed, ḥḍr is twice more associated with the construction of the str.
WH 3597
l zʿm h-strt b-ḥḍr
‘This str-shelter is for Zʿm during the encampment by water.’
HASI 22
ḥḍr ḏ ʾl tm w str
‘He of the lineage of Taym encamped by water and constructed the str-shelter.’
Finally, the str-shelter is associated with burials and ritual mourning. A text from the Cairn of the Mermaids, discussed above, couples the str with the nfs, again pointing towards the presence of ritual installations (cf. nṣb) at burials for the performance of mortuary rites.
BRCM 19.4
wgm ʿl-mʿyr w ʿl- ʾs w ʿl-{r}ḍwt w ʿl-sr ḏ- ʾl ----ʾ w bny ʾ-nfs w str
‘He grieved for Mʿyr and for ʾs and for Rḍwt and for Sr of the lineage of ----and he built the funerary monument and the str-shelter.’
The author of QZMJ 477 provides us with an illuminating description of the ritual function of the str-structure: qʿd b-h-str, which can be rendered as both “he sat inside the str-shelter” or “he halted at the str-shelter.” Indeed, another inscription KRS 2415 attests a similar construction—gls ʾ-rgm “he stopped at the funerary cairn.” The use of b- in the present inscription rather than simply taking at a direct object may therefore prefer the interpretation “he sat inside,” indicating that the str could be occupied while performing certain rituals.
QZMJ 477
qʿd b-h-str w wgm ʿl-ḫl-h ṭrq-h [h-]-qfy
‘He sat inside the str-shelter and grieved for his maternal uncle, whom the Qf-ite had struck down (or: Ṭrq the Qf-ite).’
As such, the str-shelter has a very similar distribution to the nṣb, occurring in both seasonal and mortuary contexts and preceding the performance of an animal sacrifice. I would therefore suggest that str can refer to a ritual installation, a temporary structure associated with the performance of sacral activities, finding a typological parallel in the biblical Tabernacle.
Literary sources from the early Islamic period provide further support for this hypothesis. Sean Anthony draws our attention to a passage from the Khūzistān Chronicle (~660 CE) that describes the rituals of the Arab conquerors.56 The “Ishmaelites”—as they are called—are associated with a ritual site called qwbth d-ʾbrhm, which Anthony convincingly renders as the “tent/tabernacle of Abraham” (cf. Syriac qūbtā). Such a description, moreover, seems to be compatible with early Muslim descriptions of the Meccan sanctuary itself.57 According to the Chronicler, the qwbtʾ was a place of worship and the performance of animal sacrifices, qwbrnʾ d-dbḥʾ, very much like the str-shelter of the Safaitic inscriptions. While this Arabian ritual practice had clearly been Abrahamicized by the early 7th century, we may consider the possibility that its roots are to be found in a ritual shelter like the str.
None of the known Safaitic texts provide information on the constitution or dimensions of the str-shelter. The best guess is that it was simply a tent, constructed of poles and hide. None of the sites at which str-inscriptions have been found have been excavated and so it is impossible to say more at the moment. One may suggest, however, that the cleared-out spaces associated with funerary sites may have hosted the temporary str-shelter.58
3.1 Statues
A single inscription records the offering of an image, ṣlm, as a means to obtain security from impending doom.59 The text appears to begin with mn “from” rather than the l that begins most texts,60 suggesting that the this was indeed a votive object.
RSIS 309
mn nśbt bn ʿgr h-ṣlm w tẓr mny w yṯʿ rwḥ
‘This statue is from (set up by) Nśbt son of ʿgr and Fate lay in wait so, O Yayṯeʿ, send relief.’
4 The Pilgrimage
Pilgrimages to the shrines of deities and the celebration of annual and seasonal festivals are attested throughout the ancient Near East.61 The pilgrimage was called ḥg in Safaitic and its verbal counterpart, ḥgg/yḥg. Inscriptions mentioning this rite are few as it seems that the pilgrimage sites were not located in the Ḥarrah—or at least none have so far been discovered. There are, therefore, no texts commemorating the rites performed on the pilgrimage itself; instead, the event is mentioned as a chronological anchor in dating formulae. One text commemorates the embarking on the pilgrimage in a certain year.
AL-MAFRAQ MUSEUM 24
w ḥgg snt myt mnʿt bn rḍwt w ḫrṣ ʿl-ʾhl-h f h lt w dśr slm w qbll
‘and he set off on the pilgrimage the year Mnʿt son of Rḍwt died and he kept watch for his family so O Allāt and Dusares, may he be secure and be reunited (with loved ones)’
The author’s prayer for reunion, qbll,62 is usually associated with verbs of motion and being distant from loved ones. This would imply that the pilgrimage took our author far away from his home and relatives; the performance of the pilgrimage does not seem to have included an entire tribe or family.
4.1 Time Period of the Pilgrimage
In the ancient Levant and in South Arabia, pilgrimages were performed at set times in the cultic calendar.63 Since the little information we have suggests that the nomads made pilgrimages to shrines in settled lands (see below), they would have followed the cultic calendars associated with those sanctuaries. But nothing would have prevented the authors of the Safaitic inscriptions from participating in the pilgrimage rites of various neighboring communities and cult-sites, and therefore we cannot know if the attestations of ḥgg referred to a single, annual event.
