Gesenius' Hebrew Grammar/21. The Aspiration of the Tenues
21a The harder sound of the six Begadkephath letters, indicated by a Dageš lene, is to be regarded, according to the general analogy of languages, as their older and original pronunciation, from which the softer sound was weakened (§6n and § 13). The original hard sound is maintained when the letter is initial, and after a consonant, but when it immediately follows a vowel or Šewā mobile it is softened and aspirated by their influence, e.g. פָּרַץ pāraṣ, יִפְרֹץ yiphrōṣ, כֹּל kōl, לְכֹל lekhōl. Hence the Begadkephath take Dageš lene
21b (1) at the beginning of words: (a) without exception when the preceding word ends with a vowelless consonant, e.g. עַל־כֵּן ʾal-kēn (therefore), עֵץ פְּרִי ʿēṣ perî (fruit-tree); (b) at the beginning of a section, e.g. בְּרֵאשִׁית Gn 11, or at the beginning of a sentence, or even of a minor division of a sentence after a distinctive accent (§15d), although the preceding word may end with a vowel. The distinctive accent in such a case prevents the vowel from influencing the following tenuis, e.g. וַיְהִ֕י כַּֽאֲשֶׁר and it was so, that when, Ju 115 (but וַֽיְהִי־כֵן Gn 17).
21c Rem. 1. The vowel letters ה, י, ו, א, as such, naturally do not close a syllable. In close connexion they are therefore followed by the aspirated Begadkephath, e.g. וּמָ֣צָא בָהּ, &c. On the other hand, syllables are closed by the consonantal ו and י (except קַו־תֹ֫הוּ Is 3411; שָׁלֵו֣בָֿהּ Ez 2342; אֲדֹנָי בָֿם ψ 6818), and by הּ with Mappîq; hence e.g. there is Dageš lene in עָלַי֣ פִּיהֶם and always after יְהֹוָה, since the Qerê perpetuum of this word (§ 17) assumes the reading אֲדֹנָי.
21d 2. In a number of cases Dageš lene is inserted, although a vowel precedes in close connexion. This almost always occurs with the prefixes בְּ and כְּ in the combinations בְּב, כְּכ, בְּפ (i.e. when a Begadkephath with Šewâ precedes the same or a kindred aspirate) and בְּם (see Baer, L. Psalmorum, 1880, p. 92,
on ψ 233); cf. e.g. 1 S 251, Is 109, ψ 342, Jb 192; כְג is uncertain; בְד, כְד, and בְכ according to David Qimḥi do not take Dageš, nor כְג, כב, and כְפ according to the Dikduke ha-ṭeeamim, p. 30. Sometimes the Begadkephath letters, even with a full vowel, take Dageš before aspirant (and even before ח in בַּֽחֲמִשָּׁה 1 K 1232); cf. the instances mentioned above, §20e (mostly tenues before א). In all these cases the object is to prevent too great an accumulation of aspirates. The LXX, on the other hand, almost always represent the כ and פ, even at the beginning of a syllable, by χ and φ; Χερούβ, Χαλδαῖοι, Φαρφάρ, &c.—The forms כַּֽדְכֹד (after וְשַׂמְתִּ֫י) Is 5412, and כַּֽלְכֵל (after וְנִלְאֵ֫יתִי) Jer 209 are doubly anomalous.
21e (2) In the middle of words after Šewâ quiescens, i.e. at the beginning of a syllable immediately after a vowelless consonant, e.g. יִרְפָּא yirpā (he heals), קְטַלְתֶּם ye have killed; but after Šewâ mobile, e.g. רְפָא rephā (heal thou), כָּֽבְדָה she was heavy.
21f On קָטַלְתְּ, וַיִּשְׁבְּ and similar forms, see §10i.
Whether Šewâ be vocal and consequently causes the aspiration of a following tenuis, depends upon the origin of the particular form. It is almost always vocal
(a) When it has arisen from the weakening of a strong vowel, e.g. רִדְפוּ pursue ye (not רִדְפּוּ) from רְדֹף; מַלְכֵי (not מַלְכֵּי), because originally mălăkhê, but מַלְכִּי from the ground-form malk.
Rem. Forms like שָׁלַ֫חַתְּ thou (fem.) hast sent, in which we should expect an aspirated ת after the vowel, cf. וַיִּ֫חַדְּ Ex 189, have arisen from שָׁלַחְתְּ, יִחְדְּ, &c.; Pathaḥ being here simply a helping vowel has no influence on the tenuis; cf. §28e.
- Cf. Delitzsch, Ztschr. f. luth. Theol. u. Kirche, 1878, p. 585 ff.
- Also L. Proverbiorum, 1880, Praef. p. ix; and Dikduke ha-ṭeamim, p. 30 (in German in König’s Lehrgeb., i. p. 62).
- The exceptions יָקְתְאֵל Jos 1538 (see Minḥat shay, on this passage), 2 K 147, and יָקְדְעָם Jos 1556 may perhaps be due to the character of the ק.