Top Banner
An ancient Arabian zodiac. The constellations in the Safaitic inscriptions, Part I This contribution identies a previously unrecognised mode of reckoning time in the Safaitic inscriptions - the stars. Twelve zodiacal constellations are identied, along with the constellations Orion and the Pleiades. Keywords: Ancient North Arabian, astronomy, zodiac, epigraphy, Safaitic Ahmad Al-Jallad LUCL & LIAS, Leiden University, Room 1.07B, Witte Singel 25/M de Vrieshof 4, 2311 BZ Leiden, The Netherlands e-mail: [email protected] 1. Introduction This paper identies the names of the zodiacal constel- lations in the Safaitic inscriptions 1 and their use for the reckoning of time. The North Arabian zodiac differs in many signicant ways from the contemporary Aramaic and Greek systems while sharing important similarities with the Babylonian tradition. In the following, I will present the process of decipherment, followed by an etymological discussion and a selection of relevant inscriptions. A fuller historical treatment will be carried out in a second article. 2. The seasons and the constellations Many ancient societies relied on the stars for the reckoning of time. The identication of the constellations is therefore connected to the yearly cycle as experienced by the inhab- itants of the arrah. In a brilliant 1992 article, M.C.A. Macdonald identied four seasons attested in the Safaitic inscriptions, which correspond to four out of the ve seasons known to the Rwala Bedouin, as recorded by Musil (Table 1). Three of these appear together in two inscriptions: Mu 113 ls : ʿd bn b bn ʿbd bn ʾdm w ll h- dr dʾ f ʾyf s²ty f h bʿls¹mn qbll By ʿd son of b son of ʿbd son of ʾdm and he camped here during the season of the later rains, then the dry sea- son, and then winter so, O Bʿls¹mn, show benevolence Commentary ʾy: this rare reex of the verb qy, which exhibits the sound changes q > ʾ and > , is attested only once more in MKMR 9; see Al-Jallad (2015: §3.7.4.1) and Macdonald (2004: 498) for further discussion. The same sequence is encountered in an inscription from the arrah, which is rather difcult to interpret: SIJ 1008 2 l <<>>bʾmrh {w} gdʿw[] nw{y} w ʾqwy w dʾ w qy[] w s²ty h- rbt ʿwd w h ʾ{l}{t} s¹fʾ -hm By Bʾmrh and, {O Gdʿw(divine name)}, let there be pasture and endurance while he spends the season of the later rains, the dry season, and the winter, at this raabah for a second time; and, O ʾlt, let them [i.e. the seasons] provide sustenance 1 Safaitic refers to a script used by the inhabitants of the arrah, the basalt desert of what is today northern Jordan and southern Syria and the adjacent portions of Saudi Arabia, to inscribe rock inscriptions in several closely related dialects of Old Arabic (see Huehnergard, forthcoming). 2 While the rst {w} more closely resembles a g on the copy, the inscription clearly transitions from a personal name to a divine name, and so a conjunction is expected. The g is distinguished from the w by a diameter or secant line. If the absence of the line is not the result of a copy error, then it is possible that the author forgot to carve it after forming the circle. The next word gdʿw is best restored as gdʿw[], a manifestation of the deity Gadd. The syntax of the prayer formula to Gdʿwis paralleled in C 4457. The single of this inscription was restored by Macdonald (1992a: 7). I have no explanation as to why the author began with larger letters, swerved into a progression of small letters, and then went back to carving large ones. 214 Arab. arch. epig. 2014: 25: 214230 (2014) Printed in Singapore. All rights reserved
24

Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

Jul 18, 2016

Download

Documents

Middle Eastern & Arabian Zodic
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

An ancient Arabian zodiac. The constellations in theSafaitic inscriptions, Part I

This contribution identifies a previously unrecognised mode of reckoning time in the

Safaitic inscriptions - the stars. Twelve zodiacal constellations are identified,

along with the constellations Orion and the Pleiades.

Keywords: Ancient North Arabian, astronomy, zodiac, epigraphy, Safaitic

Ahmad Al-Jallad

LUCL & LIAS, Leiden

University, Room 1.07B, Witte

Singel 25/M de Vrieshof 4, 2311

BZ Leiden, The Netherlands

e-mail: [email protected]

1. Introduction

This paper identifies the names of the zodiacal constel-

lations in the Safaitic inscriptions1 and their use for the

reckoning of time. The North Arabian zodiac differs in

many significant ways from the contemporary Aramaic

and Greek systems while sharing important similarities

with the Babylonian tradition. In the following, I will

present the process of decipherment, followed by an

etymological discussion and a selection of relevant

inscriptions. A fuller historical treatment will be carried

out in a second article.

2. The seasons and the constellations

Many ancient societies relied on the stars for the reckoning

of time. The identification of the constellations is therefore

connected to the yearly cycle as experienced by the inhab-

itants of the Ḥarrah. In a brilliant 1992 article, M.C.A.

Macdonald identified four seasons attested in the Safaitic

inscriptions, which correspond to four out of the five

seasons known to the Rwala Bedouin, as recorded by

Musil (Table 1).

Three of these appear together in two inscriptions:

Mu 113

l s:

ʿd bn ḍb bn ʿbd bn ʾdm w ḥll h- dr dṯʾ f ʾyḍ f s²ty f h

bʿls¹mn qbll

By Ṣʿd son of Ḍb son of ʿbd son of ʾdm and he camped

here during the season of the later rains, then the dry sea-

son, and then winter so, O Bʿls¹mn, show benevolence

Commentary

ʾyḍ: this rare reflex of the verb qyẓ, which exhibits the

sound changes q > ʾ and ẓ > ḍ, is attested only once

more in MKMR 9; see Al-Jallad (2015: §3.7.4.1) and

Macdonald (2004: 498) for further discussion.

The same sequence is encountered in an inscription

from the Ḥarrah, which is rather difficult to interpret:

SIJ 10082

l <<>>bʾmrh {w} gdʿw[ḏ] nw{y} w ʾqwy w dṯʾ w qy[ẓ]

w s²ty h- rḥbt ʿwd w h ʾ{l}{t} s¹fʾ -hm

By Bʾmrh and, {O Gdʿwḏ (divine name)}, let there be

pasture and endurance while he spends the season of the

later rains, the dry season, and the winter, at this raḥabah

for a second time; and, O ʾlt, let them [i.e. the seasons]

provide sustenance

1 Safaitic refers to a script used by the inhabitants of the Ḥarrah,

the basalt desert of what is today northern Jordan and southern

Syria and the adjacent portions of Saudi Arabia, to inscribe rock

inscriptions in several closely related dialects of Old Arabic (see

Huehnergard, forthcoming).

2 While the first {w} more closely resembles a g on the copy, the

inscription clearly transitions from a personal name to a divine

name, and so a conjunction is expected. The g is distinguished

from the w by a diameter or secant line. If the absence of the line

is not the result of a copy error, then it is possible that the author

forgot to carve it after forming the circle. The next word gdʿw is

best restored as gdʿw[ḏ], a manifestation of the deity Gadd. The

syntax of the prayer formula to Gdʿwḏ is paralleled in C 4457.

The single ẓ of this inscription was restored by Macdonald

(1992a: 7). I have no explanation as to why the author began

with larger letters, swerved into a progression of small letters,

and then went back to carving large ones.

214

Arab. arch. epig. 2014: 25: 214–230 (2014)

Printed in Singapore. All rights reserved

Page 2: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

Commentary

nwy: this word usually appears as a verb meaning ‘to

migrate’, but prayer formulae usually express a request for

a boon or favour. Here, nwy should probably be taken as

‘a grazing place’ or ‘pastureland’ (HALOT #6037), and

would stand opposite the known curse mḥlt ‘dearth of pas-

ture’; see for example, WH 24.

ʾqwy: this is the infinitive of the C-stem of the root √qwy,

which signifies ‘strength or endurance’ in Classical Arabic

(CAr), here perhaps, ‘grant strength, endurance’. Infini-

tives are commonly used for the purpose of commands in

the Safaitic inscriptions as an alternative to the imperative.

h-rḥbt: Macdonald (1992b: 23) has argued convinc-

ingly that the term rḥbt, when it appears with the article

h-, refers to an open area in which water collects and pro-

duces abundant herbage, a raḥabah. This is in contrast to

the term without the article, which refers to the area

known today as al-Ruḥbah.

s¹fʾ -hm: this phrase reflects a request from the deity ʾlt,

so the basic sense of this root as found in the CAr lexicons

as ‘foolish’ or ‘ignorance’ seems unsuitable. A connection

with Ugaritic spʾ ‘to eat, devour’ (del Olmo Lete & San-

mart�ın 2004: 766) is more appropriate, and in the current

context may reflect a verbal noun with the subject

expressed as a pronominal suffix.

Both of these inscriptions suggest that the year was

divided into three primary seasons (Table 2): the winter

s²ty which, as Macdonald has recognised, encompassed

the contemporary Bedouin as:

-s:

feri and a�s-�sta, stretching

from early October to mid-February. Dṯʾ would then corre-

spond to the Rwala as-sm�ak, stretching from mid-February

to mid-April. Qyẓ corresponds to the dry season, but may

also have included the early summer. The term s:

yf occurs

only three times in the Safaitic inscriptions,3 which sug-

gests that it was not a regular season for their authors.

M.C.A. Macdonald has suggested to me that it may signify

a transitional period between dṯʾ and the dry season, qyẓ.

Two other inscriptions record a series of three terms in

an identical order, mlḥ, ḏkr and ʾmt. The translations of the

editiones principes follow:

C 4985

l zdʾl bn n _gft bn ʾʿly bn ḍḥy bn ḥwq bn kwnt bn s²wʾ w

ḥḍr ẖlqt mlḥ w ḏkr w ʾmt w s¹lḥ h- ʾbl mdbr w zrʿ h- rḥ

(—)bt f h ʾlt s¹lm w nqʾt l- ḏ yʿwr

By Zdʾl son of N _gft son of ʾʿly son of Ḍḥy son of Ḥwq

son of Kwnt son of S²wʾ and he stayed in the caves of

Maliḥ, Ḏakir and Amat, and the camels were dispersed in

the desert. And he sowed the Ruhbat and, O Lat, health

and vengeance against him who obliterates4

KRS 169

l nʿrt bn ẖdmt bn trml bn s¹ry bn s¹hm w _gz b- nẖl f h yṯʿ

s¹lm w klʾ h- s¹my ḥḍ{r} mlḥ f ḏ{k}{r} f ʾmt f yʾs¹ f h

bʿls¹mn r{w}ḥ w s¹lm

By Nʿrt son of Hdmt son of Trml son of S¹ry son of

S¹hm and he was on a raid in a valley and so O Yṯʿ [grant]

security and he watched the sky {camping near a perma-

nent source of water} [at] Mlḥ and {Ḏkr} and ʾmt and

Yʾs¹ and so O Bʿls¹mn [send] {relief from adversity and

uncertainty} and [grant] security

I will comment on both inscriptions more extensively in

the next section. A few words on the interpretation of klʾ

in KRS 169 are in order, however, since it bears directly

on the interpretation of the terms mlḥ, ḏkr and ʾmt. The

translation of the editio princeps of klʾ as ‘watched’ is

probably incorrect. Instead, it should be connected with

the sense of ‘retain, detain’, and refers to the lack of rain

(HALOT #4241). A comparable use of this root is found

in Gen. 8: 2 wayyikk�al�e(ʾ) hag-ge�sem min-ha�s-�s�am�ayim

‘and the rain from the sky was restrained’. In this inscrip-

Table 1. The seasons in the Safaitic inscriptions compared to the

Rwala.

Seasons Safaitic Rwala Time period

—— —— as:

-s:

feri early October–early January

the winter s²ty a�s-�sta early January–mid-February

the season of

the later rains

dṯʾ as-sm�ak mid-February–mid-April

the early summer s:

yf as:

-s:

eyfi mid-April–early June

the dry season qyẓ al-q�eẓ early June–early October

Table 2. The division of the year by season.

Seasons Safaitic Rwala Time period

the winter s2ty as:

-s:

feri early October–early January

a�s-�sta early January–mid-February

the season of

the later rains

dṯʾ as-sm�ak mid-February–mid-April

the dry season qyẓ (including the

transitional s:

yf)

as:

-s:

eyfi mid-April–early June

al-q�eẓ early June–early October

3 These are WH 2327, 3500 and KRS 2282. The fact that all the

other seasons are attested frequently, including s²ty and dṯʾ,

makes it unlikely that the authors of the Safaitic inscriptions

were only absent during s:

yf.

