No CrossRef data available.
Article contents
The historical background of the vowel ṣere in some Hebrew verbal and nominal forms
Published online by Cambridge University Press: 05 February 2009
Extract
One of the most important sources for our knowledge of the length of vowels in the Tiberian tradition of Biblical Hebrew is a corpus of manuscripts containing transcriptions of the Hebrew Bible into Arabic letters. In most of the manuscripts the Arabic transcription employs the orthography of Classical Arabic to represent the sounds of Hebrew. Since Classical Arabic orthography used matres lectionis systematically to mark long vowels we are able to reconstruct the distribution of long and short vowels in Tiberian Hebrew. The transcriptions show us that the main factors determining vowel length were stress and syllable structure.
- Type
- Articles
- Information
- Bulletin of the School of Oriental and African Studies , Volume 57 , Issue 1 , February 1994 , pp. 133 - 144
- Copyright
- Copyright © School of Oriental and African Studies, University of London 1994
References
1 See Khan, G., Karaite Bible manuscripts from the Cairo Genizah (Cambridge, 1990)Google Scholar, ‘Vowel length and syllable structure in the Tiberian tradition of Biblical Hebrew’, Journal of Semitic Studies, XXXII, 1987, 23–82CrossRefGoogle Scholar, ‘The orthography of Karaite Hebrew Bible manuscripts in Arabic transcription’, Journal of Semitic Studies, XXXVIII, 1993, 49–70.CrossRefGoogle Scholar
2 Khan, G., ‘The pronunciation of the minor ga'ya as reflected in Karaite Bible manuscripts in Arabic transcription’ [in Hebrew], Language Studies, V–VI, Bar-Asher, M. (ed.), (Jerusalem, 1992), 465–79Google Scholar, ‘The pronunciation of the verbs 
and in the Tiberian tradition of Biblical Hebrew’, to appear in Goldenberg, G. and Raz, S. (ed.), Studies in Semitic and Cushitic linguisticsGoogle Scholar.
3 Differences in duration conditioned by vowel quality are thought to be above the threshold for auditory discrimination and thus to be audible. They are also thought to be physiologically conditioned and so constitute a phonetic universal. See Lehiste, I., Suprasegmentals (Cambridge, Mass., 1970), 18–19Google Scholar.
4 For other aspects of the historical phonology of Hebrew that are affected by this principle see my article ‘Remarks on the historical phonology of Hebrew’ (forthcoming).
5 cf. Khan, G., Journal of Semitic Studies, XXXII, 1987, 23–82CrossRefGoogle Scholar.
6 The Babylonian pataḥ corresponds to both pataḥ and segol in the Tiberian vocalization.
7 Yeivin, I., The Hebrew language tradition as reflected in the Babylonian vocalization [in Hebrew], (Jerusalem, 1985), 434–5Google Scholar.
8 Yeivin, , 602–3Google Scholar.
9 ibid., 603.
10 ibid., 505–7.
11 ibid., 514–15, 526–7.
12 ibid., 817–50.
13 ibid., 778–5.
14 By the time of Origen (A.D. 185–254) it is likely that Greek η was pronounced with the quality of the close vowel ī and Greek aι was pronounced ē. Since in the transcription of the Hexapla η corresponds not to Hebrew ī but to ē and aι corresponds to the Hebrew diphthong ay, it would appear that the transcription was originally written at an earlier period. The aforementioned shifts in the pronunciation of Greek are datable to the second century A.D. (cf. Allen, W. S., Vox Graeca: a guide to the pronunciation of Classical Greek, Cambridge, 3rd ed. 1987, 74, 79Google Scholar). This implies that the transcription that was incorporated into the Hexapla was written no later than the first century A.D. (see Janssens, G., Studies in Hebrew historical linguistics based on Origen's Secunda, Leuven, 1982, 20–1Google Scholar).
15 In classical Atticel ει was pronounced as the close mid 
but by the end of the fourth century B.C. it began to close to [ī]. It is possible that this shift facilitated the closing of ; cf. Allen, , Vox Graeca, 74Google Scholar.