Two inscriptions provide chronological information regarding the pilgrimage. The first is KNGQ 4:
KnGQ 4 = AGQ 1
l ʾgrd bn fdy w ṣyr b-ḥg
‘By ʾgrd son of Fdy and he returned to water during the pilgrimage (period).’
Since the return to permanent water—which is what the verb ṣyr signifies—normally occurred in the period of ṣyf or beginning of qyẓ, it is possible that the change of seasons was celebrated with a pilgrimage to a shrine.64
The next text is WH 1771, which configures the ḥg with a calendrical period.
WH 1771
l dʾy bn nśl w rḥḍ b-h-ngm l-yḥg
‘By Dʾy son of Nśl and he ritually cleansed during Virgo to perform a pilgrimage.’
We cannot be certain whether the author was referring to the dawn or evening rising of ngm. Its dawn rising would have occurred in late August before the arrival of the Wasmī rains. Perhaps, then, the performance of the pilgrimage was connected to rituals of rain making, ensuring that the precipitation would arrive on time and be sufficient. On the other hand, if WH 1771 and KnGQ 4 = AGQ 1 refer to the same period, then we could suppose an evening setting of Ngm.65
4.2 Pilgrimage Sites
Only one inscription, BRENV.A 1, gives the name of a pilgrimage site, sʿʿ, that is Seʿīʿ/Seia (mod. Sīʿ), which is located about 3 km SE of Kanatha (Qanawat), a city of the Decapolis. This site was home to a sanctuary of Baʿal-Samīn, the storm god, which was constructed between the 1st c. BCE to the 2nd c. CE, and involved the participation of the tribe ʿObayśat, known in Safaitic as ʿbśt, Nabataean ʿbyšt, and Greek Οβαισηνοι. Members of this lineage produced a number of Safaitic inscriptions and appear to have been active in the Ḥarrah. Indeed, a man of the ʿObayśat tribe called upon Baʿal-Samīn in the Safaitic inscription CSNS 424 for security as he participated in a rebellion against the Romans.66 Other Safaitic inscriptions demonstrate that Seia was a focus of pilgrimage in honor of Baʿal-Samīn.67
BRENV.A 1
l qʿṣn bn s{l}y bn qʿṣn bn ḫlṣ bn nhb ḏ-ʾl dʾf w ngy n⟦⟧fr snt bṭl ḥg sʿʿ
‘By Qʿṣn son of {Sly} son of Qʿṣn son of ḫlṣ son of Nhb of the lineage of Dʾf and he escaped {by fleeing} the year the pilgrimage to Sʿʿ failed.’
It is impossible to know from such evidence what bṭl exactly meant.68 Was this a seasonal pilgrimage intended to ensure the coming of rains? And did its failure mean that the nomads of the Ḥarrah suffered drought that year, as suggested by Macdonald et al.?69 Drought is a common theme in dating formulae and so it is possible that this was simply a creative way of dating one’s text to a year with little rain. But perhaps bṭl meant that our author and his group could not make it to the temple that year, maybe on account of war or the closure of the temple itself, which seems to have happened sometime before the 4th c. CE. Another inscription related to Seia suggests the latter possibility. It too is meteorological.
IS.M 198
l ḍb bn sḫr bn ʿbd bn ʾdm w tẓr h-smy w ṣlf h-{m}l snt brḥ h-ʾṣlm sʿʿ {f} h lt slm
‘By Ḍb son of Sḫr son of ʿbd son of ʾdm and he kept watch for the rains and the livestock grew thin the year the images were removed from Sʿʿ so O Allāt, may he be secure.’
Could both texts be dated to the same year and reference the same drought? Could the bṭl of BRENV.A 1 therefore refer to the failure to complete the pilgrimage because temple was closed, which incidentally happened during a year of drought, allowing the nomads to assume a causal relationship between the two events?70 Whatever the case, the link between pilgrimage and favorable meteorological conditions is strongly implied.
5 Ritual Purity
Two verbs of ritual cleansing are attested, rḥḍ “to wash” and ṭhr “to purify.”71 The rḥḍ ritual is performed before embarking on a pilgrimage (see WH 1771 above) and so presumably places one in a state of ritual purity. Ritual purity is observed before entering temples in the Sabaic inscriptions, making use of the verb rḥḍ as well.72 This custom continues into the Muslim pilgrimage, which requires the observer to enter a state of ritual purity, ʾiḥrām.
Ritual ablution seems to have been required before engaging in other rites, like mourning. The author of KRS 2415 performs rḥḍ before stopping at a funerary cairn.73 The building of cairns is closely associated with the maṣṣēbâ in the Bible, although the two have not yet occurred together in the Safaitic inscriptions.74
KRS 2415
l gfft bn nʿrt w rḥḍ f gls ʾ-rgm
‘By Gfft son of Nʿrt and he washed and then stopped at the cairn.’