4 Ryckman’s Latin is: Zaydoʾilo, filio Naghafati, filii ʾaʿlayi, filii

Daḥiyi filii ḥuw�aq�ı, filii Kuwaynati, filii �saww�a �ʾı. Et versabatur

in cavernis Malih et Dhakir et Amat. Et disperse sunt (?) cameli

in deserto. Et seminavit in Ruhbat. Et, O Ilat, salutem, et vindic-

tam ab illo qui oblitteraret.

215

ANCIENTARABIAN ZODIAC

Page 3: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

tion, the verb has an active sense: ‘and the sky withheld/

restrained it’. The third person pronoun referring to the

rain probably assimilated to the article: */wa kallaʾ-ah-

has-sam�ay/, which is also common in the inscriptions.5

This reinterpretation fits well with the fact that the

author ‘despaired’, yʾs¹, and then asked the rain-god,

Bʿls¹mn, for ‘relief’ rwḥ so that he would remain

secure, w s¹lm.6

Both C and KRS took mlḥ, ḏkr and ʾmt as toponyms.

These terms also occur independently, usually as an object

of the preposition b- or following the terms rʾy and ks¹ʾ.

For example:

AbaNS 1117

l s¹lm bn s:

bḥ w s:

yr b- ʾmt f h rḍw _gnmt

By S¹lm son of S:

bḥ and he returned to permanent water

b- ʾmt so, O Rḍw, let there be booty

KRS 1267

l khl bn ẖl bn ʾml w tẓr my b- ḏkr

By Khl son of Hl son of ʾml and he awaited rain b- ḏkr

C 523

l ʾlh bn bʾẖh bn trb w wrd h- nmrt b- ks¹ʾ mlḥ

By ʾlh son of Bʾhh son of Trb and he came to water at

Nam�arah b- ks¹ʾ mlḥ

C 2549

w ḥll h- ʿrḍ b- r{ʾ}y ḏkr nwy

and he camped in this valley b- rʾy ḏkr while migrating

with the tribe

The term rʾy has been the subject of much debate. Litt-

mann (1943: 117) suggested that rʾy signified a watering

place, in connection with the root √rwy, in his translation

of the phrase b- rʾy gml as ‘a watering-place of camels’.

While some have accepted this identification, faute de

mieux, it remains unclear how rʾy should be derived from

the aforementioned root or why it should signify a noun

of place.7 Jamme rejected this interpretation and instead

suggested the meaning ‘in sight of’ or ‘in view of’ on the

assumption that the terms ḍkr, mlḥ and ʾmt, which often

follow rʾy, were toponyms (1971: 34–35). I agree with

Jamme that rʾy as a reflex of √rwy is unlikely, but I do not

see how his translation renders a better interpretation of

these constructions. Ks¹ʾ, on the other hand, has not

enjoyed as much discussion. Most have taken it to mean a

‘track’ or ‘way’, based on a rather extreme semantic leap

from its Classical Arabic counterpart kasaʾa ‘he, or it, pur-

sued or followed another’ (Lane 2608c). If we consider,

however, the fact that mlḥ, ḏkr and ʾmt occur together in

the same order in both C 4985 and KRS 169, in light of

the sequences of seasons in Mu 113 and SIJ 1008, a better

explanation presents itself: these terms may refer to peri-

ods of time. Based on their etymologies, I would suggest

that they signal the zodiacal constellations, and if we con-

sider the time period in which the sun passed through

these constellations roughly 2000 years ago, it seems clear

that they were used to demarcate the three major seasons

attested in the Safaitic inscriptions.8

Before discussing their etymologies, a short excursus

on the apparent interaction of the sun and moon with the

zodiacal constellations is in order. This will cast light on

the meanings of b-, rʾy and ks¹ʾ. The earth’s orbital path

causes the sun’s diurnal arc to shift across the sky over

the period of a year. The series of constellations which

lies on the sun’s annual path, the ecliptic, is called the

‘zodiac’. The sun is said to be in a given constellation

when it rises in the segment of the sky occupied by that

constellation. The ecliptic constellations were identified

by the Babylonians, where they were a part of a larger

celestial coordinate system of seventeen constellations, of

which not all were ecliptic (Koch-Westenholz 1995:

132). By the fifth century BCE, the twelve-part zodiac

had emerged. This system divided the sky into twelve

30° segments, called ‘signs’, each associated with a par-

ticular ecliptic constellation. Since the ecliptic constella-

tions vary in size, the amount of time the sun spends in5 The word for sky, sam�ay, can be treated as both masculine and

feminine in the inscriptions. A parallel expression is HaNSB

119: ʿgz h- s¹my ‘and the sky withheld it [i.e. the rain], which

also occurs in the feminine HaNSB 16 w ʿgzt h- s¹my kll h- s¹nt

‘and the sky withheld it [i.e. the rain] the entire year’. On word

boundary assimilation, see Macdonald (2004: 507).6 Result clauses are formed either by w or l and the optative use

of the suffix conjugation or the prefix conjugation (see

Al-Jallad, forthcoming: §9.5.3).7 An active participle formation C�aʾiC seems possible, but in this

case we should not expect the final vowel to have been indicated

in the orthography; an underlying r�a �ʾı would appear as rʾ and

not rʾy. Moreover, a noun of place rather than an active partici-

ple pattern would have been expected.

8 The only reference in the literature I could find which considers

an alternative explanation for any of these terms is Ababneh’s

interpretation of AbaNS 1117, where he states the following

about ʾmt: ‘It seems more possible to identify ʾmt as a month

name (i.e. with astronomic sense; [sic] e.g. a star name, or one

of the lunar phases). . .’. Unfortunately, his explanation is rather

convoluted, as a month name, a star name and a lunar phase are

rather different things. Moreover, it is unclear how he came to

such a conclusion, especially since he provides no etymological

justification for his interpretation. Finally, since he accepts Jam-

me’s translation of b rʾy as ‘in sight of’, it is unclear how this

would configure with ʾmt being a month.

216

A. AL-JALLAD

Page 4: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

each is not uniform. Thus, the zodiac signs sometimes

correspond only roughly with their namesakes.9 The

moon’s path also follows the ecliptic and appears to pass

through the constellations. The moon completes a circuit

on the ecliptic every month, and the full moon will

appear in each constellation/sign once a year.

The zodiacal constellations also appear to rise and set.

When the sun occupies the same portion of the sky as a

constellation, its stars cannot be seen at dawn or dusk

because of the sun’s light. Once the sun has passed on to

the next constellation, however, the stars of the previous

one will become visible on the eastern horizon, rising

slightly before dawn. This is called the heliacal rising of a

constellation. Roughly half a year later, the constellation

will experience an acronical rising, whereby its stars

appear for the first time in the evening sky opposite the

position of the sun.10

In light of our new understanding of mlḥ, ḏkr and ʾmt, it

seems more likely that the terms rʾy and ks¹ʾ have to do

with the behaviour of celestial bodies. The heliacal rising

of the stars/constellations interested Babylonian astrono-

mers, to which they applied the term IGI = ittanmar, ‘to rise

heliacally’ (Reiner 1981: 16–17), from am�arum ‘to see’.

Safaitic rʾy corresponds well with this. There are certain

cases, however, in which rʾy must be taken as a reference

to the acronical rising of a constellation. I would therefore

suggest simply translating the term as ‘the rising’ and

determining whether it referred to heliacal or acronical

rising based on its context. The term ks¹ʾ is comparable to

CAr kusʾun ‘the latter part of a month; its last ten days, or

about that period’ (Lane 2608), Hebrew k�ıse and Syriac

kes�aʾ, both ‘full moon’ (HALOT #4329). Since we have

no evidence to suggest that the month was divided into

three ten-day portions11 or that the terms of the zodiac

even refer to months, I would suggest that Safaitic ks¹ʾ

refers to the full moon when it occupies the constellation/

sign with which it is in construct. Full moons occur when

the moon rises directly opposite the sun, and so we can

roughly determine the time of year to which an inscription

containing ks¹ʾ before the name of a zodiacal constellation

refers. Finally, I would suggest that the preposition b- sig-

nifies the period of time in which the sun rises in a given

constellation.

The question as to whether mlḥ, ḏkr and ʾmt refer to the

actual constellations, to a 30� segment of the sky (i.e. a

sign) or had simply become names for the months

remains. Only a single published Safaitic inscription

attests a clear reference to a month name:

KRS 1965

l ghm bn zhrn bn ʿḏ bn tm bn ʿḏ w ts²wq ʾ[l-] ḥbb b-

s²hr s¹bt f h lt qbll

By Ghm son of Zhrn son of ʿḏ son of Tm and he longed

{for} a friend in the month of S¹bṭ, so, O Lt, show benevo-

lence

‘S¹bṭ’ obviously renders the winter month �səb�aṭ, and

indicates that at least this author was using a version of the

Babylonian calendar. If he is not an anomaly, then it would

appear that the Babylonian month names were in use

among the nomads, suggesting that the zodiacal terms had

not become names of the months. It remains unclear, how-

ever, whether an ecliptic coordinate system consisting of

twelve 30° parts was in use or whether the terms referred

to the constellations themselves. The fact that the authors

of these inscriptions were clearly concerned with heliacal

and acronical risings supports the latter interpretation.

Indeed, there is nothing in the inscriptions that would

require the existence of a system of zodiac signs.

The identification of the constellations is made primar-

ily on etymological grounds, with due consideration to

each term’s context within an inscription. Given this limi-

tation, there is no way to be sure that the Arabian constel-

lations comprised the same stars as their classical

counterparts. While these issues will be dealt with in more

detail in Part II of this paper, they directly affect the way

we interpret the chronological dimension of the astronomi-

cal texts. Variation in the size of the constellations would

influence the amount of time the sun would have spent in

each, challenging our ability to assign accurate dates to the

various celestial references. Moreover, the time of year the

sun spent in each constellation would have shifted by a

few days over the estimated period of Safaitic epigraphic

activity (first century BCE–fourth century CE) due to the

phenomenon of axial procession. Nevertheless, I think it is

safe to assume that the constellations correspond to their

9 Due to the phenomenon of axial procession, the zodiac signs

today do not correspond to the constellations on which they

were based. The sign Aries, for example, corresponds to the

constellation Pisces. Since the Safaitic inscriptions were

produced in the neighbourhood of 2000 years ago, however,

the discrepancy between sign and constellation was much less

pronounced. For further reading, see Kelley & Milone 2005:

26–27.10 Heliacal setting refers to apparent setting of a constellation with

the sun slightly before the sun’s light renders it invisible. The

acronical setting occurs when the stars of a constellation set

opposite the sun at dawn. Neither of these appear to be men-

tioned in the Safaitic inscriptions.11 This division was in use in Ancient South Arabia; see Stein

2005 for further discussion.

217

ANCIENTARABIAN ZODIAC

Page 5: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

classical counterparts, but to assign rough dates to the dif-

ferent astronomical events (i.e. early April for the begin-

ning of Aries rather than a specific date). Table 3 lists the

comparable incarnations of the zodiac along with the dates

of both the signs and the position of the sun. While the

fixed dates of the zodiac signs are about a month removed

from the constellations today, the difference would have

been negligible some 2000 years ago. The astronomical

dates estimate the period in which the sun would have

passed through each constellation in the second century

CE.12

3. The constellations

Aries = ḏkr

The Safaitic term corresponds perfectly with the Ara-

maic ארכד , but a loan can probably be ruled out based on

the preservation of the initial interdental. The use of this

root to signify a ram is unknown in CAr, where ḏakar has

a more general sense, referring to the male of a species as

opposed to the female (Lane 970a). The origins of the ram

symbol, replacing the original ‘hired farm-labourer’ is

unclear and continues to be debated (Rogers 1998:19).

The sun would have passed through Aries during the

month of April, placing it at the end of dṯʾ.

Libra = ʾmt

This word is probably related to the root √ʾmt, which

means in CAr to ‘measure or compute quantity’ (Lane

94c), and renders it a suitable match for Libra, the scales.13

This meaning, however, was not known to the Arab lexi-

cographers. In the earliest periods, the stars of Libra con-

stituted the claws and tail of Scorpio, but by the ‘three

stars each’ period (c.1100 BCE), Libra was recognised as

its own constellation (Rogers 1998: 24–25). An echo of its

former unity with Scorpio, however, is preserved in a sin-

gle inscription (see the discussion under ʿqbt). The sun

passes through Libra from mid-September to mid-October,

which corresponds with the end of qyẓ, and therefore sig-

nifies the end of the dry season.

