Gesenius' Hebrew Grammar/58. The Pronominal Suffixes of the Verb
Cf. the statistics collected by H. Petri, Das Verbum mit Suffixen im Hebr., part ii, in the נביאים ראשנים, Leipzig, 1890. W. Diehl, Das Pronomen pers. suff.... des Hebr., Giessen, 1895. J. Barth, ‘Beiträge zur Suffixlehre des Nordsem.,’ AJSL. xvii (1901), p. 205 f. Brockelmann, Semit. Sprachwiss., i. 159 f.; Grundriss, p. 638 ff.
58a 1. The pronominal suffixes appended to the verb express the accusative of the personal pronoun. They are the following:—
|To a form ending in||To a form in the Perf.||To a form in the Imperf.|
|a Vowel.||ending in a Consonant.||ending in a Consonant.|
|Sing.||1. com.||־֫נִי||־ַ֫ ני (in pause ־ָ֫ נִי)||־ֵ֫ נִי||me.|
|2. m.||־֫ךָ||־ְךָ (in pause||־ֶ֥֫ ךָ, also ־ָךְ)||thee.|
|f.||־ךְ||־ֶ֫ ךְ ־ֵךְ, rarely ־ָךְ||־ֵךְ|
|3. m.||־֫הוּ, וֹ||־ָ֫ הוּ, (הֹ) וֹ||־ֵ֫ הוּ||him.|
|Plur.||1. com||־֫נוּ||־ָ֫ נוּ||־ֵ֫ נוּ||us.|
|2. m.||־כֶם||־ְכֶם||you (vos).|
|3. m.||הֶם,  ם||־ָם (from ־ָ֫ הֶם, ־ָ֫ ם||־ֵם (from ־ֵ֫ הֶם)||eos.|
|poet.||־֫מוֹ||־ָ֫ מוֹ||־ֵ֫ מוֹ|
|f.||־ן||־ָן, ־ַ֫ ן||.....1||eas.|
58b 2. That these suffixes are connected with the corresponding forms of the personal pronoun (§ 32) is for the most part self-evident, and only a few of them require elucidation.
58c The suffixes נִי, נוּ, הוּ, הָ (and ךָ, when a long vowel in an open syllable precedes) never have the tone, which always rests on the preceding syllable; on the other hand, כֶם and הֶם always take the tone.
58d In the 3rd pers. masc. ־ָ֫ הוּ, by contraction of a and u after the rejection of the weak ה, frequently gives rise to ô (§23k), ordinarily written וֹ, much less frequently הֹ (see §7c). In the feminine, the suffix הָ should be pronounced with a preceding a (cf. below, f, note), as ־ָ֫ הָ or ־ֶ֫ הָ, on the analogy of āhû; instead of ־ָ֫ הָ, however, it was simply pronounced ־ָהּ, with the rejection of the final vowel, and with Mappiq, since the ה is consonantal; but the weakening to ־ָה is also found, see below, g.
58e 3. The variety of the suffix-forms is occasioned chiefly by the fact that they are modified differently according to the form and tense of the verb to which they are attached. For almost every suffix three forms may be distinguished:
(a) One beginning with a consonant, as ־֫נִי, ־֫הוּ, ו (only after î), ־֫נוּ, ם (הֶם), &c. These are attached to verbal forms which end with a vowel, e.g. יִקְטְל֫וּנִי; קְטַלְתִּ֫יהוּ, for which by absorption of the ה we also get קְטַלְתִּיו, pronounced qeṭaltîu; cf. §8m.
58f (b) A second and third with what are called connecting vowels (־ַ֫ נִי, ־ֵ֫ נִי), used with verbal forms ending with a consonant (for exceptions, see §59g and §60e). This connecting vowel is a with the forms of the perfect, e.g. קְטָלַ֫נִי, קְטָלָ֫נוּ, קְטָלָם (on קְטָלֵךְ, the ordinary form of the 3rd masc. perf. with the 2nd fem. suffix, cf. below, g); and e (less frequently a) with the forms of the imperfect and imperative, e.g. יִקְטְלֵ֫הוּ, קָטְלֵם; also with the infinitive and participles, when these do not take noun-suffixes (cf. §61a and h). The form וֹ also belongs to the suffixes of the perfect, since it has arisen from ־ָ֫ הוּ (cf., however, §60d). With ךָ, כֶם, the connecting sound is only a vocal Šewâ, which has arisen from an original short vowel, thus ־ְךָ, ־ְכֶם, e.g. קְטָֽלְךָ (qeṭālekhā), or when the final consonant of the verb is a guttural, ־ֲךָ, e.g. שְׁלָֽחֲךָ. In pause, the original short vowel (ă) reappears as Seghôl with the tone ־ֶ֫ ךָ (also ־ָ֫ ךְ, see g). On the appending of suffixes to the final וּן of the imperfect (§47m), see §60e.