There are several other occurrences of the verb rḥḍ, but the inscriptions give us no specific information as to the purpose of ablution. Perhaps the most interesting of this category is C 4443, a long mourning inscription detailing both sacred (mourning for the dead) and secular activities. After grieving for several lost loved ones, including those murdered by an enemy tribe, he records pasturing and then washing during ṯbr, a time period, perhaps Sagittarius, and then keeping watch for the enemy. This sequence seems to follow the pattern we have seen with the ḏbḥ and nṣb texts, that is, the performance of a ritual (ablution) before undertaking a dangerous activity (keeping watch for enemies).
C 4443
w wgm ʿl-ʾm-h w ʿl-dd-h w ʿl-ḫl-h w ʿl-ʿm w ʿl-ʾnʿm qtl-h {ʾ}l ṣbḥ f wlh ʿl-bn ḫl-h trḥ w rʿy h-ḍʾn w rḥḍ b-ṯbr w ḫl h-ś[n]ʾ f h lt slm w wgd ʾṯr ʾḫ-h f ndm
‘and he grieved for his mother and his paternal uncle and his maternal uncle and ʿm and ʾnʿm, who the lineage of Ṣabāḥ murdered, and he was distraught with grief for the son of his maternal uncle, who had perished, and he pastured the sheep and washed during Ṯbr (Sagittarius) and kept watch for enemies so, O Allāt, may he be secure and he found the traces of his brother and was devastated by grief’
Other inscriptions record the performance of ritual washing, with no further information. These can be compared to the simple ḏbḥ inscriptions, a commemoration of a ritualistic act which would have been clearly understood by its author and original audience.
MKWS 21
l qnʾl bn ʾrśt w {r}ḥḍ
‘By Qnʾl son of ʾrśt and he washed.’
IS.H 641
l śgʾ bn śrk bn śdd w ⟦.⟧ rḥḍ b-wqʿt grmʾl
‘By Śgʾ son of Śrk son of Śdd and he washed in the pool of Grmʾl.’
The verb ṭhr is attested only once but in a clearly ritualistic context.75 The author of MA 1 performs ṭhr to enter a state of ritual purity in order to perform an animal sacrifice to Roḍay. While it is unclear what actions ṭhr would entail, both it and rḥḍ appear to imply the requirement of a shift from the profane to the sacred to perform religious rituals.76
MA 1
w ṭhr w ḏbḥ {l-}rḍy w ġnm nqt
‘and he entered into a state of ritual purity and made an animal sacrifice to Roḍay so may he gain a she-camel as spoil’
Conceptually, the performance of an ablution in the desert could be conceived of as a type of sacrifice. With water being scarce, the use of this precious resource for religious purposes could be understood in similar terms as the sacrifice and immolation of livestock for the deity.
6 Offerings
The inscriptions record several other types of offerings to the gods, but these are less frequently attested than the ḏbḥ-rite.77 The texts simply register the performance of the offering and petitions to the gods, but do not provide any further contextualizing evidence. At face value, they too appear to be personal acts of devotion.
6.1 Unspecified Offerings
Several other verbs could signify the presentation of an offering to the gods, qdm,78 qrb,79 qbl,80 and ntn.81 None of these attestations indicate what exactly was offered, but that they were intended to bring relief to those suffering from straitened circumstances is suggested by the prayers following them. This is much is stated explicitly in H 2411 and possibly WH 1731.
MNSA 2
qdm f h lt slm
‘he made an offering so, O Allāt, may he be secure’
H 2411
qrb brkt w ʾḫs f nʿm snt bʿlsmy
‘he made an offering (or: drew near to) at Brkt because he was suffering from scarcity so may Baʿal-Samāy show grace (this) year.’
C 1581
w qrb
‘and he made an offering’
KRS 1836
h bʿlsmn ḥlmt m-ḏ qbl
‘O Bʿlsmn, may that which he has offered bring forbearance.’
WH 1731
ʿny w ntn82
‘he suffered and so made an offering’
6.2 Burnt Offerings and Libations
Immolation of the sacrificed animal’s corpse was an important sacrifice type in the Hebrew Bible, there called ʿôlâ “that which goes up (in smoke).”83 The practice is attested at Ugarit as well with the verb, šrp.84 Three verbs associated with the burnt offering occur in Safaitic: ʾṣly, ʾsrf and ṣʿd.85
The burnt offering in SIJ 293 precedes an oath and could be a way to guarantee its fulfillment.
SIJ 29386
ʾṣly w ʾqsm b-ʾlh ḥy l-hdy ʿẓm
‘he made a burnt offering and swore by Allāh, who is living (or: (his) life), that he will lead with greatness’
BS 456
l PN w ʾsrf
‘by PN and he made a burnt offering’
The verb ṣʿd was suggested by Winnett to signify a burnt offering as well, which would be the Arabic equivalent of ʿôlâ. It is attested only three times, all in unclear contexts and so it could possibly refer to a feature of the landscape as well or simply the verb “to ascend.”
ASFF 456/SIJ 432
w ṣʿd
‘and he made a burnt offering’ (?)
WH 604.1
l PN h-ṣʿdt
‘by PN is this burnt offering’ (?)
The libation is perhaps only attested once in the inscriptions, with no explicit connection to the sacrifice or mention of the type of liquid poured out.87
C 4454
l ḥml bn nśbt w ṣbb b-ksʾ {g}ml
‘By Ḥml son of Nśbt and he made a libation during the evening setting of gml (Gemini).’