Aquarius = mlḥ

The connection of mlḥ with Aquarius can be made on

the basis of analogy with the previous two signs, namely,

that they correspond to the final constellation of their

Table 3. The Zodiac.

Latin English Greek

Aramaic

(Qumran) Babylonian Sign dates Astronomical dates

Aries Ram Κqιός ארכד the Hireling MUL(-LU2)ḪUN-GA2 agru mid-March–mid-April early April–late April 25 days

Taurus Bull Τaυqος ארות Bull of An MULGU4-AN-NA

(mul.mul) alu/is le

mid-April–mid-May late April–late May 37 days

Gemini Twins Dίdυlοι אימואת the Great Twins MULMA�S-TAB-BA-

GAL-GAL m�a�su/t�uʾ�am�u rabutu

mid-May–mid-June late May–late June 31 days

Cancer Crab Κaqj~ımος הנטרס the Crab MULNAGAR alluttu mid-June–mid-July late June–early July 20 days

Leo Lion Λέxm הירא the Lion MULUR-A/ MUL

UR-GU-

LA n�e�su / urgulu

mid-July–mid-August early July–mid-August 37 days

Virgo Maiden Πaqhέmος הלותב the Furrow MULKI absinnu mid-August–mid-September mid-August–late September 45 days

Libra Scales Ζυcός אינזומ the Scales GI�S-ERIN2 zib�an�ıtu/zib�anu mid-September–mid-October late September–late October 23 days

Scorpio Scorpion Σjοqpιός אברקע the Scorpion MULGIR2-TAB zuqiq�ıpu mid-October–mid-November late October–late October 7 days

Ophiucus* late October–mid-November 18 days

Sagittarius Archer Τοξόsgς אתשק Pabilsag MULPA-BIL-SAG Pabilsag mid-November–mid-December mid-November–mid-December

32 days

Capricorn Goat-horned Αἰcόjeqxς אידג the Goat-Fish MULSUḪUR.

MA�S2ku6 suḫurm�a�su

mid-December–mid-January mid-December–mid-January 28 days

Aquarius Water-bearer Ὑdqοvόος אלוד the Great One MULGU-LA rammanu mid-January–mid-February mid-January–mid-February 24 days

Pisces The fishes Ἰvhύeς אינונ the Tails mulKUN

ME�S zibb�atu mid-February–mid-March mid-February–early April 38 days

*Ophiucus is the 13th classical constellation that crosses the ecliptic, but it was omitted from the system of zodiacal signs.

Note: The Aramaic names come from the Palestinian Jewish Aramaic text studied by Greenfield and Sokoloff (1989), which is the earliest Aramaic

source containing the names of the Zodiac. The Babylonian zodiac is discussed in Koch-Westenholz 1995: 164; the Akkadian names are from the

Chicago Assyrian Dictionary and Schramm 2010.

12 The current astronomical dates are generally a month later. The

calculation of these dates, along with the heliacal and acronical

rising of each constellation was done by using the software

STELLARIUM © based on the view from the Ḥarrah of northern

Jordan in and around 150 CE.

13 I initially thought ʾmt translated as Virgo ‘the maiden’, but

M.C.A. Macdonald pointed out that the term better fits Libra,

as ʾmt only has the sense of ‘female slave’ in Arabic and other

Semitic languages. I thank him for this suggestion, which is

incidentally a better fit for the division of seasons suggested at

the beginning of this paper.

218

A. AL-JALLAD

Page 6: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

respective seasons. The remaining season, the winter s²ty,

ends in mid-February, and therefore corresponds to Aquar-

ius. This connection can also be supported on etymologi-

cal grounds, although the relationship is not as transparent

as with the previous two constellations. Aquarius was

originally depicted as a water-pourer, whom the MUL.

APIN called ‘the great one’. This figured passed on to the

Greeks, who left him unnamed and depicted him with a

single vase pouring water onto Piscis Austrinus (Rogers

1998: 25). The Semitic root mlḥ refers primarily to salt,

and in Arabic, it can signify water that has become salt.

The term mall�aḥ refers to a sailor or seaman, which puts it

in the same semantic field as ‘water bearer’, but this is still

not very close. I would instead suggest that mlḥ refers to

the vessel in which salt was carried, or perhaps even to a

salt worker. The Safaitic inscriptions contain references to

transporting salt, and this could have been a more relevant

image to their authors than that of a ceramic vase.14 While

it is impossible to know for sure how they transported salt,

in many traditional societies today this is achieved by car-

rying it in large baskets or flat pans, which are sometimes

placed on the head or shoulders. Such an image could

have been the basis for the name of this constellation. On

the other hand, an interesting connection between a water

bearer and salt is found in Euting’s Tagebuch einer Reise

in Inner-Arabien (1896: 73–74, 102–103). There, he

describes a scene near K�af in W�ad�ı Sirḥ�an in which a slave

drew water from a well in a saltpan and then poured it on

the ground in order to leach out the salt. Thus, mlḥ could

also refer to the role played by the water bearer in the

manufacture of salt.15

From this, we can reconstruct the following correspon-

dences and translate our inscriptions accordingly

(Table 4).

Let us now turn to the inscriptions that contain these

three terms together:

C 4985

l zdʾl bn ngft bn ʾʿly bn ḍḥy bn ḥwq bn kwnt bn s²wʾ w

ḥḍr ẖlqt mlḥ w ḏkr w ʾmt w s¹lḥ h- ʾbl mdbr w zrʿ h- rḥ

(—)bt f h ʾlt s¹lm w nqʾt l- ḏ yʿwr

By Zdʾl son of Ngft son of ʾʿly son of Ḍḥy son of Ḥwq

son of Kwnt son of S²wʾ and the sun was present in the

portions of Aquarius, Aries and Libra [i.e. an entire year],

during which he pastured the camels on slḥ-plants in the

inner desert and sowed this Rḥbt so, O ʾlt, let there be

security and a curse upon him who would efface

Commentary

The author recounts the activities of his year through

the use of two parallel constructions. He begins by stating

that an entire year has gone by: ‘the sun passed through

the portions of Aquarius, Aries and Libra’ and then, in a

parallel circumstantial clause introduced by w, mentions

two activities which must have taken place at different

times of the year: ‘pasturing the camels in the desert and

sowing the Rḥbt’.

ḥḍr: in most cases, ḥḍr should be interpreted as ‘he

camped near permanent water’ (Lane 590b). This interpre-

tation does not fit the context, however, as it would imply

that the author camped at permanent water while also pas-

turing the camels in the inner desert, and sowing a shallow

depression where water collects (the Rḥbt). Instead, I

would take ḥḍr in contexts such as these to refer to the

‘presence’ of the sun, in line with the basic sense of this

root (Lane 588c). Grammatically, ḥḍr is probably a suffix-

conjugated verb. While one would expect feminine

concord with ‘sun’, agreement in Safaitic is erratic, and

several nouns that are feminine in other Semitic languages

take masculine concord in Safaitic (see e.g. s¹my below).

hlqt: the root √hlq has the sense of ‘to divide, measure,

or portion’ (Lane 799c; HALOT # 2919), and here I would

suggest that it refers to the portion of the sky occupied by

or allotted to each of these constellations. Thus, the phrase

ḥḍr hlqt would mean ‘it [the sun] was present in the por-

tions of (Aquarius, Aries and Libra)’ or in other words, the

sun made its yearly journey, passing through the last con-

stellation of each season.

s¹lḥ: the primary definition of Arabic s¹lḥ ‘to void

excrement’ is not appropriate in this context, nor is the

D-stem ‘to arm with a weapon’ (Lane 1402a–b). Instead, I

would interpret this as a denominal verb meaning ‘to pas-

ture on ʾisl�aḥ’ a type of plant which is said to cause the

milk of the camels to become abundant, or it can refer to a

type of herb growing on tracts of sand (Lane 1402c–

1403a). The latter meaning is especially suitable in this

context, as the author pastured in the inner desert.

nqʾt: the etymology of this term remains unknown, but

it surely signifies something negative as it appears

exclusively in curses. I have therefore chosen to translate

it neutrally as ‘a curse’.

KRS 169

l nʿrt bn ẖdmt bn trml bn s¹ry bn s¹hm w _gz b- nẖl f h yṯʿ

s¹lm w klʾ h- s¹my ḥḍ{r} mlḥ f ḏ{k}{r} f ʾmt f yʾs¹ f h

bʿls¹mn r{w}ḥ w s¹lm

14 E.g. HaNSB 184 w mlḥ f ʿny kbr ‘and he traded salt and toiled

continuously’. The vessel for carrying water would have also

been a leather sack in the desert, and not a ceramic jug.15 I thank M.C.A. Macdonald for this reference.

219

ANCIENTARABIAN ZODIAC

Page 7: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

By Nʿrt son of Hdmt son of Trml son of S¹ry son of S¹hm

and he raided this valley, so, O Yṯʿ, may he be secure; and

the sky withheld its rain during [the sun’s] presence in

Aquarius, then Aries, then Libra, and so he despaired, so O

Bʿls¹mn, let there be relief that he remain secure

or put another way:

. . . and the sky withheld its rain in the winter, then the

season of the later rains, and then the dry season, so he des-

paired, so O Bʿls¹mn, let there be relief that he remain

secure

Commentary

_gz b- nhl: the verb ‘to go on a raid’ usually appears as

_gzz, which must be a D-stem of a geminate root √ _gzz. This

inscription may then attest a G-variant of this root, or may

simply be the result of a writing error.

ḥḍr: the term ḥḍr is also attested in KRS 169, where it

can be understood in two ways. First, one could take it as

an infinitive, acting as an adverb of circumstance in con-

struct with the three constellations mlḥ f ḏ{k}{r} f ʾmt thus,

‘and the sky withheld [the rain] while camping near per-

manent water in Aquarius, then Aries and then Libra’ or it

could also refer to the sun’s position, ‘the sky withheld

[the rain] during the [the sun’s] presence in Aquarius,

Aries and Libra’.

Thus, in these contexts, the three zodiacal constellations

would signal the end of each of the three seasons men-

tioned in the Safaitic inscriptions (Table 5).

Another inscription attests a sequence of mlḥ and ḏkr,

which also nicely fits our understanding of the time

periods which they reference.

RWQ 326

l ʾnʿm bn ḥmṭn bn ʾs¹d bn mtʿ bn s¹md bn ẖlf bn ʾs¹ḥm

bn ʿzz bn hrb bn hṯlt bn h:

gg bn nʿmʾl bn ʿm bn ybʾs¹ w ʿlf

mlḥ f ḏkr nbṭ f yʾs¹ m s¹my. . .

By ʾnʿm son of Ḥmṭn son of ʾs¹d son of Mtʿ son of

S¹md son of Hlf son of ʾs¹ḥm son of ʿzz son of Hrb son of

Hṯlt son of Ḥgg son of Nʿmʾl son of ʿm son of Ybʾs¹ and

he fed [his animals] on dry fodder in Nabataea in Aquarius

[the winter] then Aries [the season of later rains] and so he

despaired on account of the sky [i.e. the weather]. . .

Commentary

ʿlf: the verb ʿlf should be connected with Classical

Arabic ʿalafa ‘he fed (the beasts) on fodder’. While the

object is absent here, it is attested in other inscriptions,

for example, LP 722 ʿlf h- mʿzy s¹nt bʾs¹ ‘he fed the

goats on dry fodder in the year of misery’. The need to

feed animals on dry fodder during the rainy seasons of

s²ty and dṯʾ suggests a drought, and explains why the

author expresses despair, yʾs¹, on account of the

weather, m s¹my.

nbṭ: locative nouns in Safaitic are often unmarked, and

so nbṭ can be construed as ‘in Nabataea’, and probably

refers to an area under Nabataean control. It is less likely,

although possible, that the author was describing the Na-

bataeans feeding their animals on dry fodder.

m s¹my: one would expect the article h- or ʾ- to precede

s¹my in this construction. The elision of the article’s onset

in this position is unlikely, and so its omission is probably

a writing error.