16 In the Babylonian tradition final open syllables vocalized with segol in Tiberian Hebrew are sometimes vocalized with qameṣ: 
(
, Ps. 74: 23), 
(
, Ezek. 2: 9). Conversely, where Tiberian Hebrew has a final open syllable vocalized with qameṣ (
, –) the Babylonian tradition sometimes has pataḥ: 
(
, Ps. 148: 8) (Yeivin, , 685–97Google Scholar). This interchange between qameṣ and pataḥ could not have arisen by analogy with 
roots. The masc. sing, active participle Qal of verbs has ṣere in the final syllable not qameṣ. It would appear that in the Babylonian tradition the final vowel in these forms was sometimes opened even further, with the result that it was treated as [a] by the 
quality shift and shifted to 
.
17 The reading of the transcription of the vowel in the final syllable of 
(Ps. 46. 10) is not certain. Mercati, G. (Psalterii Hexapli Reliquiae, The Vatican, 1958)Google Scholar reads оνк.σσες with epsilon.
18 As we have seen, in at least one context (viz. forms such as μασε = 
) epsilon represented a long vowel.
19 Janssens, , Studies in Hebrew historical linguistics, 70Google Scholar.
20 Ch. Sarauw, , Über Akzenl und Silbenbildung in den älteren semitischen Sprachen (Copenhagen, 1939), 56 ffGoogle Scholar. According to Sarauw, Philippi's law affected short i in every stressed closed syllable. This was an extension of earlier formulations of the law, in which it was said to operate in more restricted contexts. For a summary of these earlier formulations see Blau, J., ‘On the chronology of Lex Philippi [in Hebrew], Ninth World Congress of Jewish Studies (Jerusalem, 1986), Division D, vol. i, 1–2Google Scholar.
21 Yeivin, , The Hebrew language tradition, 383Google Scholar.
22 Sarauw, , Über Akzent und Silbenbildung, 78Google Scholar, Brønno, E., Sludien über hebräische Morphologie und Vokalismus, auf Grundlage der Mercatischen Fragmente der zweiten Kolumne der Hexapla des Origenes (Leipzig, 1943), 448Google Scholar.
23 cf. Sarauw, , Über Akzent und Silbenbildung, 84–85Google Scholar, Blau, , Pröceedings of the ninth World Congress of Jewish Studies, vol. I (Jerusalem, 1986), 1–4Google Scholar; Qimron, E., Lěšonénu, L, 1986, 248-9Google Scholar.
24 Other forms which have pataḥ in Tiberian Hebrew resulting from Philippi's law such as 
(<*bint), and 
(construct <*zaqin) do not occur in the extant portions of the Hexapla transcription. It is significant that the epenthetic vowels in segolate forms are not represented in the Hexapla.
25 The form 
written with scriptio plena in the Isaiah scroll from Qumran (IQIsa, Is. 42: 16, MT: 
) also reflects a pronunciation that has not been affected by Philippi's law. See Qimron, , Lěšonénu, L, 248Google Scholar.
26 Sarauw, , Über Akzent und Silbenbildung, 56 ff.Google Scholar
27 ibid., 90.
28 ibid., 80.
29 Blau, J., ‘On pausal lengthening, pausal stress shift, Philippi's law and rule ordering in Biblical Hebrew’, Hebrew Annual Review, v, 1981, 1–11Google Scholar. He argues, however, that the pataḥ in the final closed syllable of some, imperfect consecutive forms in, pause arose originally in pause, e.g. pausal 
vs. context 
; pausal 
vs context 
. Here the pataḥ arose by the operation of Philippi's law after pausal lengthening had taken place and after the pausal stress shift to the closed ultima.
30 Qimron, , ‘Interchanges of ṣere/pataḥ in Biblical Hebrew’ [in Hebrew] Lésonénu, L, 1986, 77–101Google Scholar.
31 In addition to the levelling of pausal and contextual forms of original qiṭl the segolate nouns underwent further analogical processes, e.g. contextual qēṣel vs. pausal qāṣel; see Sarauw, , Über Akzent und Silbenbildung, 85–6Google Scholar; Blau, , Hebrew Annual Review, V, 1981, p. 3 n. 8Google Scholar.