6.3 Images
Safaitic inscriptions often accompany images of desert life, from animals to raids and festivals.88 The associated texts usually comprise the signature of the artist, introduced by the lām auctōris l-, which signifies possession or authorship. On their own, the interpretation of such carvings is difficult—scholars interpreted their purpose along similar lines as the texts.89 A few inscriptions, however, do suggest that such carvings could be a type of offering to a deity, perhaps comparable to the offering of cultic objects among settled peoples, such as clay figurines.90 Such drawings are accompanied by a verbal adjective qṣy “to dedicate” followed by the name of a god.91
LP 317 = IS.M 92 (Figure 10)
l ḥr bn qnʾl bn qḥś bn ḥḍg h-nqt qṣyt l-nh{y}
‘By Ḥr son of Qnʾl son of Qḥś son of Ḥḍg is this she-camel, which has been dedicated to Nhy.’
KRS 1307 (Figure 11)
l ḫṭst bn skrn bn ḫṭst bn zkr h-gml qṣy l-śʿhqm
Figure 10
The Safaitic inscription LP 317
Courtesy: OCIANA
Figure 11
The Safaitic inscription KRS 1307
Courtesy: OCIANA
‘By Ḫṭst son of Skrn son of Ḫṭst son of Zkr is this camel, which has been dedicated to Shayʿhaqqawm.’
C 1658
l wq{r} bn yʿl h-gmln qṣyn l-ʾlt w l-rḍw f h yṯʿ ʿwr m ʿwr h-[ḫ]ṭṭ
‘By Wqr son of Yʿl are these two camels dedicated to ʾAllāt and Roḍaw, so O Yayṯeʿ blind whosoever effaces these carvings.’
It may be the case that the participle qṣy was optional, as the purpose of an image was clear in its original context, not requiring redundant written elaboration. Prayers accompanying some signed rock art also open the possibility that they were votive images.
ASFF 428
l rgl bn hmt bn ʿm bn mṯʿ bn ʿmd h-gml w tẓr h rḍw flṭ-h mn sḫl
‘By Rgl son of Hmt son of ʿm son of Mṯʿ son of ʿmd is this camel and he kept watch; O Roḍaw deliver him from weakness.’
Rock art is also associated with burials. BRCM 14 mentions an image of a horseman with the verb nṣb and the construction of a funerary installation. This context could allow for the interpretation of these images as burial offerings.
BRCM 14 (Figure 12)
l ʾbḥt bn gls bn fśgt bn glḥn w nṣb w l-h [h-]frs w bny ʿl-ʾs
‘By ʾbḥt son of Gls son of Fśgt son of Glḥn and he set up a cult stone and this (image of a horse) is by him and he built over ʾs.’
If the str-shelter is in fact a cultic installation, then the inscription ABSWS 81 (Figure 13) may also connect a drawing of a camel with cultic ritual. While the exact archaeological context was not recorded, the stone seems to lie within the boundaries of a stone enclosure, perhaps where the str-shelter was set up.
In most cases, however, we cannot be sure if carved images at mortuary installations were originally components of the burial ritual or if they were latter additions by passersby.
Figure 12
The Safaitic inscription BRCM 14
Photo: P. Bikai. Courtesy: OCIANA
Figure 13
The Safaitic inscription AbSWS 81
Photo: S. Abbadi. Courtesy: OCIANA
7 Vows and Oaths
The verb nḏr—the common verb for taking a vow across Semitic—is attested only once in KWQ 42; it may be significant that it occurs in the context of water or perhaps this is a situation in which it is better to understand ḥḍr as “during a festival.”92
KWQ 42
l ʿzm bn mrʾt w nḏr ḥḍ{r}
‘By ʿzm son of Mrʾt and he made a vow {while camping by water/during a festival}.’
The qsm oath (see SIJ 293 in § 6.2) is also attested, cf. Quran 5:53. See the discussion of this inscription above.
Finally, oaths of self-amputation are thrice attested, twice in the context of retribution. The oath-taker offers to cut off his hand in promise of vengeance against those who have wronged him.
C 25
wgd ʾṯr ʾḫ-h f wlh ʿl-h f hy lt w h śms ʾṭn ʾl-km yd-h l-ṯʾr m-ḏ ʾslf
‘he found the trace of his brother so he was distraught with grief for him, so, O Allāt and Śams, may he cut off his hand for you (in promise) for vengeance against him who has committed this act’
AKSD 5
ṯqb yd-h w dśr ʿl-ksr wqʿ-n
‘may he cut off his (own) hand so that Dusares act against any destroyer of our inscription’
C 4453
w qṣʿ yd-h f ḥy w rḍw
‘and may he cut off his hand and life that Roḍaw …’
None of the texts provide information about other rituals that might accompany such oaths.93
8 Sacred Water
One inscription seems to refer to a sacred water source associated with Baʿal-Samīn; BESS19 1 records a so far unique invocation to the storm god.
