restricted access Chapter 13: Semitic Sibilants

From: Black Athena

Rutgers University Press colophon
In lieu of an abstract, here is a brief excerpt of the content:

312 BLACK ATHENA SEMITIC SIBILANTS CHAPTER 13 INTRODUCTION I n Chapter 8 I looked at the progress of the Egyptian letter ß conventionally transcribed /s=/ from /h°/ to /s=/. I drew an analogy from transcriptions from the Hebrew /s=/ into Greek cq, sc, cs, and s.1 The situation of sibilants within Semitic is even more complicated than that. At this point, I shall not be treating the voiced or emphatic sibilants , which will be considered with individual loans. I shall restrict myself to those that are unvoiced and unemphatic. It is generally recognized that Proto-Semitic had three of this type, conventionally labeled /s1/, /s2/ and /s3/. It has also been generally considered that /s1/ corresponds to the Canaanite and Aramaic letter s=in, /s2/ to the Hebrew svin and /s3/ to samekh. In Phoenician, unlike the more conservative Hebrew, svin merged with s=in. In Hebrew svin remained independent until much later when it and the Aramaic svin merged with samekh. It is, therefore, maintained that up to then Hebrew retained the original values.2 In both Arabic and Ge’ez the modification was different. In both these, /s3/ merged with /s1/ and /s2/ remained independent but corresponded phonetically to the old /s3/. Correspondences with cognates in other Afroasiatic languages go against the conventional wisdom that in Proto-Semitic /s1/ was originally /s=/. These cognates suggest that the generally more conservative [CH. 13] SEMITIC SIBILANTS 313 Arabic and Ge'ez have preserved the original correspondence /s1/= /s/. In the First Millennium BCE loans and transcriptions from Akkadian /s1/ tended to be rendered as /s1/(s3) in Canaanite. This is generally seen as the result of an Akkadian shift /s±/ to /s/. This correspondence could be explained equally well or better by Akkadian having kept the original /s1/=/s/, while the Canaanite /s1/shifted to /s=/.3 It is difficult to say when exactly the shift took place but it would seem to have been during the last half of the Second Millennium BCE. Canaanite Phoenician Hebrew Arabic Aramaic Ge'ez S S1 S1 S3 S1 S1 S2/3 S1 S1/3 S: S2 S2 S2 S2 S2 S+ S3 S1/2 S3 S3 S1 S3 S2 Alphabetic transcription of these sibilants further increases the complexity . The earliest letter forms appear to have been S, “checkerboard ” and s. The first, a later horizontal form of which ç, became the Canaanite s=in. The Greek letter, however, took its name sigma with metathesis from the Phoenician letter name samekh. The Semitic letter samekh itself, the “checkerboard,” was altered by the shaft slipping down below the three horizontals to form samekh: i–––. In the more conservative Greek and Italic alphabets, the modifications were the less drastic modifications of the “checkerboard,” X, X or xi.4 I have questioned above whether this letter always had the value /ks/. Rather, the /k/ was probably a soft fricative /kh/, not a plosive. Thus, it may simply have stood for a fricative plus sibilant /khs/.5 Furthermore, it is striking that in all the Mycenaean forms identified with a later xi the vowels are repeated: kese, kisi, kusu etc. Syllabaries are by their nature unable to represent double consonants. Thus, such repetitions should not be segmented into k-s but, like xi itself, seem to represent fricatives rather than stops united with sibilants.6 LOANS OF SIBILANTS FROM CANAANITE INTO GREEK Greek lacked this multiplicity of unvoiced sibilants. Therefore, in loans from West Semitic or Akkadian before 1200 BCE, /s1/-/s/, later /s3 /, 314 BLACK ATHENA was rendered as /s/. For instance, s=s=mn, the conventional Ugaritic and Phoenician transcription, was written sasama in Linear B and shvsamon in alphabetic Greek. The Canaanite and Hebrew s=ôs=an or s=us=an “lily” became sou'son soûson (4).7 After 1200 BCE, borrowings from Canaanite and Aramaic /s1/-/s=/ still sometimes remained /s/ but equally often became /skh/, /khs/, /ks/ and, very likely, /sk/.8 This temporal pattern is the opposite of Greek borrowings from the Egyptian sign transcribed as /s=/. For examples of Greek renditions of the Egyptian /s=/ as first /k/ and /kh/, see Chapter 8 above.9 For renditions from the later development /skh/, there are s=nw “rope, net” in Late Egyptian “circuit, enclosure, cartouche,” Mycenaean kono, koino and the Greek scoi'no" (H) “reed, grass, rope, net, bind.” Then there is scediva (H) “raft” from the Egyptian s=dw “raft.” Chantraine provides no etymology for either. For a Canaanite example...