3.1. Inscriptions containing ḏkr, mlḥ, ʾmt

independently

Now that we have established the identity of these three

terms, let us turn our attention to their occurrences follow-

ing b- and ks¹ʾ / rʾy:

ḏkr:

KRS 1267

l khl bn ẖl bn ʾml w tẓr my b- ḏkr

By Khl son of Hl son of ʾml and he awaited rain when

the sun was in Aries [April, the end of dṯʾ]

Commentary

my: while my is clear on the photograph and fits the

context well, it should also be noted that the expression tẓr

mny ‘he awaited fate’, the exact sense of which is not

clear, is common and that the author in this case may have

simply forgotten to carve the n. It is also possible that

s¹my was also intended, but this word is usually preceded

Table 4. The identification of ʾmt, ḏkr, mlḥ.

Latin English Greek Aramaic Babylonian Safaitic

Aries Ram Κqιός ארכד Agrarian worker agru ḏkr ‘ram’

Libra Scales Ζυcός אינזומ The Scales zib�an�ıtu ʾmt ‘measure’

Aquarius Water-bearer Ὑdqοvόος אלוד the great one rammanu mlḥ ‘salt- bearer’

Table 5. The constellations and the seasons.

Seasons Safaitic Rwala Time period

the winter s²ty mlḥ as:

-s:

feri early October–early January

a�s-�sta early January–mid-February

the season of

the later rains

dṯʾ ḏkr as-sm�ak mid-February–mid-April

the dry season qyẓ ʾmt as:

-s:

eyfi mid-April–early June

al-Q�eẓ early June–early October

220

A. AL-JALLAD

Page 8: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

by the article h- or ʾ- and would, at any rate, have had the

same meaning as my in this context, namely, to await the

rains.

KRS 2851

l g{d}d bn —m—m bn ʿbd w rʿy h- nẖl b- ṯlg b- {k}s¹ʾ

ḏkr

By {Gdd} son of {—m—m} son of ʿbd and he pastured

in this valley on snow during the {full moon} of Aries

(mid-October)

Commentary

The editor interpreted the last word as {m}s¹ʾ, but the

first letter looks more like a poorly formed k. The full

moon occupies Aries around mid-October, which is rather

early for snowfall, but then again it may be this very fact

that motivated the author to mention it in an inscription.

C 4973

l ʿmd bn ms¹k bn ʿmd bn mlk w wld h- mʿz {w} rʿy -h

bql b- ḏkr

By ʿmd son of Ms¹k son of ʿmd son of Mlk and he

helped the goats to give birth {and} pastured them on

spring herbage when the sun was in Aries.

mlḥ:

C 523

l ʾlh bn bʾẖh bn trb w wrd h- nmrt b- ks¹ʾ mlḥ

By ʾlh son of Bʾhh son of Trb and he came to water at

Nam�arah during the full moon of Aquarius.

Commentary

The full moon sits in Aquarius when the sun rises in

Leo, in the summer.

C 3230

{l} ( ) ʿm bn ḥy w rʿy {h-} ʾbl b- rʾy mlḥ ʿl- [h-] nẖl w

wgm ( ) [ʿ][l-] ʿm {-h} —

By ʿm son of Ḥy and he pastured the camels during the

rising of Aquarius over (?) [the] valley; and he grieved

[for] {his} grandfather

Commentary

The inscription does not contain enough information to

determine whether rʾy refers to acronical or heliacal rising.

The phrase ʿl [h-] nhl could also be taken as ‘against/bor-

dering the valley’.

WH 479

l s¹{w}r bn ẓhln w wrd b- mlḥ hm w brkt f flt:

rḍw

By {S¹wr} son of Ẓhln and he went to water at Hm and

Brkt during Aquarius (mid-January–mid-February) so,

may Rḍw deliver

Commentary

The fact that the author went to permanent water in the

middle of the winter may suggest the presence of a

drought. This seems to be suggested by the fact that he

went to two places to water, perhaps indicating that the

water supply was diminished.

brkt: Macdonald (1992b: 37–38) convincingly identi-

fied this as the name of an area in the seasonal migrations

of the nomads.

ʾmt:

KRS 896

l ys¹lm bn grmʾl bn qḥs² w wgd s¹fr s:

ʿd f ngʿ w wrd h-

nmrt b- ʾmt ʾẖrt

By Ys¹lm son of Grmʾl son of Qḥs² and he found the

inscription of S¹ʿd, so he grieved in pain and went to water

at Nam�arah when the sun was in Libra [the dry season]

thereafter

Commentary

ngʿ: this verb is used frequently in grieving formulae

and reflects the N-stem of the root √wgʿ ‘to be in pain’.

This derivation is supported by the fact that the verb wgʿ

itself is used in the same grieving context; compare, for

example, WH 946 w wgʿ ʿl _gyrʾl with WH 960 w ngʿ ʿl

_gyrʾl, both ‘and he grieved in pain for _Gyrʾl’.

ʾẖrt: this term is attested only once more in the Safaitic cor-

pus, in KRS 1444 where it clearly means ‘offspring’. Such a

meaning seems impossible here. I would suggest that it is an

adverb of time, meaning ‘thereafter’ related to the basic sense

of the root √ʾhr ‘last, after, following’ (Lane 31–32).

AbaNS 1117

l s¹lm bn s:

bḥ w s:

yr b- ʾmt f h rḍw _gnmt

By S¹lm son of S:

bḥ and he returned to permanent water

when the sun was in Libra so, O Rḍw, let there be booty

KRS 1770

l ʾḏnt bn ʾs¹ bn kmd bn ʾs¹ w wrd f nyt {b-} ʾmt{n} f rwḥ

h b<ʿ>ls¹mn

By ʾḏnt son of ʾs¹ son of Kmd son of ʾs¹ and he went to

water then migrated with the tribe while the sun was in

Libra, so let there be relief, O Bʿls¹mn

Commentary

nyt: this is the infinitive of nwy ‘to migrate (with the

tribe)’, which is used here in an infinitive chain construc-

tion (see Al-Jallad, forthcoming: §9.4.1), where only the

first verb of a sequence of related actions is finite.

ʾmt{n}: the n of this inscription is slightly higher than

the rest of the letters and is connected to the following f,

which may indicate that it was unintentional. If it was

intentional, then it would reflect a rare occurrence of nuna-

tion in Safaitic; another possible example is attested in

KRS 1551, mḥltn ‘dearth of pasture’.

C 4903

l knt bn zʿq s¹+<<t>>+{n}{t} ms¹ʿ {ʾ}s¹y h- rḥbt b- ʾmt

h- dṯʾ

221

ANCIENTARABIAN ZODIAC

Page 9: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

By Knt son of Zʿq in the year. . . this Rḥbt during the

Libra of the season of the later rains

Commentary

This text must refer to the acronical rising of Libra,

which occurs in April, that is, the Libra of the spring. The

interpretation of the ms¹ʿ ʾs¹y is unclear. The Arabic dictio-

naries know ms¹ʿ as one of the names of the North Wind

but no definition of ʾs¹y √ʾsw/y or √ʾws is suitable for this

context. I have therefore left it untranslated.

3.2. The other zodiacal constellations that follow rʾy

and ks¹ʾ

There are six other terms preceded by rʾy and ks¹ʾ and it

stands to reason that these too refer to periods of time indi-

cated by the heliacal rising and the position of the sun and

moon with respect to a given zodiacal constellation.

ḏl = Pisces

The term ḏl does not correspond to the fish of the Greek

or West Semitic zodiac, but is a better match for the Baby-

lonian name zibb�atu ‘tails’, as encountered in the MUL.

APIN (Rogers 1998: 27). The Safaitic ḏl is cognate with

CAr ḏaylun, which means ‘tail’, or more generally, ‘the

latter, hinder, or last part of anything’ (Lane 990c). Only

one inscription known to me contains this term:

LP 712

l tʾm bn s¹ʿd bn ḏr w wrd b- rʾy ḏl {y}s²rb

By Tʾm son of S¹ʿd son of Ḏr and he came to water dur-

ing the rising of Pisces to drink

Commentary

The prefix conjugation in result clauses is usually intro-

duced by w or l. Its absence in the orthography in this

inscription suggests the writing of two identical letters at

word boundaries as if they were one and, further, the loss

of final short vowels, thus: */bereʾ�ayḏaylleya�srab/.

yʾmr = Capricorn

This term is cognate with CAr yaʾm�urun which is

defined by the lexicographers as ‘a kind of mountain-

goat’ or ‘a certain wild beast’, ‘a beast resembling a

goat, having a single branching horn in the middle of

its head’ and ‘a certain beast of the sea’ (Lane 99a).

The uncertainty with which the dictionaries treat this

word may attest to its antiquity, and perhaps to the fact

that it referred to a mythical creature rather than an

actual animal. Indeed, that its definition encompasses

both a goat and a sea creature corresponds nicely with

the Babylonian goat-fish hybrid.

HaNSB 197

l hnʾ bn wrd bn s²hyt bn ʾs¹ w ʾs²rq m- ḥrn b- ʾbl -h s²ʿr

b- rʾy yʾm{r} f h bʿls¹mn _gyrt w s¹lm w ʿwr m ʿwr

By Hnʾ son of Wrd son of S²hyt son of ʾs¹ and he

migrated to the desert from the Ḥrn with his camels to

spring herbage during the heliacal rising of Capricorn

[early-mid February], so, O Bʿls¹mn, let there be abun-

dance and security, and blind whosoever would efface [the

inscription]

Commentary

The acronical rising of Capricorn takes place in July,

making it an unlikely period to find herbage in the desert.

C 4276

l s¹ẖr bn ʾws¹ʾl bn s¹ẖr w ẖlt:

mn ẖyl b- rʾy {y}ʾmr f rgʿ

hʾs¹ m- rql s¹nt s¹rt hʾs:

f rqln m- ẖl (—)

Commentary

C reads the second y as {h}, but given the context, this

is probably the result of a copy error or damage on the

rock. Even with this restoration, the exact sense of the

inscription remains elusive as it contains several hapax le-

gomena and is unformulaic. I am unable to offer a satisfac-

tory interpretation at this time.

RWQ 290

l gld bn— h bn s¹gʿw w wrd ʾdyt h- bly b- ks¹ʾ yʾmr

By Gld son of — son of S¹gʿw and he went to water in

the valleys of the camel-graves when the full moon was in

Capricorn

Commentary

Interestingly, in his commentary on this inscription,

al-R�us�an identified ks¹ʾ as a month name, but took the

first y of yʾmr as part of the word, ks¹ʾy, which he

equated with ʾayl�ul. He then left the final three letters,

ʾmr, unexplained.

ʾdyt h- bly: ʾdyt probably renders */ʾawdeyat/, the plural

of wadi ‘valley’. The second term bly is found in a few Sa-

faitic inscriptions (e.g. WH 163, 165), and refers to a par-

ticular form of a grave where the dead man’s camel was

hamstrung and left to die either beside the grave or in a pit

next to it, possibly to provide a suitable mount in the after-

life (Macdonald 1994: 762). A Nabataean burial inscrip-

tion makes reference to a blwʾ, which Hayajneh (2006)

connected with Safaitic bly and indeed Classical Arabic

baliyyah.

Virgo = h-ngm

Scholars have always assumed that if ngm had an astro-

nomical referent, it would be identical to CAr na�gam, the

Pleiades. Its appearance simply with the preposition b-

suggests, however, that it refers to one of the twelve zodia-

cal constellations. Again here we find a better connection

with the Babylonian rather than the Greek and the Qumr�an

Aramaic zodiacs. The Babylonian sign depicted the god-

dess Shala holding a sheaf of corn. While the Greek

222

A. AL-JALLAD

Page 10: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

zodiac focused on the image of the Shala, the virgin, the

ear of corn became the symbol of this constellation in the

eastern zodiacs, �sebelt�a in Syriac, h�o�sag in Pahlavi and

indeed kany�a in Sanskrit (Laffitte 2003: 115). The Safaitic

sides with the latter group. I would connect ngm with CAr

root √n�gm, which refers to ‘the emergence’ or ‘appear-

ance’ t:

alaʿa/ẓahara, of—among other things—a ‘plant’, ‘a

canine tooth’, ‘a horn’ or ‘a planet’ (Lisan s.v.). As such,

h-ngm may be a nominal form referring to that which the

seed-produce put forth, similar indeed to the root √s¹nbl,

which gives: sanbala z-zarʿu ‘The seed-produce put forth

its sunbul [ears]’ as well as a nominal form sunbul (Lane

1440b). Classical Arabic na�gm (Lane 3028c) refers to the

‘sprouts from the roots [of a tree or shrub]’ or to triticum

repens, couch or dog grass. If I am correct in identifying

this term with Virgo, then the original sense of this root

must have been wider. The reason why the name of this

constellation sometimes occurs with the definite article is

at present unclear.