58g Rem. 1. As rare forms may be mentioned sing. 2nd pers. masc. ־ְכָה Gn 277, 1 K 1844, &c., in pause also ־ֶ֫ כָּה (see below, i); fem. כִי, ־ֵ֫ כִי ψ 1034, 1376. Instead of the form ־ֵךְ, which is usual even in the perfect (e.g. Ju 420, Ez 2726), ־ָךְ occurs as fem. Is 609 (as masc. Dt 617, 2845, Is 3019, 555 always in pause); with Munaḥ Is 546, Jer 2337.—In the 3rd masc. הֹ Ex 3225, Nu 238; in the 3rd fem. ־ָה without Mappîq (cf. §91e) Ex 23, Jer 4419; Am 111, with retraction of the tone before a following tone-syllable, but read certainly שָׁמַר לָנֶ֫צַח.—The forms ־֫מוֹ, ־ָ֫ מוֹ, ־ֵ֫ מוֹ occur 23 times, all in poetry (except Ex 2331) [viz. with the perfect Ex 1510, 2331, ψ 736; with the imperfect Ex 155 (מוּ for מוֹ), 157,9,9,12,15,17,17, ψ 25, 2110,13, 225, 4517, 806, 14010; with the imperative ψ 511, 5912,12, 8312]. On the age of these forms, see §91l 3; on ־ַן and ־ָן as suffixes of the 3rd fem. plur. of the imperfect, §60d.—In Gn 489 קָֽחֶם־נָא (cf. וַיַּכֶּם־שָׁם 1 Ch 1411 according to Baer), ־ֵם has lost the tone before Maqqeph and so is shortened to ־ֶם.—In Ez 448 וַתְּשִׂימוּן is probably only an error for וַתְּשִׂימוּם.
58h 2. From a comparison of these verbal suffixes with the noun-suffixes (§ 91) we find that (a) there is a greater variety of forms amongst the verbal than amongst the noun-suffixes, the forms and relations of the verb itself being more various;—(b) the verbal suffix, where it differs from that of the noun, is longer; cf. e.g. ־֫נִי, ־ַ֫ נִי, ־ֵ֫ נִי (me) with ־ִי (my). The reason is that the pronominal object is less closely connected with the verb than the possessive pronoun (the genitive) is with the noun; consequently the former can also be expressed by a separate word (את in אֹתִי, &c.).
58i 4. A verbal form with a suffix gains additional strength, and sometimes intentional emphasis, when, instead of the mere connecting vowel, a special connecting-syllable (ăn) is inserted between the suffix and the verbal stem. Since, however, this syllable always has the tone, the ă is invariably (except in the 1st pers. sing.) modified to tone-bearing Seghôl. This is called the Nûn energicum (less suitably demonstrativum or epentheticum), and occurs principally (see, however, Dt 3210 bis) in pausal forms of the imperfect, e.g. יְבָֽרֲכֶֽנְהוּ he will bless him (Ps. 7215, cf. Jer 522), אֶתְּקֶ֫נְךָּ Jer 2224; יְכַ֫בְּדָ֥נְנִי he will honour me (ψ 5023) is unusual; rarely in the perfect, Dt 2413 בֵּֽרְכֶךָּ. On examples like דָּנַ֫נִּי Gn 306, cf. §26g, §59f. In far the greatest number of cases, however, this Nûn is assimilated to the following consonant (נ, כ), or the latter is lost in pronunciation (so ה), and the Nûn consequently sharpened. Hence we get the following series of suffix-forms:— 1st pers. ־ַ֫ נִּי (even in pause, Jb 714, &c.), ־ֶ֫ נִּי (for ־ַ֫ נְנִי, ־ֶ֫ נְנִי).
2nd pers. ־ֶ֫ ךָּ (Jer 2224 in pause ־ֶנְךָּ) and, only orthographically different,
3rd pers. ־ֶ֫ נּוּ (for ־ֶ֫ נְהוּ), fem. ־ֶ֫ נָּה for ־ֶ֫ נְהָ.
