Gesenius' Hebrew Grammar/6. Pronunciation and Division of Consonants

by Wilhelm Gesenius

Layout 2

§6. Pronunciation and Division of Consonants.

P. Haupt, ‘Die Semit. Sprachlaute u. ihre Umschrift,’ in Beiträge zur Assyriologie u. vergleich. semit. Sprachwissenschaft, by Delitzsch and Haupt, i, Lpz. 1889, 249 ff.; E. Sievers, Metrische Studien, i, Lpz. 1901, p. 14 ff.

a 1. An accurate knowledge of the original phonetic value of each consonant is of the greatest importance, since very many grammatical peculiarities and changes (§ 18 ff.) only become intelligible from the nature and pronunciation of the sounds. This knowledge is obtained partly from the pronunciation of the kindred dialects, especially the still living Arabic, partly by observing the affinity and interchange of sounds on Hebrew itself (§ 19), and partly from the tradition of the Jews.^[1]

The pronunciation of Hebrew by the modern German Jews, which partly resembles the Syriac and is generally called ‘Polish’, differs considerably from that of the Spanish and Portuguese Jews, which approaches nearer to the Arabic. The pronunciation of Hebrew by Christians follows the latter (after the example of Reuchlin), in almost all cases.

b The oldest tradition is presented in the transcription of Hebrew names in Assyrian cuneiform; a later, but yet in its way very important system is seen in the manner in which the LXX transcribe Hebrew names with Greek letters.^[2] As, however, corresponding signs for several sounds (ט, ע, צ‍, ק, שׁ) are wanting in the Greek alphabet, only an approximate representation was possible in these cases. The same applies to the Latin transcription of Hebrew words by Jerome, according to the Jewish pronunciation of his time.^[3]

On the pronunciation of the modern Jews in North Africa, see Bargès in the Journ. Asiat., Nov. 1848; on that of the South Arabian Jews, J. Dérenbourg, Manuel du lecteur, &c. (from a Yemen MS. of the year 1390), Paris, 1871 (extrait 6 du Journ. Asiat. 1870).

c 2. With regard to the pronunciation of the several gutturals and sibilants, and of ט and ק, it may be remarked:—

1. Among the gutturals, the glottal stop א is the lightest, corresponding to the spiritus lenis of the Greeks. It may stand either at the beginning or end of a syllable, e.g. אָמַר ʾāmár, יֶאְשַׁם yäʾšám. Even before a vowel א is almost lost to our ear, like the h in hour and in the French habit, homme. After a vowel א generally (and at the end of a word, always) coalesces with it, e.g. קָרָא qārā for an original qārăʾ, Arab. qărăʾä; see further, § 23 a, 27 g.

d ה before a vowel corresponds exactly to our h (spiritus asper); after a vowel it is either a guttural (so always at the end of a syllable which is not final, e.g. נֶהְפַּךְ nähpakh; at the end of a word the consonantal ה has a point—Mappîq—in it, see § 14), or it stands inaudible at the end of a word, generally as a mere orthographic indication of a preceding vowel, e.g. גָּלָה gālā; cf. §§ 7 b and 75 a.

e ע is related to א, but is a much stronger guttural. Its strongest sound is a rattled, guttural g, cf. e.g. עַזָּה, LXX Γάζα, עֲמֹרָה Γόμοῤῥα; elsewhere, a weaker sound of the same kind, which the LXX reproduce by a spiritus (lenis or asper), e.g. עֵלִי Ἡλί, עֲמָלֵק Ἀμαλέκ.^[4] In the mouth of the Arabs one hears in the former case a sort of guttural r, in the latter a sound peculiar to themselves formed in the back of the throat.—It is as incorrect to omit the ע entirely, in reading and transcribing words (עֵלִי Eli, עֲמָלֵק Amalek), as to pronounce it exactly like g or like a nasal ng. The stronger sound might be approximately transcribed by gh or ^rg; but since in Hebrew the softer sound was the more common, it is sufficient to represent it by the sign ʿ, as אַרְבַּע ʾarbaʿ, עַד ʿad.

f ח is the strongest guttural sound, a deep guttural ch, as heard generally in Swiss German, somewhat as in the German Achat, Macht, Sache, Docht, Zucht (not as in Licht, Knecht), and similar to the Spanish j. Like ע it was, however, pronounced in many words feebly, in others strongly.