HaNSB 218

l ḍhd bn nr bn ḥrb bn ḍhd bn kṯbt bn ḥmyn w wgm ʿl-

s¹lm w ʿl- _gyr w ʿl- ʾʿdg w ʿl- ʿwḏn w ʿl- bʿl w ʿl- ḥny w ʿl-

s¹ny f wny w ngs² h- ʾbl m- ḥrn mn ṯlg b- rʾy ngm

By Ḍhd son of Nr son of Ḥrb son of Ḍhd son of Kṯbt

son of Ḥmyn and he grieved for S¹lm and for _Gyr and for

ʿwḏn and for Bʿl and for Ḥny and of Sny and so became

weak; and he drove the camels from the Ḥawr�an on

account of snowfall during the rising of Virgo

Commentary

The heliacal rising of Virgo takes place in early Octo-

ber, which seems a little early for snowfall, although

note the snowfall in October in KRS 2851. Virgo’s ac-

ronical rising occurs in late February, when snow would

also be rare, but still possible. As with KRS 2851, the

explicit mentioning of snow may indicate that it was a

remarkable event at the time of year to which the

inscription refers.

CSA 2.1

l hmlk bn ʾlh bn s¹ny bn hmlk bn ḥrb w rʿy mdbr {f} s:

yr

b- h- ngm f h lt s¹lm

By Hmlk son of ʾlh son of S¹ny son of Hmlk son of Ḥrb

and he pastured in the desert, {then} he returned to perma-

nent water while the sun was in Virgo, so, O Lt, may he be

secure

WH 3053

l dʾy bn ns²l w rḥḍ b- h- ngm l- yḥg

By Dʾy son of Ns²l and he washed while the sun was in

Virgo in order to perform a pilgrimage

ʾʾly = Taurus

The term ʾʾly is clearly an elative formation, ʾafʿal,

derived from the root √ʾly; this corresponds to CAr �alan,

which is an adjective applied to ‘a ram with large but-

tocks’ (Lane 87c). This, in and of itself, is not a convinc-

ing rendition of the ‘bull’ or the ‘bull of heaven’, but

Safaitic ʾʾly is also cognate with the Akkadian name of this

constellation, alu, which refers to the bull as a mythologi-

cal being (CAD 377). The final y along with the two glot-

tal stops clearly indicate that the Safaitic term was not a

direct borrowing from Akkadian, but perhaps reflects a

calque or translation.

KRS 1706

l ys¹lm bn ʿwḏn bn mlk w ʿyd h- ḍʾn b- ḥrn b- rʾy ʾʾly f h

bʿls¹mn w h ḏs²r w lt w s²ʿhqm fs:

yt w ʿwr ḏ yʿwr

By Yslm son of ʿwḏn son of Mlk and he placed the

sheep in an enclosure in the Ḥawr�an during the rising of

Taurus so, O Bʿls¹mn and Ḏs²r and Lt and S²ʿhqm, let

there be deliverance and blind him who would efface

Commentary

ʿyd: I would take this as a D-stem of the root √ʿwd, the

basic sense of which is ‘to circle, enclose, encompass,

etc.’ (Leslau 1987: 77a). The Safaitic D-stem could refer to

the ‘encircling’ of the sheep, in other words, putting them

into an enclosure.

C 974

l ḥny bn qs¹m bn ḥn{y} bn lʿṯm bn zʾkt bn ʾs¹lm bn mbdy

bn kwnt w mtʿ l- ẖlqt b- h- mʿzyn b- r{ʾ}y ʾʾly f rʿy h- bq{r}

w ẖr{s:

} s²nʾ f h lt s¹{l}m

By Ḥny son of Qs¹m son of Ḥn{y} son of Lʿṯm son of

Zʾkt son of ʾs¹lm son of Mbdy son of Kwnt and it (a pla-

net?) shown brightly for a period in Orion during the ris-

ing of Taurus, then he pastured the cattle and watched out

for enemies so, O Lt, may he be secure

Commentary

This inscription, which is difficult to interpret, seems to

be describing the behaviour of a planet with respect to

Orion during the rising of Taurus. The translation of mtʿ l-

hlqt b- h- mʿzyn is very tentative.

mtʿ: if we consider the phrase mataʿa n-nah�aru ‘the day

became advanced, the sun being held high’ (Lane 3016c)

with the sense the verb takes when applied to wine,

mataʿa n-nab�ıḏu ‘the wine’s red colour was intense’, it

would seem that the verb’s basic sense has to do with

brightness or intensity of colour. Indeed, brightness was a

notable feature of the planets and stars for the Babylonians

(Reiner 1981: 18). The verb mataʿa must refer to an

understood body, perhaps one of the planets. Venus, for

example, will pass through Orion during the heliacal rising

of Taurus.

223

ANCIENTARABIAN ZODIAC

Page 11: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

ẖlqt: the term ẖlqt must be understood temporally here

rather than spatially, thus ‘a period’ rather than ‘a portion’.

A parallel is attested in C 2820: w brḥ ẖlqt s²ty h- dr ‘and

he departed this place for the period of winter’.

b- h- mʿzyn: most understand mʿzyn as a variant plural

form of mʿz ‘goat’, the normal plural of which is mʿzy. It is

better to understand the final -n as an adjectival suffix,

*�an, which would render the term analogous with CAr

maʿʿ�azun ‘a possessor, or master of, miʿzan [or goats]’

(Lane 2724c), and thus a suitable match for the Babylo-

nian Orion, ‘the loyal shepherd of Heaven’.

Gemini = gml

At first glance, this would seem to be the equivalent

of Babylonian gamlu, the crook (Auriga). Since gml also

follows the preposition b-, it is more likely that the term

refers to a constellation on the ecliptic. It is tempting to

view gml as a borrowing from Latin GEMINI. The confu-

sion of n and l is unexpected, but not without parallel,

especially in loanwords. But the penetration of a Latin

term into the Arabian zodiac is problematic. The other

zodiac names reflect etymological equivalents to the

Babylonian or Greek and it is therefore unclear why

speakers would bring over a Latin term for this constel-

lation alone. Moreover, the use of Latin was highly

restricted in the Near East. While one occasionally finds

a Safaitic-Greek bilingual inscription, no Latin texts

have been produced by the inhabitants of the desert.

The connection of root √gml with ‘twins’ can be made

on etymological grounds by appealing to CAr �gumlatun,

�gumul, which can refer to ‘any aggregate unseparated’

(Lane 460b–c). While this might not be a direct transla-

tion of the Babylonian, it does capture the basic sense of

two conjoined parts. On the other hand, the constellation

itself looks remarkably like a camel (Fig. 1). It has two

vertical lines which fork at the bottom, resembling legs,

and a crossbar three quarters of the way up with one end

extending further past the right parallel line than the end

crossing the left. This resembles a neck on the right side

and a tail on the left. Finally, the two lines draw closer

above the crossbar resembling a hump. Safaitic gml may

in fact refer to a camel and reflect a local innovation on

account of the constellation’s resemblance to a culturally

significant animal.16

SIJ 827

l ẓʿn bn mlk bn qms:

t bn ʾd bn s¹hwt w s:

yr m- mdbr

b-rʾy gml

By Ẓʿn son of Mlk son of Qmṣt son of ʾd son of S¹hwt

and he returned to permanent water from the desert during

the heliacal rising of Gemini [early July]

KRS 1560

l grf bn ʾws¹ w rʿy bql b- gml

By Grf son of ʾws and he pastured herbage when the

sun was in Gemini

LP 435

l grmʾl bn ḏʾb bn kn w ws¹q -h h- s¹l b- rʾy gml s¹nt

ws¹q ʾl qdm ʾl hrm f h lt s¹lm

By Grmʾl son of Ḏʾb son of Kn and the torrent drove

him away during the acronical rising of Gemini [mid-

December] the year the people of Qdm struggled with the

people of Hrm so, O Lt, may he be secure

Commentary

The phrase ws¹q h- s¹l was understood by Littmann

and others to mean ‘the torrent drove him away’. If this

is correct, then rʾy would signal the acronical rather than

heliacal rising of Gemini. Musil recorded an expression

among the Rwala ilya t:

alaʿat as-sheyl l�a t�amen as-seyl

‘when Canopus rises, trust not the Creek’.17 Since

Gemini also rises in mid-December, its appearance on

the evening horizon in the winter could have been

equivalent to the sighting of Canopus by the Rwala, and

Fig. 1.

Gemini.

16 This reimagining can be compared to the representation of

Sagittarius as a teapot in North America today. 17 See Macdonald (1992b: 2) for references and discussion.

224

A. AL-JALLAD

Page 12: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

signalled the beginning of the period in which torrents

become a threat.

LP 436

l ʿḏ bn s¹wr bn nqm w ws¹q -h h- s¹l b- rʾy gml f h lt

s¹lm l- ḏ s¹ʾr

By ʿḏ son of S¹wr son of Nqm and the torrent drove

him away during the acronical rising of Gemini so, O Lt,

let there be security to him who would leave (this inscrip-

tion) untouched

ʿqbt = Scorpio

The Safaitic ʿqbt is very close to the expected ʿqrb(t),

scorpion. The unexpected deletion of the r may point

towards taboo deformation, which is a common phenome-

non with dangerous animals. This hypothesis is impossible

to prove, however, since no inscriptions containing a

reference to the actual insect have been discovered. The

appearance of the name ʿqrb does not constitute counter-

evidence for this thesis since, as Macdonald has correctly

stated before, ‘the literal meaning [of a name] can seldom

have had much relevance to either the givers or receivers

of a name at the time it was bestowed, and once it was

attached to an individual it ceased to mean anything in

context except that particular person’ (1999: 254). Scorpio

originally encompassed the constellation of Libra, the stars

of which made up its claws; an awareness of this identifi-

cation persisted until Greek times (Rogers 1998: 25).

A single inscription seems to attest to an awareness of

a greater Libra-Scorpio among the inhabitants of the

Ḥarrah: MU 89 b- ks¹ʾ ʾmt ʿqbt ‘during the full moon of

Libra-Scorpio’.

ʿqbt is the most frequently attested constellation in

the Safaitic corpus, suggesting that its appearance and

disappearance in the sky had some cultural significance

for the authors of these inscriptions. If this is true, then

it may also have contributed to its phonological defor-

mation. Unlike the other constellations, ʿqbt occurs with

rʾy and ks¹ʾ in all but one case, which makes sense

given that the sun spends only seven days in Scorpio

proper.18

C 1895

l s²ʿ bn nẓmt b[n] ʿgz w wrd h- nmrt b- ks¹ʾ ʿqbt f hy

ʾ{y} [l]t19 rwḥ

By S²ʿ son of Nẓmt son of ʿgz and he came to water at

Nam�arah during the full moon of Scorpio [early May] so,

{O} [Lt], let there be relief

C 1927

l ʾʾs¹d bn mḥlm bn rbʾl bn ʾnʿm w ḥḍr h- dr w tẓr h-

s¹my b- r{ʾ}y ʿqbt f h bʿls¹mn rwḥ w ʿwr w wqd l- ḏ yʿwr

h- s¹fr

By ʾʾs¹d son of Mḥlm son of Rbʾl son of ʾnʿm and he

camped here at permanent water and awaited the rains dur-

ing the heliacal rising of Scorpio [mid-December], so O

Bʿls1mn let there be relief; and blindness and scorching

heat be upon him who would efface this writing

C 3818

l {ʾ}flt:

bn {g}rm{ʾ}l bn flt:

w ṯlg b- h- dr b- {r}{ʾ}y ʿqbt

By {Flt:

} son of {Grmʾl} son of Flt:

and it snowed here

during the heliacal rising of Scorpio [mid-December]

KRS 1719

l mrʿ bn ʿbd bn ʿḏ bn s²rb w kmʾ b- rʾy ʿqbt b

By Mrʿ son of ʿbd son of ʿḏ son of s²rb and he gathered

truffles during the acronical rising of Scorpio [mid-May]

b. . .

Commentary

Truffles abound in the season of the later rains, from

February to April, but they can continue to grow as late as

June. Perhaps it was this especially long season that com-

pelled the author to mention it in an inscription. The final

b could indicate that the inscription is incomplete.