BESS19 194
l rġḍ bn ġsm bn śḥl w ḫrṣ ʿl-ḥmlt sqm f h bʿlsmn ḥnn nqʿ-k f-yql ʾl-h ṣdq
‘By Rġḍ son of Ġsm son of Śḥl and he kept watch over Ḥlmt who had fallen ill so, O Baʿal-Samīn, show mercy though your watering places so that his folk will say he is just.’
The term nqʿ is attested for the first time here but a related word, augmented with the feminine t, nqʿt, is attested several times as the goal of the verb of motion wrd, “to go to water.”95 This movement is often configured with periods in the rainy seasons, implying that such inscriptions were carved during droughts, and that nqʿt was a place of permanent water to which the nomads would flock when the rains failed. The present noun nqʿ could reflect a plural “watering places” or perhaps a different derivative from the same root. Whatever the case, the use of sacred water connected to a deity to alleviate illness is a well-attested cross-cultural practice. In the Arabian context we may furnish two examples. The first is a legend recounted by ibn al-Kalbī which goes as follows:96
It was al-Harith who used to be the custodian of the Ka’bah. But when ʿAmr ibn-Luhayy came [to Mecca] he disputed his right to its custody, and with the aid of the children of Ishmael, fought the Jurhumites, defeated them, and cleared them out of the Kaʿbah; he then drove them out of Mecca, and took over the custody of the Sacred House (al Bayt) after them. He then became very sick, and was told, “There is a hot spring in al-Balqāʾ, in Syria (al-Shaʾm); if you would go there, you would be cured.” So he went to the hot spring, bathed therein, and was cured. During his stay there, he noticed that the inhabitants of the place worshipped idols. He, therefore, queried them saying, “What are these things?” To which they replied, “To them we pray for rain, and from them we seek victory over the enemy.” Thereupon he asked them to give him [a few of those idols], and they did. He took them back with him to Mecca and erected them around the Kaʿbah.
trans. Faris 1952: 7
While certainly not historical, it illustrates the trope of the healing power of waters associated with gods. The second is the Zamzam well of Mecca, which Muslims believe to have healing properties.97
According to OCIANA (accessed 25/3/2021), there are 91 occurrences of the verb ḏbḥ, to which we may add a handful of other reference to sacrifice using synonyms.
Ryckmans (1950–1951).
Macdonald (forthcoming).
All data are based on OCIANA, consulted on 25/3/2021.
For an outline of sacrifice in Antiquity, see Ekroth (2014). On sacrifice in the religions of ancient Canaan and Israel, see Nakhai (2001), and across South Arabia as well, see J. Ryckmans (1993); Robin (2012, § 2d). Sacrifice figures prominently in literary accounts of pre-Islamic Arabian religion (Hoyland 2001: 162–166).
A related function is attested in South Arabia; sacrifice was meant to produce a “divine response” through an oracle (Hoyland 2001: 154).
Canaanite zbḥ, Aramaic dbḥ, Ugaritic dbḥ (DNWSI, 301–302). Ancient South Arabian ḏbḥ (Beeston et al. 1982: 37–38).
SafDict, 66, 78b. The verb hrq, whose Classical Arabic cognate means “to pour,” could reflect a sacrifice and libation.
I quote only the relevant portions of the cited inscriptions in the next sections for the sake of space. In most cases, the genealogies, which can be quite lengthy, have been omitted.
Camel sacrifice is common in The Book of Idols, e.g. the tale of the idol Saʿd (ibn al-Kalbī 1913: 37) and also the Quranic tale of Ṣāliḥ, Q 26:155–157.
Ryckmans (1950–1951: 435–436).
For an important typological parallel of this type of sacrifice from ancient Greece, see Jameson (2014, ch. 6 “Sacrifice before Battle”). I thank Prof. Duncan MacRae for this reference.
This is my interpretation as discussed in Al-Jallad (2017); Hayajneh (2016) suggests the understanding of ḥll as “to return to a profane condition.”
This is my reading and interpretation; the editio princeps (OCIANA) read the dating formula as s¹nt {ʾ/k}{k}{r}{ḫ}ṣfrṣ h-mdnt and does not provide an interpretation. The verb ṣdy is attested for the first and only time here. Considering the context, one may connect it to Classical Arabic taṣdiyatun “the clapping with the hands,” perhaps a type of prayer. Lane gives an example of the tD-stem in such a context: ṣalātu-humu t-taṣaddī wa-l-mukāʾ “their prayer is the clapping with the hands and whistling” (Lane, 1670b).
Ryckmans (1950–1951: 436).
Ababneh and Harahsheh (2015).
Ababneh and Harahsheh (2015: 48).
See, for example, 1 Sam 9 12–14; Kings 3:4; 11:17. For a discussion of the High Place and the sacrality of landscape, see Eksell (2002: 133–137); see also Healey (2001: 73) and Alpass (2013: 68–73) on the role of high places in Nabataean religion. On the distribution of the inscriptions and rock art, see Macdonald (1992b); Brusgaard (2019, ch. 6).
See the discussion in Healey (2001: 48) and Alpass (2013: 68–73), and the references there.
SafDict, 129a.