g As regards ר, its pronunciation as a palatal (with a vibrating uvula) seems to have been the prevailing one. Hence in some respects it is also classed with the gutturals (§ 22 q, r). On the lingual ר, cf. o.

h 2. The Hebrew language is unusually rich in sibilants. These have, at any rate in some cases, arisen from dentals which are retained as such in Aramaic and Arabic (see in the Lexicon the letters ז, צ and שׁ).

i שׁ and שׂ were originally represented (as is still the case in the unpointed texts) by only one form ש; but that the use of this one form to express two different sounds (at least in Hebrew) was due only to the poverty of the alphabet, is clear from the fact that they are differentiated in Arabic and Ethiopic (cf. Nöldeke in Ztschr. f. wissensch. Theol., 1873, p. 121; Brockelmann, Grundriss, i. 133). In the Masoretic punctuation they were distinguished by means of the diacritical point as שׁ (sh) and שׂ (ś).^[5]

k The original difference between the sounds שׂ and ס^[6] sometimes marks a distinction in meaning, e.g. סָכַר to close, שָׂכַר to hire, סָכַל to be foolish, שָׂכַל to be prudent, to be wise. Syriac always represents both sounds by ס, and in Hebrew also they are sometimes interchanged; as סָכַר for שָׂכַר to hire, Ezr 4; שִׂכְלוּת for סִכְלוּת folly, Ec 1.

l ז (transcribed ζ by the LXX) is a soft whizzing s, the French and English z, altogether different from the German z (ts).

m 3. ט, ק‍, and probably צ are pronounced with a strong articulation and with a compression of the larynx. The first two are thus essentially different from ת and ך, which correspond to our t and k and also are often aspirated (see below, n). צ is distinguished from every other s by its peculiar articulation, and in no way corresponds to the German z or ts; we transcribe it by ṣ; cf. G. Hüsing, ‘Zum Lautwerte des צ,’ in OLZ. x. 467 ff.

n 3. Six consonants, the weak and middle hard Palatals, Dentals, and Labials

ב ג ד כ‍ פ ת (בְּגַדְכְּפַת)

have a twofold pronunciation, (1) a harder sound, as mutes, like k, p, t, or initial b, g (hard), d; and (2) a softer sound as spirantes.^[7] The harder sound is the original. It is retained at the beginning of syllables, when there is no vowel immediately preceding to influence the pronunciation, and is denoted by a point, Dageš lene (§ 13), placed in the consonants, viz. בּ b, גּ g, דּ d, כּ‍ k, פּ p, תּ t. The weaker pronunciation appears as soon as a vowel sound immediately precedes. It is occasionally denoted, esp. in mss., by Rāphè (§ 14 e), but in printed texts usually by the mere absence of the Dageš. In the case of ב, כ‍, פ, ת, the two sounds are clearly distinguishable even to our ear as b and v, k and German (weak) ch, p and ph, t and th (in thin). The Greeks too express this twofold pronunciation by special characters: ךּ κ, כ‍ χ; פּ π, פ φ; תּ τ, ת θ. In the same way ג should be pronounced like the North German g in Tage, Wagen, and ד like th in the, as distinguished from גּ and דּ.

For more precise information on the cases in which the one or the other pronunciation takes place, see § 21. The modern Jews pronounce the aspirated ב as v, the aspirated ת as s, e.g. רַב rav (or even raf), בַּיִת bais. The customary transcription (used also in this Grammar) of the spirants ב, כ‍, ת by bh, kh, th is only an unsatisfactory makeshift, since it may lead (esp. in the case of bh and kh) to an erroneous conception of the sounds as real aspirates, b–h, k–h.

o 4. According to their special character the consonants are divided into—

	(a) Gutturals	א ה ע ח;
	(b) Palatals	ג כ‍ ק;
	(c) Dentals	ד ט ת;
	(d) Labials	ב פ;
	(e) Sibilants	ז שׁ שׂ ס צ‍;
	(f) Sonants	ו י, ר ל, נ‍ מ‍.

In the case of ר its hardest pronunciation as a palatal (see above, g, end) is to be distinguished from its more unusual sound as a lingual, pronounced in the front of the mouth.