KRS 2489

l m _gyr bn s:

rm bn nʿmy bn wqr w tẓr ʾ- s¹my b- ʿqbt

By M _gyr son of Ṣrm son of Nʿmy son of Wqr and he

awaited the rains when the sun was in Scorpio

KRS 1551

l ʾs¹ẖr bn ẖdmt bn trml bn s¹ry bn s¹lm w ʾlmn b- ʿqbt f

h lh rwḥ w mḥltn l- ḏ yʿwr h- s¹fr

By ʾs¹ẖr son of Hdmt son of Trml son of S¹ry son of

S¹lm and Mars is in Scorpio so, O Lh, let there be relief;

and a dearth of pasture to him who would efface this

writing

Commentary

ʾlmn: the editor left this word unexplained. Its syntax

would suggest that it is a verb, but the root √lmn does not

exist, and a derivation from the root √ʾlm would leave the

18 The term ʿqbt also occurs independently in curse formulae,

where it should be understood as a noun meaning ‘punishment’

(Lane 2099a). While the two terms are identical in consonantal

garb, they most likely had different vocalizations. Van den

Branden (1960) devised a rather different interpretation of ʿqbt,

but this was based on the misreading of rʾy ʿqbt as rhy ʿqbt,

and so there is no need for a lengthy discussion of his interpre-

tations.

19 The copy clearly shows ʾ{y}{b}t, but such a deity is unknown

from elsewhere. I have, therefore, amended the third letter to

[l], which gives us the well-known divine name Lt. The voca-

tive particle ʾy has been previously attested, e.g. KRS 1526.

225

ANCIENTARABIAN ZODIAC

Page 13: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

final n unexplained. Instead, I would suggest a connection

with Akk dLumnu, which refers to Mars (Reiner 1981:

13). The initial alif may reflect a compound ʾel-lumnu, a

phonological reproduction of the determinative, an elative

formation or even the definite article ʾ-. The term appears

once more where it can be equally interpreted as Mars:

KRS 1374 w tẓr h- s¹[m]y w ʾlmn ‘and he awaited the

{rains} and Mars’.

mḥltn: dearth of pasture is a common component of

curse formulae, but in most cases it appears simply as mḥlt,

without the final -n. The present mḥltn could reflect an idi-

omatic use of the dual or a rare attestation of nunation.

3.3. The remaining zodiacal constellations

The other constellations must be identified on an etymo-

logical basis as they do not occur following either rʾy or

ks¹ʾ and, as such, their identification remains tentative.

Cancer = s¹rṭn

Cancer is attested in three variant forms. C 3531 attests

a short form s¹rt:

, while LP 1190 terminates in a t, s¹rt:

t. C

1156 contains s¹{r}t:

{n}, but it is unclear if the restoration

of the second letter as an r is correct. Since all of the West

Semitic names of this constellation contain an n, it seems

best to restore the name of this constellation as s¹rt:

n, and

to explain the variant forms individually.

C 3531

l ʿmd bn mlk bn ʿmd bn ms¹k bn [ʿ]md bn mlk — h- bqr

b- s¹rt:

f h lt s¹lm

By ʿmd son of Mlk son of ʿmd son of Ms¹k son of

{ʿmd} son of Mlk — the cattle when the sun was in

Cancer so, O Lt, may he be secure

Commentary

Assuming that the absence of the final n on the copy

was not the result of a copy error, then it could have been

assimilated to the following letter. Word boundary assimi-

lation in a similar context is attested in KRS 2340: w ndm

ʿl ʾb -h w ʿl ʾdwd -h ʾrbʿt r _gm[n] mny ‘and he was devas-

tated on account of his father and his four paternal uncles

who were taken to the grave by fate’, where the passive

participle r _gm has a plural antecedent, and so should be

restored as r _gm[n], with the assimilation of the final n to

the following m.

LP 1190

l ʾnʿm bn ʾs¹{d} bn ʾnʿm s¹nt ḥdr ʾl s¹rt:

[n]

By ʾnʿm son of {ʾs¹d} son of ʾnʿm the year he camped

at permanent water until the sun was in Cancer

Commentary

s¹rt:

[n]: while the final t is clear on the copy, the copyist

may have mistaken a stray line through the n as part of the

letter. I would therefore suggest the following emendation:

s¹rt:

[n].

ʾl: the preposition ʾl can mean ‘until’ in Safaitic, similar

to CAr ḥatt�a (see Al-Jallad, forthcoming: §5).

There are two other possible attestations of s¹rt:

n, but

both their readings and context within the inscriptions

make their identification uncertain, and other explanations

exist. Both occur following the phrase dṯʾ hwrd; the first

term is clearly the verb ‘to spend the season of the later

rains’ while the second is open to interpretation.

C 1187

l ʿqrb bn m{d}{ʿ} bn s²m[t] [b][n] ẖ{l}ʾl w ḥll h- [d][r]

w d{ṯ}ʾ hw[r]d b- s¹lt:

By ʿqrb son of {Mdʿ} son of S²mt {son of} Hlʾl and he

camped here and spent the season of the later rains hw[r]d

b- s¹lt:

C 1156

l ẖlʾl [b][n] s²mt bn bn{t} bn ẖlʾl bn bnt w wgm ʿl- ʿb

{d} {w} {ʿ}l- ʾnʿm w ʿl- ʿbṯn w ʿl- ml{k} w ʿl- ʿd w ʿl- ḥb

{w} dṯʾ h{w} r( )d b- s¹{r}t:

{n}

By Hlʾl [son of] S²mt son of {Bnt} son of Hlʾl son of

Bnt and he mourned for ʿbd and for ʾnʿm and for ʿbṯn and

for Mlk and for ʿd and for Ḥb and he spent the season of

the later rains h{w} r( )d b- s¹{r}t:

{n}

If s¹{r}t:

{n} and s¹lt:

have the same referent, then it

seems likely that it was the constellation Cancer, which

would mean that s¹lt:

should be read as s¹{r}t:

[n]. It is pos-

sible to see in the l of s¹lt:

a malformed r, and one would

then only have to assume that the copyist overlooked the

final n, which is understandable considering that its shape

is a small line or dot. As is clear from other inscriptions,

the term h-wrd can refer to a toponym, for example, C

744 rʿy {h}wrd ‘he pastured Hwrd’. The ensuing chrono-

logical mismatch between Cancer—which the sun occu-

pies in late June and early July—and dṯʾ, however,

should motivate us to reinterpret hwrd in its present con-

text. I would suggest reparsing the clause as dṯʾ -h w rd

b- s¹rt:

n. In C 1187, the h would reflect a pronominal

clitic referring back to h- dr ‘this place’, thus ‘he camped

in this place, and spent the season of the latter rains at it’.

The absence of any antecedent for this pronoun in C

1156 could simply mean that its referent was understood,

that is, the location of the inscription. The next sentence,

w rd b- s¹rt:

n translates as ‘and he returned when the sun

was in Cancer’. The verb rd is simply the suffix conjuga-

tion of the root √rdd, ‘to return’. This reinterpretation fits

well our understanding of the year. The author seems to

have spent the season of the later rains in an area that,

while well watered in the spring, would have become dry

226

A. AL-JALLAD

Page 14: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

in the summer months, and so would explain his depar-

ture in late June.

If, however, one insists on maintaining a single interpre-

tation of the phrase dṯʾ h- wrd in all of the Safaitic inscrip-

tions, then perhaps the term following b- in both

inscriptions should be equated with the term s¹lt:

n, the

exact meaning of which is unclear (see C 66, 1952, 2802;

LP 424, 540, 1013). The basic sense has to do with ‘gov-

erning’ and ‘authority’, and so perhaps in this context it

refers to an area or district visibly under Roman rule.

Finally, since the r in these cases is not hooked, it is also

possible to read the word as s¹bt:

, that is, the month corre-

sponding to February–March. This, however, would not

explain the final {n} in C 1156.

Leo = h- ʾs¹d

Only a single inscription contains a possible reference

to Leo. The reasons for this are at present unclear.

C 28

l drʾl bn mty bn rgḍ bn hḏr w byt ẖyt:

f tẓr h- ʾs¹d f h ʾlt

s¹lm

By Drʾl son of Mty son of Rgḍ son of Ḥḍr and he spent

the night while travelling, then he awaited Leo, so O ʾlt,

may he be secure

Commentary

tẓr: this verb is the Gt-stem of the root √nẓr ‘to guard’

and is generally translated as ‘to await’ (cf. CAr intaẓara).

It is used primarily with meteorological phenomena

h- s¹my ‘the rains’ and mny ‘fate’. The literal meaning of

ʾs¹d ‘lion’ would make little sense in this context. In

Sabaic, ʾs¹d can mean ‘warriors’ or ‘soldiers’ (Beeston

et al. 1982: 7), but there is no evidence for such a meaning

in the Safaitic inscriptions. The attested term for a troop is

ms¹rt, and warriors seem to be called ṯbr or qtl.

h- ʾs¹d: the context in which Leo is mentioned seems to

suggest that the author was awaiting its rise on the hori-

zon, even though rʾy is not explicitly mentioned (cf. Mars

in KRS 1374). If this is correct, then the inscription was

probably composed in August.

Sagittarius = rmy or ṯbr, or both?

The identification of Sagittarius is complicated by the

various names associated with this constellation. The

MUL.APIN refers to it as Pabilsag, a relatively minor

Sumerian god. An alternative Babylonian tradition

identified the constellation as Nedu, the ‘soldier’, but it is

unclear if he possessed a bow and arrow. The image of this

constellation with the Greeks was that of a satyr archer,

and it was not until classical times that the familiar

centaur archer became associated with this sign (Rogers

1998: 26–27).

Two candidates for Sagittarius are found in the inscrip-

tions. The first is ṯbr, which I have suggested was the term

for soldier, perhaps a verbal adjective ṯabb�ar from the

sense ‘to drive away, expel’ (Lane 330c).20 This would be

a suitable match for Babylonian Nedu.21

There are a couple of inscriptions in which the term ṯbr

is better taken as a period of time rather than literally as

‘soldier’.

C 1758:

l ʿm bn (ḥ)r bn ʿm bn ḥ{n} ḏ- ʾl bʿr ( ) w wgm ʿl- ʾs²yʿ -h

ḏ ns¹l b- ṯbr

By ʿm son of Ḥr son of ʿm son of Ḥn of the lineage of

Bʿr and he grieved for his companions who went away

when the sun was in Sagittarius

Commentary

ns¹l: the n-stem of √s¹ll, cf. CAr ‘it (a thing) became

pulled forth, drawn out gently; he slipped away, or stole

away’ (Lane 1396a).

C 4443

w wgm ʿl- ʾm -h w ʿl- dd -h w ʿl- ẖl -h w ʿl- ʿm w ʿl-

ʾnʿm qtl -h (ʾ)l s:

bḥ f wlh ʿl- bn ẖl -h t( )rḥ w rʿy h- ḍʾn w

rḥḍ b- ṯbr w ẖl h- s²[n]ʾ f h lt s¹lm w wgd ʾṯr ʾẖ -h f ndm

And he grieved for his mother and for his paternal uncle

and for his maternal uncle and for ʾnʿm whom the people

of Ṣbḥ have killed, so he was distraught over his cousin,

who died; and he pastured the sheep, and washed during

Sagittarius, and then struck down the enemy, so O, Lt, let

there be security; and he found the inscription of his

brother, so he was devastated

Commentary

The phrase rḥḍ b- ṯbr parallels rḥḍ b- h- ngm found in

WH 3053 and may indicate a special type of bathing or

cleansing carried out at certain times of the year.

ẖl h- s²[n]ʾ: I connect hl with CAr halla ‘to perforate’

or ‘pierce through’; the phrase ẖallah�u bir-rumḥi means

20 For example: NST 3 l s²ʿ bn khl w wgm ʿl- s²gʾ ʾẖ -h rgm mny

w ʾty nẓr h- ṯbrn mn- ʾl s:

bḥ s¹nt ḥrb -hm ʾl ʿwḏ b- wʿl w gnmt l-

ḏ dʿy w ʿwr ds²r ḏ ẖbl. ‘By S2ʿ son of Khl and he grieved for

his brother, taken to the grave by fate; and the warriors from

the people of Ṣbḥ came to guard the year the people of ʿwḏ

waged war against them during/at wʿl so may he who would

read aloud have booty and may Ds²r blind him who would

scratch out’. For the interpretation of the verb dʿy, see Al-Jal-

lad, forthcoming: Glossary).21 A probably unrelated ṯ�abir is attested as the name of an agricul-

tural marker star which appeared between 10 and 22 May

1079, falling under the season of Ṣayf ‘spring’ (Piamenta

1990–1991, I: 55). I thank Professor J. Lentin for this

reference.