There is much controversy surrounding the dating of this text, the existence of ʿAbīd b. Sharyah, and even on the vocalization of his name (an alternate vocalization ʿUbayd exists). The text as we have it is known from one 16th c. manuscript and its contents are traced back to al-Barqī (d. 249/863). On these matters, see Crosby (2007). It is unclear how much of this material in fact goes back to a historical ʿAbīd, if such a person did in fact exist, but its contents are largely folkloric and, with the exception of certain theonyms, anthroponyms, basilonyms, and tribal names, do not reflect a faithful transmission of pre-Islamic knowledge.
This line he attributes to one Abū Saʿīd al-Muʾmin.
The occasion for the assembly of the tribe implies a sacral communal activity, such as a pilgrimage.
Thanksgiving sacrifices are recorded in South Arabia as well (J. Ryckmans 1993; Hoyland 2001: 154; Robin 2012, § 2d.) and one can understand the ʿAqīqah sacrifice performed by Muslims today on the occasion of the birth of a child along these lines as well.
For a broad comparative perspective, see the masterful study of Mettinger (1995); on nṣb-stele in Phoenicia, see Doak (2015).
The substantive nṣb is attested throughout West Semitic, where it refers generally to a stele, usually funerary; see DNWSI, 750. In Ancient South Arabian, the nṣb also has a funerary function (SD, 99).
Healey (2001: 156). For a comprehensive treatment of the Nabataean Betyls, see Wenning (2001).
See Hutter (1993).
Mettinger (1995: 130–131).
Sanders (2013: 99–100).
Also cf. Acts 15:29 and 1 Corinthians 8; see Reynolds (2018: 189).
Hoyland (2001: 186).
Ibn al-Kalbī (1913: 8); see also Wellhausen (1897: 101). Littmann (1943: 56) interprets this phrase as setting up a statue for the god, but it seems more likely that the deity was represented by a natural, unworked stone. See also Al-Azmeh (2014: 214).
The path to funerary stele is beautifully illustrated in the Katamuwa stele, where the nbš “life force” of the deceased resides in the nṣb, just as the presence of a deity would have occupied such a stele. See Pardee (2009) for the edition of the text and Sanders (2013) for further analysis.
This might be compared to the sacrificial altars in South Arabia; see Maraqten (2021: 452–456).
Nakhai (2001: 49).
See Macdonald et al. (1996: 456 and forthcoming) for further discussion.
Littmann interprets this strange dating formula to refer to a drought (1943: 56). Note the similar wording of the narrative in SG 1, which may refer to the same event: SG 1: ṭrd h-ḏʾb ẓlʿ m-mn ʿkd yglḥ “he drove away the wolf, which was seeking to mate, from Mn after it attacked” (SafDict, 71). In light of the latter inscription, it seems more likely that the “wolf” refers to some marauding group rather than a drought. The h preceding the first lt could be a writing error as the name is usually preceded by the vocative, or it could reflect a variant pronunciation, hallāt. One might suggest that the two spellings in this text, lt and hlt, should be regarded as separate deities, but as far as I know, there is only one other case where hlt occurs outside of a vocative context, SIJ 840. The narrative and prayer of this inscription read: w mṭy f h hlt slm w ġnmt “and he set off on a journey so O hlt grant security and spoil.” On the same stone, however, another inscription (SIJ 841) carved in the same hand contains the same narrative and prayer yet this time invoking lt: w mṭy f h lt ġnmt w slm “and he set off on a journey so, O lt, grant spoil and security.” Thus, I think it is better to explain the divine name hlt in the former inscription as the result of dittography.
Harding (1953); SafDict, 61.
SafDict, 41; and earlier in Al-Jallad (2015: 298).
This text is RWQ 325 l ʾḏnt bn ʿbd bn kn h-sbqy w ʾkl lḥm smn w zm w nẓr “By ʾḏnt son of ʿbd son of Kn the Sbq-ite and he consumed fatty meat and a dish of milk and kept guard.” One Hismaic inscription (KJC 46), however, seems to record a votive offering of a meal to the gods Dusares and Kutbay. We will return to the relationship between the nṣb and the nfs in their mortuary context, specifically how they related to the notion of an afterlife, in § 5.
Compare also with Classical Arabic manṣibun and modern Arabic manṣib, both meaning “a place where something is planted” or “rank,” “office.”
On the development of the npš-stele, see Steiner (2015). See Mouton (1997) on its spread in Arabia. Hayajneh (2018) surveys the Arabian evidence and publishes a new Safaitic (or Thamudic) nfs-text from the region of Gerasa on a stone slab.
RWQ 328 l ṯʿl bn ʿmd bn mlk bn qḥś h-nfs t f h lt ʿwr ḏ-yʿwr “This funerary monument is for Ṯʿl son of ʿmd son of Mlk son of Qḥś so O Allāt blind him who would efface (it).”
The most common form of the word is nfst, but I have argued that the final t may be understood as the feminine demonstrative as the word appears as nfs in all other languages and the basic form nfs is also attested in Safaitic. The plural construction is also occasionally followed by the plural demonstrative ʾly, h-ʾfs ʾly “these funerary monuments,” further supporting the idea that this formula includes the use of a proximal demonstrative; see the discussion under the lemma nfs in SafDict, 103.
This mission was led by Ahmad Al-Jallad and Ali al-Manaser.
The site is described in Appendix 2.