On the twofold pronunciation of r in Tiberias, cf. Delitzsch, Physiol. und Musik, Lpz. 1868, p. 10 ff.; Baer and Struck, Dikduke ha-t^eamim, Lpz. 1879, p. 5, note a, and § 7 of the Hebrew text, as well as p. 82.

p In accordance with E. Sievers, Metrische Studien, i. 14, the following scheme of the Hebrew phonetic system is substituted for the table formerly given in this grammar:—

i. Throat sounds (Gutturals): ח ע ה א.

ii. Mouth-sounds:		w.	m.	e.	w.	m.
1. Mutes and Spirants:	Palatal	גּ	כּ‍	ק	ג	כ‍
	Dental	דּ	תּ	ט	ד	ת
	Labial	בּ	פּ	—	ב	פ
2. Sibilants:	...	ז	שׁ שׂ ס	צ‍
3. Sonants:	...	ו י	ר ל	נ‍ מ‍

q Rem. 1. The meaning of the letters at the top is, w. = weak, m. = middle hard, e. = emphatic. Consonants which are produced by the same organ of speech are called homorganic (e.g. ג and כ‍ as palatals), consonants whose sound is of the same nature homogeneous (e.g. ו and י as semi-vowels). On their homorganic character and homogeneity depends the possibility of interchange, whether within Hebrew itself or with the kindred dialects. In such cases the soft sound generally interchanges with the soft, the hard with the hard, &c. (e.g. ד = ז, ת = שׁ, ט = צ‍). Further transitions are not, however, excluded, as e.g. the interchange of ת and ק (ת = כ = ק). Here it is of importance to observe whether the change takes place in an initial, medial, or final letter; since e.g. the change in a letter when medial does not always prove the possibility of the change when initial. That in certain cases the character of the consonantal sound also influences the preceding or following vowel will be noticed in the accidence as the instances occur.

r Rem. 2. Very probably in course of time certain nicer distinctions of pronunciation became more and more neglected and finally were lost. Thus e.g. the stronger ע ^rg, which was known to the LXX (see above, e), became in many cases altogether lost to the later Jews; by the Samaritans and Galileans ע and ח were pronounced merely as א, and so in Ethiopic, ע like א, ח like h, ש like s.

s Rem. 3. The consonants which it is usual to describe especially as weak, are those which readily coalesce with a preceding vowel to form a long vowel, viz. א, ו, י (as to ה, cf. § 23 k), or those which are most frequently affected by the changes described in § 19 b–l, as again א, ו, י, and ן, and in certain cases ה and ל; finally the gutturals and ר for the reason given in § 22 b and q.

1. Cf. C. Meinhof, ‘Die Aussprache des Hebr.,’ in Neue Jahrb. f. Philol. u. Pädag., 1885, Bd. 132, p. 146 ff.; M. Schreiner, ‘Zur Gesch. der Ausspr. des Hebr.,’ in ZAW. 1886, p. 213 ff.
2. Cf. Frankel, Vorstudien zu der Septuag., Lpz. 1841, p. 90 ff.; C. Könneke, ‘Gymn.-Progr.,’ Stargard, 1885. On the transcription of eleven Psalms in a palimpsest fragment of the Hexapla at Milan, see Mercati, Atti della R. Accad., xxxi, Turin, 1896. [Cf. Burkitt, Fragments of... Aquila, Cambr. 1897, p. 13.]
3. Numerous examples occur in Hieronymi quaestiones hebraicae in libro geneseos, edited by P. de Lagarde, Lpz. 1868; cf. the exhaustive and systematic discussion by Siegfried, ‘Die Aussprache des Hebr. bei Hieronymus,’ in ZAW. 1884, pp. 34–83.
4. It is, however, doubtful if the LXX always consciously aimed at reproducing the actual differences of sound.
5. The modern Samaritans, however, in reading their Hebrew Pentateuch pronounce שׂ invariably as שׁ.
6. The original value of ס, and its relation to the original value of שׂ and שׁ, is still undetermined, despite the valuable investigations of P. Haupt, ZDMG. 1880, p. 762 f.; D. H. Müller, ‘Zur Geschichte der semit. Zischlaute,’ in the Verhandlungen des Wiener Orient. Congresses, Vienna, 1888, Semitic section, p. 229 ff.; De Lagarde, ‘Samech,’ in the NGGW. 1891, no. 5, esp. p. 173; Aug. Müller, ZAW. 1891, p. 267 ff.; Nöldeke, ZDMG. 1893, p. 100 f.; E. Glaser, Zwei Wiener Publicationen über Habaschitisch-punische Dialekte in Südarabien, Munich, 1902, pp. 19 ff.—On the phonetic value of צ‍ see G. Hüsing, OLZ. 1907, p. 467 ff.
7. So at any rate at the time when the present punctuation arose.