227

ANCIENTARABIAN ZODIAC

Page 15: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

‘he pierced him with the spear and transfixed his heart’

(Lane 778a). It is unclear if this sentence bears any rela-

tionship with the previous one, that is, whether washing

was connected with going into battle.

A single inscription containing a prayer to Bʿls¹mn, the

rain-god, attests a possible epithet of Sagittarius: h- rmy,

possibly */har-ramm�ay/ ‘archer’.

HN 62

l ḥmlg bn s¹hm bn qdm w ʾty h- rmy b- ḥqb -h f h

bʿls¹mn rwḥ

By Ḥmlg son of S¹hm son of Qdm and Sagittarius has

come with his quiver so, O Bʿls¹mn, let there be relief

Commentary

This inscription may preserve an expression among

the ancient inhabitants of the Ḥarrah concerning the

coming of the rains. If ʾty h- rmy signifies the heliacal

rising of Sagittarius, then this would take place at the

end of December, when rain or snow would have been

common.

ḥqb -h: Safaitic ḥqb should be connected with CAr

ḥaq�ıbatun ‘a bag or receptacle’ (Lane 610c), which in this

context signifies a quiver.

3.4. Other celestial bodies

The present study has focused on the constellations that lie

on the ecliptic, but other constellations and perhaps even

the names of planets and individual stars could await dis-

covery in the corpus. I have already suggested the identifi-

cation of mʿzyn as Orion and of ʾlmn as Mars. One

striking absence is the Pleiades. Known as zappu in the

MUL.APIN and an-na�gamu in CAr, the constellation does

not seem to be found in the patterns discussed above.

Since ngm has been identified as Virgo, the Pleiades must

have gone by another name. Aramaic knows this constel-

lation by the name k�ım�a. This term for the Pleiades is

attested as early as 2400 BCE in the form k�a-ma-t�u (Laf-

fitte 2003: 112). The reflex of this name in the Safaitic

inscriptions seems to be km. The absence of the final t sug-

gests that the name was drawn from Aramaic or some

other Northwest Semitic language in which the feminine

ending was realised as a vowel.

C 2572

l khl bn ḥml bn ns²bt bn ktm w ẖrs:

h- km f hy lt s¹lm

By Khl son of Ḥml son of Ns²bt son of Ktm and he

watched for the Pleiades, so, O Lt, may he be secure

Commentary

The heliacal rising of the Pleiades occurs in spring in

the northern hemisphere, and so the author may have been

awaiting the arrival of dṯʾ.

4. Conclusions

The Arabian zodiac as identified in this paper sits

somewhere between the Greek and West Semitic zodi-

acs and the Babylonian (Tables 6–8). Unlike the South

Arabian and Arabic constellations mentioned in the Isla-

mic sources, the names attested in the Safaitic inscrip-

tions are not direct ports from the Aramaic. Instead,

they seem to reflect a common heritage with the Baby-

lonian, which may be the result of the long historical

Table 6. The Arabian zodiac in context.

Latin English Greek

Aramaic

(Qumran)

South

Arabian Classical Arabic Sumero-Akkadian Arabian

Aries Ram Κqιός ארכד nyny al-ḥamal the Hireling lu.hun.ga

(hun, lu, lu) agru

ḏkr ‘ram’

Taurus Bull Τaῦqος ארות ṯwrn al-ṯawr Bull of An gu4.an.na

(mul.mul) alu/is le

ʾʾly ‘bull’

Gemini Twins Dίdυlοι אימואת al-gawz�aʾ/al-s:

�urah the Great Twins ma�s.tab.

ba.gal.gal m�a�su/t�uʾ�am�u rabutu

gml ‘twins?’/‘camel’

Cancer Crab Κaqj~ımος הנטרס s2rt:

n al-sarat:

�an the Crab al.lul (alla) alluttu s1rt:

n ‘crab’

Leo Lion Λέxm הירא al-ʾasad the Lion ur.gu.la (a, ur.a) urgulu/n�e�su h- ʾs1d ‘lion’

Virgo Maiden Πaqhέmος הלותב s1bltn al-sunbulah the Furrow ab.s�ın (absin) absinnu/�ser’u h- ngm ‘seed-produce’

Libra Scales Ζυcός אינזומ al-m�ız�an/al-zub�an�a the Scales zib�an�ıtu (r�ın) ʾmt ‘scale’

Scorpio Scorpion Σjοqpιός אברקע al-ʿaqrab the Scorpion g�ır.tab zuqaq�ıpu ʿqbt ‘scorpion’

Sagittarius Archer Τοξόsgς אתשק ḥẓyn al-qaws Pabilsag Pa.bil.sag (pa) divine name h- rmy/ṯbr ? ‘archer’/‘soldier’

Capricorn Goat-horned Αἰcόjeqxς אידג al-gad�ı the Goat-Fish suḫur.ma�s

(ma�s) suḫurm�a�su

yʾmr ‘sea-goat’

Aquarius Water-bearer Ὑdqοvόος אלוד al-dal�u the Great One gu.la (gu) s:

inundu/

kuurku/rammanu

mlḥ ‘salt vessel’,

‘salt worker’?

Pisces The fishes Ἰvhύeς אינונ ʾs3rn al-ḥ�ut/al-ri�s�aʾ the Tails (the field) kun.me�s

(zib.me�s, iku) zibb�atu/

zibb�at sin�un�utu

ḏl ‘tail(s)’

228

A. AL-JALLAD

Page 16: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

interaction between the peoples of the Ḥawr�an and the

deserts of North Arabia with Mesopotamian civilisation.

Part II of this paper will deal with the historical

implications of this finding.

AcknowledgementI would like to thank my friend M.C.A. Macdonald for his very

helpful comments and corrections on an earlier draft of this paper

and Dr. Markus Schmalzl (ALMA Regional Centre || Allegro) for his

kind help with astronomical matters. I also owe thanks to Prof.

J�erome Lentin, Prof. Andrzej Zaborski, Dr. Maarten Kossmann, and

Dr. Marijn van Putten for their comments on an earlier draft of this

paper. All errors remain my own.

Sigla

AbaNS Safaitic inscriptions in Ababneh 2005.

C Safaitic inscriptions in Corpus Inscriptionum Semiticarum.

Pars V. Paris, 1950–1951.

CSA Safaitic inscriptions in Clark 1984–85.

HALOT Koehler et al. 1994–1996; 1995–2000.

HaNSB Safaitic inscriptions in Ḥar�aḥi�sah 2010.

KRS Safaitic inscriptions recorded by G.M.H. King on the Basalt

Desert Rescue Survey and published on http://krcfm.orient.

ox.ac.uk/fmi/iwp/cgi?-db=AALC_BDRS&-loadframes

Lane Lane 1863–93.

Lisan Ibn Manẓ�ur [n.d.].

LP Safaitic inscriptions in Littmann 1943.

MUL.APIN Hunger & Pingree 1989.

NST Safaitic inscriptions in Harding 1951.

RWQ Safaitic inscriptions in al-R�us�an 2005.

SIJ Safaitic inscriptions in Winnett 1957.

WH Safaitic inscriptions in Winnett & Harding 1978.

ReferencesAbabneh, M. 2005. Neue safaitische

Inschriften und deren bildliche

Darstellungen. (Semitica et Semito-

hamitica Berolinensia, 6). Aachen: Shaker.

Beeston, A.F.L., al-Ghul, M., M€uller, W.W. &

Ryckmans, J. 1982. Sabaic Dictionary

(English–French–Arabic). Louvain-la-

Neuve: �Editions Peeters.

Clark, V.A. 1984–85. New Safaitic Inscriptions

from Sakaka and Azraq. Abr-Nahrain 23:

14–21.

del Olmo Lete, G. & Sanmart�ın, J. 2004. A

Dictionary of the Ugaritic Language in the

Alphabetic Tradition. Leiden: Brill.

Euting, J. 1896. Tagbuch einer Reise in Inner-

Arabien. Vol. 1. Leiden: Brill.

Greenfield, J.C. & Sokoloff, M. 1989.

Astrological and Related Omen Texts

in Jewish Palestinian Aramaic. Journal

of Near Eastern Studies 48/3:

201–214.

Ḥar�aḥi�sah, R. 2010. Nuq�u�s s:

af�aʾiyyah mina

ʾl-b�adiyah al-ʾurduniyyah. Amman: Ward

Books.

Harding, G.L. 1951. New Safaitic Texts.

Annual of the Department of Antiquities of

Jordan 1: 25–29.

Hayajneh, H. 2006. The Nabataean Camel

Burial Inscription from Wadi Ram, Jordan.

Based on a Drawing from the Archives of

Professor John Strugnell. Die Welt des

Orients 36: 104–115.

Huehnergard, J. Forthcoming. Arabic in its

Semitic Context. In Al-Jallad, A.M. (ed.),

Arabic in Context. Leiden: Brill.

Hunger, H. & Pingree, D. 1989. MUL.APIN:

An Astronomical Compendium in

Cuneiform. Horn, Austria: Verlag

Ferdinand Berger & S€ohne Gesellschaft

m.b.H.

Ibn Manẓ�ur [n.d.]. Lis�an al-ʿarab. Beirut: D�ar

Ṣ�adar.

Al-Jallad, A.M. 2015. Graeco-Arabica I: the

southern Levant. Pages 1–60 in Nehm�e, L.

(ed.), �Ecrit et �ecriture dans la formation

des identit�es en monde syriaque et arabe

IIIe–VIIe si�ecles. Leuven: Peeters.

Al-Jallad, A.M. Forthcoming. An Outline of a

Grammar of the Safaitic Inscriptions.

Leiden: Brill.

Jamme, A. 1971. The Grammatical Usage of

the Safaitic Verb rʿy, to pasture. Annali

dell’Istituto Orientale di Napoli 31:

21–40.

Kelley, D.H. & Milone, E.F. 2005.

Exploring Ancient Skies: A Survey of

Ancient and Cultural Astronomy. New

York: Springer.

Koch-Westenholz, U. 1995. Mesopotamian

Astrology: An Introduction to Babylonian

and Assyrian Celestial Divination.

Copenhagen: University of Copenhagen.

Koehler, L., Baumgartner, W., Richardson,

M.E.J. & Stamm, J.J. 1994–1996. Hebrew

and Aramaic Lexicon of the Old Testament

I–IV. Leiden: Brill.

Koehler, L., Baumgarner, W., Richardson,

M.E.J. & Stamm, J.J. 1995–2000.

Hebrew and Aramaic Lexicon of the Old

Testament I–V. (Electronic edition).

Leiden: Brill.

Lane, E.W. 1863–93. An Arabic-English

Lexicon. London: Williams and Norgate.

Laffitte, R. 2003. Les noms s�emitiques des

signes du zodiaque: de Babylone �a

Baghdad. Comptes rendus du GLECS 34:

97–118.

Leslau, W. 1987. Comparative Dictionary of

Geʾez (Classical Ethiopic): Geʾez–English,

English–Geʾez, with an Index of the

Semitic Roots. Wiesbaden: Harrassowitz.

Table 7. Parazodiacal constellations.

Constellation Safaitic Comparative evidence

Orion h- mʿzyn cf. Akkadian MULSIPA.ZI.AN.NA ‘loyal

shepherd of heaven’

The Pleiades h- km cf. Aramaic k�ım�a; PNWS *kimatu,

Akk. k�ımtu ‘family’

Table 8. Planets.

Planet Safaitic Comparative evidence

Mars ʾlmn cf. Akkadian dLumnu ‘evil one’

229

ANCIENTARABIAN ZODIAC

Page 17: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

Littmann, E.W. 1943. Safaitic Inscriptions.

Syria. (Publications of the Princeton

University Archaeological Expeditions to

Syria in 1904–1905 and 1909. Division

IV, Section C). Leiden: Brill.

Macdonald, M.C.A. 1992a. The Seasons and

Transhumance in the Safaitic Inscriptions.

Journal of the Royal Asiatic Society 2: 1–11.

Macdonald, M.C.A. 1992b. North Arabian

epigraphic notes I. AAE 3: 23–43.

Macdonald, M.C.A. 1999. Personal names in

the Nabataean realm: a review article.

Journal of Semitic Studies 44/2: 251–289.

Macdonald, M.C.A. 2004. Ancient North

Arabian. Pages 448–533 in Woodard, R.

