etc.), Rahman {'merciful,' perhaps of monotheistic times, CIS 6), Hainan (8) and others whose nam* is thought to signify water-nymphs (153, etc.).
This pantheon appears to resemble that of the Italians before Greek influence : the gods were to some extent hypostases of operations or objects, and there was supposed to be some special merit in enumerating them. Of this last process the terminations of many inscriptions offer illustra- tions. The more important of their temples had names, after which the god was often called. The oti'erings to them consisted, as we have seen, of lavish gifts to the temples; but sacrifices of tin ordinary sort (CIS 290) and offerings of incense (194) also form the subject of allusions. Sometimes it took the form of self-presentation on the part of the worshipper, whatever may have been the import of that act. The earliest instance is said to be in a bustrophedon inscription (ZD3IG xxii. 425), and the most elaborate, that contained in the inscription of Hadakan (CIS 37), in which the author declares that he puts the god in possession of himself, his famUy, his and their property, and all the property belonging to his clan. If the inscription HI 2678 (p. 26) be rightly interpreted by Mordtmann, this act could be performed re- peatedly ; and the inscr. CIS 126 would probably explain it more clearly, if we knew the meaning of the words. The plan of erecting stones ia honour of the gods also finds illustration (CIS 100); and most of the texts we have are musnads, or tablets dedicated to the gods, sometimes with other oti'erings. The office of priest (iiym) seems some- times to have been united with that of tribal head (CIS 41. 1), but at other times was probably dele- gated to humbler individuals. That pUgilmages were made in honour of the gods appears from the month Dhu Hijjat or Mahajjat ; the former of which is the only month-name which the Saba;ans share with the Moslems (the Sakean twelve are enumerated by Midler in MM 51). Prayers are ordinarily designated by the common Semitic word for petition, but the other word (amid), which occurs often, perhaps implies stereotyped formulae. From the inscr. CIS 126 it would appear that the gods were also appeased |by certain forms of per- sonal abstinence, and from one of those edited by Winckler (I.e.) it might appear that they had some share in the administration of justice. The Sah:eans also had certain ideas of ceremonial purity, violation of which had to be atoned for by public acknowledgment on tablets placed in temples : some curious specimens of these are given in SA pp. 20-25.
iv. Language. — Of the S. Arabian inscriptions, a few are couched in a dialect scarcely distinguish- able from classical Arabic. This is the case with the texts dealing with ceremonial purity, to which reference has been made. The Sabaean texts seem to resemble most closely the dialect known as Ethiopic ; and indeed Ethiopic may be regarded as the form of Saba^an first given literary shape by Christian missionaries, altuough, unless the dates on the Marib inscriptions (Glaser, MVAS 6) are absolutely misleading, Sabajan must have con- tinued in use for a centmy or two after the com- mencement of Ethiopic literature. Owing to the absence of vowels, we know little of the pronun- ciation or the grammatical finesse of Saba^an ; but it clearly differed from the classical Arabic idiom in many particulars ; in some of which it pre- served what classical Arabic lost, while more often it seems to represent a later stage of development tlian the latter. Its alphabet retains a sioilant lost to Arabic ; and in certain cases the weak letters have still consonantal value in Sabsean (as in Ethiopic) where they have lost it in Arabic Instead of the prefixed article which govenu
SHEBA
SHEBNA
483
Arabic syntax, Sabsean has an affix, similar to that in ase in Aramaic ; both of which bear a curious likeness to the Armenian system. For the nunation which in Arabic supplies, to some ex- tent, the place of an indefinite article, Sabacan has miinnticm. Probably in this matter Arabic retains the older termination, whereas the two languai:es may have developed or borrowed their definite articles independently. The emploj-ment of the dual would appear to have been as regular in Sabxan as in ^Vrabic, though the mode of express- ing it differed somewh.at. The Salxpan syntax has also some remarkable peculiarities, to which nothing in Arabic corresponds, thouyh they might be illustrated from Helirew. We have already seen (in art. Language of the OT) that, like Ethiopic, Sabsan occasionally agrees in its vocabu- lary with Canaanitish against Arabic ; and there are also cases in which it agrees remarkably with the Aramaic vocabulary, although in the most striking of these (see CIS 79) the common words are perhaps borrowed from Aramaic, since the in- scription shows signs of having been written by a foreigner. Though there is still much about both grammar and vocabulary that is obscure, the progress made in the study since Osiander's time compares favourably with that achieved in other regions of epigraphy.
D. S. Maegoliouth.
SHEBA (MP ; B ^i/ica, A Zd;3ef ; Sal)ee).—A town, according to AV, wliich was allotted to Simeon (Jos 19-), and is mentioned between Beersheba and ^Ioladah. This was apparently the view of Euse- bius and Jerome (Onom. s. Za^t). RV, however, and the edition of 1611, read ' Beersheba or Sheba' ; and this is in agreement with the number of towns (13) said to have been allotted to Simeon (Jos 19^"*), and with the omission of Sheba from the list in 1 Ch 4**. It is not unlikely that yjy] is due to dittograpby from vzv ixn, or it may be a corruption of PDPi (cf. LXX B) of Jos W\ So Dillm. ad loc.