M.C.A. Macdonald (2006) advanced a cogent argument that the nfst referred to the name of the deceased inscribed on stone, similar to the meaning of npš in the Aramaic of the Ḥawrān. At the time only two occurrences of bny h-nfst existed. In light of the present discovery, however, such an interpretation seems difficult to maintain, at least for every occurrence. One text clearly states the name of the dead person associated with the ʾfs and the other texts on the stone record “building” the ʾfs. This panel demonstrates that the nfs is some type of installation.
A full archaeological and epigraphic examination of the site, and another related one, are in preparation. Their publication has been delayed by the COVID-19 phenomenon, which has prevented the team from returning to the field.
The site clearly served a funerary function as several other groups of texts surrounding the enclosure recording grieving/funerary narratives. These are edited in Appendix 2.
Nehmé (2010: 459–460); Hayajneh (2017).
SafDict, 124.
SafDict, s.v.
For a comprehensive treatment of Israel’s Tabernacle, see George (2009); see also the classic study of Cross (1947).
str = a denominal verb “to construct the str,” ʾḫḏ h-str “he took possession of the str”; bny h-str “he constructed the str”; l PN h-str “this str belongs to/was constructed by PN.”
It is tempting in these cases to interpret ḥḍr along the lines of the Ancient South Arabian cognate, namely, “to perform a pilgrimage” and the C-stem hḥḍr “to hold a festival for a deity” or “to offer,” see the discussion in Maraqten (2021: 433–434), but there is no independent evidence in Safaitic to support such an understanding. In most cases, ḥḍr is ambiguous being the only verb in the narrative, yet in all cases that provide any context, the understanding of “to encamp by water” is best supported; see Macdonald (1992d: 29); SAFDICT, 84.
Anthony (2018: 35–36).
Anthony (2018: 36) provides the description of Ibn Jurayj, an early Meccan scholar, who describes the primitive Kaʿbah as an ʿarīš, “a tabernacle-like structure.” See also Rubin (1986).
See the nfs complex in Appendix 2.
This word is cognate with Classical Arabic ṣanam; see SafDict 129a. The term, ṣlmʾ, is much more common in Nabataean as a votive object (Healey 2001: 159).
The introductory l is called the lām auctōris and is usually translated as “by” but can often mean “for,” as in the case of grave inscriptions; see Macdonald (2006) for a discussion. In the present inscription, the first ⟨m⟩ takes the slender form, which differs from the other ⟨m⟩ in the text. The slender form is typical of the Fine Script while the present inscription takes the Common letter shapes. If the first m is not associated with the text, then it might be possible to consider the first vertical stroke of the text the letter l, the lām auctōris.
See the chapters in McCorriston (2011). For Arabian pilgrimages, see Maraqten (2021 and 2015) on the pilgrimage to the temple of Awām at Maʾrib, Yemen; Al-Ghul (1984) discusses the pilgrimage to Itwat. On pilgrimages in the Hebrew Bible and in the archaeological record, see Nakhai (2001).
SafDict, 109a.
The pilgrimage to Awām seems to have taken place sometime in February–March (Maraqten 2021: 452); the Sabaean month ḏ-mḥgtn further indicates that the pilgrimage was to take place at a fixed time of the year. In North Arabia as well, spring festivals appear to have been observed (Alpass 2013: 42–44; Healey 2001: 161; Wellhausen 1897: 79–101). The Bible records three seasonal/agricultural pilgrimages: Deuteronomy 16 describes these three: Pesach and Unleavened Bread (Deut 16:1–8); Shavuot/Weeks (Deut 16:9–12); Sukkot/Booths (Deut 16:13–17). The annual pilgrimage of pre-Islamic times continues into Islamic practice at Mecca, which takes place from the 8th to the 12th or 13th of the month ḏū l-ḥiǧǧah. On the Muslim pilgrimage, see Peters (1994).
The seasons of ṣyf likely began in mid-April and stretched to June; during this period, depending on the year, there would have been herbage remaining from the seasons of the later rains, dṯʾ, and surface water as well. During qyẓ, the desert would have been hostile, lacking water and herbage. See Macdonald (2020) and (1992c).
See Al-Jallad (2014; 2016) on the Safaitic parapegma.
Macdonald (2003: 278) cautiously suggests a connection between the Safaitic ʿbśt and the group mentioned in Greek and Nabataean texts; CSNS 424 makes this link highly probable.
Healey (2001: 65). While the term for “temple” is not attested in Safaitic, a Hismaic text from Wadi Ram, published by Farès and Zayadine (1998), commemorates the construction of the temple (bt) of Allāt by the tribe of ʿd, likely Quranic ʿĀd: w bny bt lt ḏ ʾl ʿd “and he of the lineage of ʿd helped construct the temple of Allāt.” The same term for temple is used in South Arabia as well; see Robin (2012: 19–20), and in the Quran, e.g. 22:26.
For a thorough etymological discussion of this root, see Prioletta and Hull (2020).
Macdonald, Muazzin and Nehmé (1996: 463 and n. 77); Macdonald (2003: 278).