D. (ed.), The Cambridge Encyclopaedia of

the World’s Ancient Languages.

Cambridge/New York: Cambridge

University Press.

Piamenta, M. 1990–1991. Dictionary of post-

classical Yemeni Arabic. (2 volumes).

Leiden: Brill.

Reiner, E. 1981. Babylonian Planetary

Omens: Part Two. In collaboration with

David Pingree. Malibu: Undena

Publications.

Rogers, J.H. 1998. Origins of the ancient

constellations: I. The Mesopotamian

traditions. Journal of the British

Astronomical Association 108: 9–28.

al-R�us�an, M. 2005. Nuq�u�s s:

afawiyyah min

w�ad�ı qas:

�ab bil-ʾurdunn. PhD thesis, Ar-

Riy�aḍ, J�amiʿat al-Malik Saʿ�ud, Kingdom

of Saudi Arabia.

Schramm, W. 2010. Akkadische Logogramme.

G€ottingen: Universit€atsverlag G€ottingen.

Stein, P. 2005. Once again, the division of the

month in Ancient South Arabia.

Proceedings of the Seminar for Arabian

Studies 35: 279–286.

Van den Branden, P.A. 1960. HSMYʾ-SMYʾ et

RHY ʿQBT ʿQBʾ dans les textes de Safa et

de Hatra. Al-Machriq: 217–230.

Winnett, F.V. 1957. Safaitic Inscriptions from

Jordan. Toronto: University of Toronto

Press.

Winnett, F.V. & Harding, G. 1978. Inscriptions

from Fifty Safaitic Cairns. Toronto:

University of Toronto Press.

230

A. AL-JALLAD

Page 18: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

Addendum to A. Al-Jallad, An Ancient Arabian Zodiac, Part I (AAE 25 214-

30).

(10/28/2014, published at https://leidenuniv.academia.edu/AhmadAlJallad)

a. More notes on Ks¹ʾ

b. More attestations of h- ʾs¹d

c. Another attestation of the Pleiades kmt

d. The Safaitic year

a. More notes on ks¹ʾ

In my article, An Ancient Arabian Zodiac: The Constellations in the Safaitic Inscriptions, part I (AAE 25: 214-230, (2014)), I considered the term ks¹ʾ to mean ‘full moon’. This meaning was supported by cognates in Hebrew and Aramaic (p. 216), and by the time periods to which

the inscriptions referred. For example, we seem to have two inscriptions which likely date

to the same remarkable event, but using different references. Thus:

HaNSB 218: w ngs² h- ʾbl m- ḥrn mn ṯlg b- rʾy ngm

‘and he drove the camels from the Ḥawrān on account of snowfall during the

heliacal rising of Virgo (~ mid-Oct)’

KRS 2851: w rʿy h- nẖl b- ṯlg b- {k}s¹ʾ ḏkr

‘and he pastured in the valley on snow during the full moon of Aries (~ mid-

late Oct)

It is also possible to interpret ks¹ʾ as the ‘latter part’ or ‘end’ of the constellation’s journey

across the celestial dome before it sets on the western horizon at sunrise, putting the

meaning within the semantic range of the Arabic root ksʾ (Lane 2608c). In other words, ks¹ʾcould refer to the cosmic setting of the asterism. This event would occur approximately

half a year after its rising, overlapping roughly with the period in which the full moon

would have occupied the same constellation. Our understanding of the chronology of these

inscriptions would, therefore, remain unchanged. The interpretation of the term as

Page 19: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

‘cosmic setting’ would form a nice pair, rʾy ‘rising’ and ks¹ʾ ‘setting’. However, in support of

the original meaning ‘full moon’, one can point towards a single attestation of what seems

to be a parallel zodiac tradition in AAUI 229, in which the author states w rʿy h- ʾnẖl b- qmr h-

ʾns¹y. Since the pasturing formula does not usually indicate associates or friends, it is

probably best to take the term following b- as a time period. In this case, b- qmr h- ʾns¹y ‘the

moon of ʾns¹y’ ‘Virgo’ would correspond to b- ks¹ʾ ngm. The term ʾns¹y could be connected

with CAr ʾanīsah, with perhaps a hypocoristic ending, meaning ‘young woman’, ‘maiden’. Thus, the translation would be: ‘and he pastured in the valleys during the (full) moon of

Virgo’. As it stands, it seems impossible to choose between the two translations with

absolute certainty.

All published texts containing ks¹ʾ

C 523

l ʾlh bn bʾẖh bn trb w wrd h- nmrt b- ks¹ʾ mlḥ

By ʾlh son of bʾẖh son of Trb and he went to water at Namārah during the full moon/cosmic

setting of Aquarius (end of July).

SESP.S 9

l mḥlm bn ʿbd bn mʿz bn ms²ʿr bn s¹ʿd bn wtr bn mlk w wrd b- ks¹ʾ ḏkr f h bʿls¹m{n} rwḥ w ẖrṣ h- rm f h

lt s¹lm w h- ẖṭṭ

By Mḥlm son of ʿbd son of Mʿz son of Ms²ʿr son of S¹ʿd son of Wtr son of Mlk and he went to water during the full moon/cosmic setting of Aries (mid-late October) so, O Bʿls¹mn, let there be ease and he kept watch for the Romans, so, O Lt, may he be secure, and this writing

too.

KRS 2851

l g{d}d bn —m—m bn ʿbd w rʿy h- nẖl b- ṯlg b- {k}s¹ʾ ḏkr

By {Gdd} son of {—m—m} son of ʿbd and he pastured in this valley on snow during the {full

moon/cosmic setting of} of Aries (mid-late October).

Page 20: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

C 1895

l s²ʿ bn nẓmt b[n] ʿgz w wrd h- nmrt b- ks¹ʾ ʿqbt f hy {ʾ}{l}t rwḥ

By S²ʿ son of Nẓmt son of ʿgz and he went to water at Namārah during the full moon/cosmic

setting of scorpio (early May) so, O ʾlt, let there be relief.

KRS 1881

l ʿqdt bn ʿbd w rʿy h- nẖl nwy w wrd b- ks¹ʾ {s¹}{n}n ʿ{q}bt h- ʾḍyt h- ʿqbt h- ẖṭṭ

By ʿqdt son of ʿbd and he pastured in the valley while migrating and went to the watering

hole during the full moon of s¹nn (thorns? claws?) of Scorpio (May)...

Mu 75

l gd bn wʿy bn bny bn ns²ʿʾl bn s¹ʿd bn s²nʾ bn ḥrb bn bʾs¹h bn bʿr w wgm ʿl- wrd w rʿy h- bql b- ks¹ʾ ʾmt ʿqbt f h lt w rḍy s¹lm m- [[]] + s² + nʾ w ʿwr l- ḏ yʿwr h- s¹fr

By Gd son of Wʿy son of Bny son of Ns²ʿʾl son of S¹ʿd son of S²nʾ son of Ḥrb son of Bʾs¹h son of Bʿr and he grieved for Wrd and pastured on fresh herbage during the full moon/cosmic rising of Libra-Scorpio (May?) so, O Lt and Rḍy, may he be secure against enemies but may

he who would efface this writing go blind.

C 4454

l ḥml bn ns²bt w ṣbb b- ks¹ʾ {g}ml

By Ḥml son of Ns²bt and he poured water during the full moon/cosmic setting of Gemini

(early December).

Page 21: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

RWQ 290

l gld bn ...hn bn s¹gʿ w wrd ʾdyt h bly b- ks¹ʾ yʾmr

By Gld son of …hn son of S¹gʿ and he went to water in the valleys of the Baliyyah during the

cosmic setting of Capricorn (early June).

b. More attestations of ʾs¹d = Leo

At the time of writing the article, I was only aware of one astronomical occurrence of ʾs¹d. A

few more have come to light; interestingly, both of these seem to anthropomorphize the

constellation, pointing perhaps towards a kind of mythology connected with the zodiac.

ASWS 124

l qṣy bn qʿṣn bn ʾḥlm bn rwḥ w ʾs¹lf h- rwy m h- ḥm f ẖz h- ʾs¹d f h s²ʿhqm fṣyt l -h

By Qṣy son of Qʿṣn son of ʾḥlm son of Rwḥ and the sweet water was brought to an end from

the heat, for Leo was hostile so, O S²ʿhqm, may he be delivered.

RSIS 80

l ʿqrb bn ʿbd bn nʿmn bn kn w mr b- ḍf f ʿdy h- ʾs¹d f ẖbl rbʿt f h lt brkt ʾẖk ḏ-

By ʿqrb son of ʿbd son of Nʿmn son of Kn and he passed by Ḍf; and Leo transgressed, and so

he ruined the spring, so, O Lt, bless ʾẖk ḏ …

c. Another attestation of the Pleiades, kmt

Another possible attestation of the Pleiades was found in an unpublished inscription from

Jebel Qurma, recording during the Leiden University survey of 2012. It reads:

l gbl bn ʾmyw rʿy h- kmt nwy

By Gbl son of ʾmy and he pastured at/during Pleiades while migrating

Page 22: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

The description seems abbreviated and must refer to either the rising or setting of the

asterism. Unlike the single previously known attestation (Al-Jallad 2014:228), the feminine t

is indicated in this form.

Page 23: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

d. The Safaitic Year

The relationship between the zodiac and the seasons is as follows (from Al-Jallad 2015, §24.8)

Season Latin Arabian Date of the sign Astronomical date (in

antiquity)

dṯʾ = Season of the

later rains

Pisces ḏl

‘tail(s)’ Mid-February – mid March

Mid February – early

April

38 days

Aries ḏkr

‘ram’ mid-March – mid- April

Early April – Late April

25 days

qyẓ =

Sum

mer

ṣyf = early

summer

Taurus ʾʾly

‘bull’ mid-April – mid- May

late April – late May

37 days

Gemini gml

‘twins?’ or ‘camel’ mid-May – mid- June

late May – late June

31 days

Cancer s¹rṭ[n] ?

‘crab’ mid-June – mid- July late June – early July

20 days

brḥt = late

summer

Leo h- ʾs¹d

‘lion’ mid-July – mid- August

early July – mid August

37 days

Virgo h- ngm

‘seed-produce’ mid-August – mid-

September

mid August – late

September

45 days

Libra ʾmt

‘scale’ mid-September – mid-

October

late September – late

October

23 days

s²ty = Winter Scorpio ʿqbt

‘scorpion’ mid-October – mid-

November

late october – late

october

7 days

Ophiucus Late October to mid-

November

18 days

Sagittarius h- rmy and/or ṯbr ?

‘archer’/’soldier’ mid-November to mid-

December

mid-november to mid-

december

32 days

Capricorn yʾmr ‘sea-goat’

mid- December – mid

January

mid-december – mid

January

28 days

Aquarius mlḥ

’salt vessel’ or ’salt worker’

mid- January – mid

February

Mid-January – mid

February

24 days

Page 24: Al-Jallad.2014.an Ancient Arabian Zodiac Part I Addendum-libre

Sigla

ASWS Awad, M. 1999. Dirāsat nuqūš ṣafawiyyah min ǧanūb wādī sārah al-bādiyah al-

ʾurdunniyyah aš-šimāliyyah. Masters Thesis. Irbid, Institute of Archaeology and

Anthropology, Yarmouk University.

C Ryckmans, J. (ed.) 1950-1. Corpus Inscriptionum Semiticarum Pars V. Paris:

Imprimerie Nationale.

KRS Safaitic inscriptions recorded by G.M.H. King on the Basalt Desert Rescue

Survey.

Mu Inscriptions recorded on the SESP surveys 1996–2003 (to appear on OCIANA).

RSIS Schirin, R. 2013. Nueu safaitische Inschriften aus Süd-Syrien (SSHB 16). Aachen:

Shaker Verlag.

RWQ Al-Rousan, M. 2005. Nuqūš ṣafawiyyah min wādī qaṣṣāb bi-l-ʾurdunn. PhD

dissertation. Ar-Riyāḍ, Ǧāmiʿat al-Malik Saʿūd, Kingdom of Saudi Arabia.

SESP Inscriptions recorded on the SESP surveys 1996–2003 (to appear on OCIANA).

Bibliography

Al-Jallad, A. 2014. An ancient Arabian zodiac. The constellations in the Safaitic inscriptions,

Part I. AAE 25: 214-230.

----. An outline of the grammar of the Safaitic inscriptions, (Studies in Semitic Languages and

Linguistics 80). Leiden: Brill.