[1st pers. plur. ־ֶ֫ נּוּ (for ־ֶ֫ נְנוּ), see the Rem.]
In the other persons Nûn energetic does not occur.
58k Rem. The uncontracted forms with Nûn are rare, and occur only in poetic or elevated style (Ex 152, Dt 3210 [bis], Jer 522, 2224); they are never found in the 3rd fem. sing. and 1st plur. On the other hand, the contracted forms are tolerably frequent, even in prose. An example of ־ֶ֫ נּוּ as 1st plur. occurs perhaps in Jb 3115 [but read ־ֵנוּ and cf. §72cc], hardly in Ho 125; cf. הִנֶּ֫נּוּ behold us, Gn 4416, 5018, Nu 1440 for הִנְנוּ (instead of הִנְּנוּ; see §20m).—In Ez 412 the Masora requires תְּעֻגֶ֫נָה, without Dageš in the Nûn.
58l That the forms with Nûn energicum are intended to give greater emphasis to the verbal form is seen from their special frequency in pause. Apart from the verb, however, Nûn energicum occurs also in the union of suffixes with certain particles (§100o).
This Nûn is frequent in Western Aramaic. In Arabic the corresponding forms are the two energetic moods (see §48b) ending in an and anna, which are used in connexion with suffixes (e.g. yaqtulan-ka or yaqtulanna-ka) as well as without them.
- According to Diehl (see above), p. 61, כֶם occurs only once with the perfect (see §59e), 7 times with the imperfect, but never in pre-exilic passages, whereas the accus. אֶתְכֶם occurs 40 times in Jer. and 36 times in Ezek.—הֶם occurs only once as a verbal suffix (Dt 3226, unless, with Kahan, Infinitive u. Participien, p. 13, אַפְאִיהֵם from פָּאַהּ is to be read), while the forms כֶן (2nd f. pl.) and ־ֵן and הֶן (3rd f. pl.), added by Qimḥi, never occur.
- We have kept the term connecting vowel, although it is rather a superficial description, and moreover these vowels are of various origin. The connective a is most probably the remains of the old verbal termination, like the i in the 2nd pers. fem. sing. קְטַלְתִּ֫יהוּ. Observe e.g. the Hebrew form qeṭāl-ani in connexion with the Arabic qatala-ni, contrasted with Hebrew qeṭālat-ni and Arabic qatalat-ni. König accordingly prefers the expression ‘vocalic ending of the stem’, instead of ‘connecting syllable’. The connective ē, aŒ¬, as Prätorius (ZDMG. 55, 267 ff.) and Barth (ibid. p. 205 f.) show by reference to the Syriac connective ai in the imperf. of the strong verb, is originally due to the analogy of verbs ל״י (מְחֵנִי = מְחֵינִי from meḥainî), in which the final ê was used as a connecting vowel first of the imperat., then of the impf. (besides many forms with a, §60d), and of the infin. and participle.
- Thus in Ps 2 ־מוֹ occurs five times [four times attached to a noun or preposition, §§91f, 103c], and ־ֵם only twice.
- It is, however, a question whether, instead of a connecting syllable, we should not assume a special verbal form, analogous to the Arabic energetic mood (see l, at the end) and probably also appearing in the Hebrew cohortative (see the footnote on §48c).—As M. Lambert has shown in REJ. 1903, p. 178 ff. (‘De l’emploi des suffixes pronominaux...’), the suffixes of the 3rd pers. with the impf. without waw in prose are ־ֶ֫ נּוּ and ־ֶ֫ נָּה, but with waw consec. ־ֵ֫ הוּ and ־ֶ֫ הָ or ־ָהּ; with the jussive in the 2nd and 3rd pers. always ־ֵ֫ הוּ, ־ֶ֫ ההָ, the 1st pers. more often ־ֶ֫ נּוּ than ־ֵ֫ הוּ, and always ־ֶ֫ נָּה.
- According to Barth ‘n-haltige Suffixe’ in Sprachwiss. Untersuchungen, Lpz. 1907, p. 1 ff., the connecting element, as in Aramaic, was originally in, which in Hebrew became en in a closed tone-syllable.
- So König, Lehrgeb., i. p. 226.
- On נוֹ = ־ֶנּוּ Nu 2313, see §67o.