Four pedestals at the site supported statues, one of which was of Herod the Great, and two reliefs of Mithra were recovered from the site and are now housed at the Damascus museum;
SafDict, 117a, 137a–b. In Classical Arabic, this root has largely been replaced by ġsl, although it retains the meaning “to wash” as a transitive verb. The passive form ruḥiḍa “to be wet, covered in water” survives as an idiom referring to one covered in sweat from a fever (Lane, 1052c).
Maraqten (2021: 448). Given the South Arabian parallel, I find it less likely that the term rḥḍ to be a non-sequitur referring to secular bathing or being ill.
The rgm cairn seems to have had a funerary function as almost all details about the structure suggest (SafDict, 115). Another term for a cairn is ṣwy (SafDict, 130).
Nakhai (2001: 44); Gen 31:43–54, where Laban and Jacob set up a maṣṣēbâ and then a gal, a pile of stones. The terms gl and glt (WH 1873 and KWQ 37, respectively) are attested in Safaitic, which are treated neutrally in the SafDict as “stone.” But I would cautiously suggest interpreting it as a type of cairn, cognate with the Hebrew gal and Syriac gallā, “a heap of stones.”
This is the primary verb used to express ritual purity in the Quran, Biblical Hebrew, and is attested in Sabaic as well (Maraqten 2021: 447–448; SD 153).
The rules for the performance of sacrifices in the Hebrew Bible, including matters of ritual purity, can be found in Leviticus 1–7 and in Deuteronomy; on this, see Anderson (1992a–b). The Ugaritic texts are concerned with matters of bodily purity when concerning the king and these are similar to the stipulations in the Hebrew Bible; see Pardee (2002: 234).
For an overview of the types of offerings made in pre-Islamic Arabia, see Hoyland (2001: 163–166).
Classical Arabic qaddama “to place before,” “to offer.”
Aramaic qurbānā “sacrifice,” and in the D-stem “to offer, dedicate”; DNWSI, 1028–1030.
Perhaps qabbala “to cause to receive,” cf. Classical Arabic G-stem qabila “to receive.” Alternatively, it could be connected to the infinitive qbll “to be reunited” (SafDict, 109), but such a meaning does not seem to fit the current context.
Canaanite ntn, Hebrew nāṯan “to give.”
The alternative understanding of this word would be to take ntn as Classical Arabic natana and natnun “to stink.”
See for example Exodus 20:20.
Nakhai (2000: 42); “to burn by fire” DNWSI 1194. The sound correspondences between Arabic s and NWS ś suggest that the verb was a borrowing. A connection with the Classical Arabic ʾasrafa “he transgressed” (Lane, 1351a) is also possible; see section § 4.
The verb sbʾ was suggested to have such a meaning in the SafDict, but it seems more likely to me now that it should be interpreted as a simple verb of movement, compare with Sabaic sbʾ “to undertake an expedition” (SD 122) and Classical Arabic subʾatun “a long journey” (Lane, 1287a). Thus, HASI 24 would read w sbʾ m-ḥrn w ḫyṭ f ʾmgd w ʾḫlṣ “he set off on a long journey from the Ḥawrān and travelled quickly so may he attain bounty and deliverance.”
On the interpretation of this text, see Al-Jallad (2021b).
Numbers 28:6–7 indicates that libations accompany the twice-daily burnt offerings required of Israel; 1 Kings 18:33 records the pouring of a libation over the burnt offering.
On the classification of rock art motifs and their distribution in the landscape, see Brusgaard (2019).
Compare Macdonald (2006) to Eksell (2002).
See Lacerenza (1988–1989: 142–144).
Littmann (1943: 78) interpreted such texts as “bills of sale,” but there is no mention of the transfer of money and property and it is hard to understand what function a public bill of sale on a rock in the middle of the desert would serve. Ryckmans (1950–1951: 436) suggests that these were drawings of animals dedicated to the gods, but given that none of the other narrative components are accompanied by illustrations, I think this tips the scale of understanding in favor of the carving itself being an offering, which I have suggested in (2015b: 204). Charloux, Guagnin, and Norris go on to suggest that a similar interpretation could be applied to the monumental camels of North Arabia, including those associated with Thamudic B inscriptions (2020: 102).
Northwest Semitic ndr/nzr (DNWSI 717–719); Hismaic nḏr (Graf and Zwettler 2004; Al-Jallad 2020); and frequently in Dadanitic nḏr, see Kootstra (2019) and OCIANA for references.
For example, in Gen 31:44–49, the construction of a cairn accompanies the taking of oaths while oaths are made following a sacrifice in Jonah 1:16. For more, see Hutter (1993); Bloch-Smith (2005).
See Appendix 2 for the edition of this text.
SafDict, 105b–106a.
I thank Mr. Asad Uz Zaman for bringing this connection to my attention.
For example, Sahih al-Bukhari 3261: “Abū Ǧamrah al-Ḍubaʿī narrated: ‘I used to sit with Ibn ʿAbbās in Mecca. Once I had a fever and he said (to me), “Cool your fever with Zamzam water, for Allāh’s Messenger said: ‘It, (the Fever) is from the heat of the (Hell) Fire; so, cool it with water (or Zamzam water).’ ” Compare also with the story of Naʿamān the Aramaean in 2 Kings 5.