Gesenius' Hebrew Grammar/93. Paradigms of Masculine Nouns
93a Masculine nouns from the simple stem may, as regards their form and the vowel changes connected with it, be divided into four classes. A synopsis of them is given on pp. 264, 265, and they are further explained below. Two general remarks may be premised:
(a) That all feminines without a distinctive termination (§122h) are treated like these masculine nouns, e.g. חֶ֫רֶב f. sword, like מֶ֫לֶךְ m. king, except that in the plural they usually take the termination ־וֹת; thus חֲרָבוֹת, constr. חַרְבוֹת (and so always before suffixes, see §95).
93b (b) That in the plural of the first three classes a changeable vowel is always retained even before the light suffixes as a lengthened pretonic vowel, whenever it also stands before the plural ending ־ִים. All suffixes, except כֶם, כֶן, הֶם, הֶן (־ֵיכֶם, ־ֵיכֶן, ־ֵיהֶם, ־ֵיהֶן), are called light. Cf. §92e.
93c 1. Paradigm I comprises the large class of segholate nouns (§84aa–a). In the first three examples, from a strong stem, the ground-forms, mălk, sĭphr, qŭdš have been developed by the adoption of a helping Seghôl to מֶ֫לֶךְ (with ă modified to è), סֵ֫פֶר (ĭ lengthened to ē), קֹ֫דֶשׁ (ŭ lengthened to ō). The next three examples, instead of the helping Seghôl, have a helping Pathaḥ, on account of the middle (d, f) or final guttural (e). In all these cases the constr. st. sing. coincides exactly with the absolute. The singular suffixes are added to the ground-form; but in c and f an ŏ takes the place of the original ŭ, and in d and f the guttural requires a repetition of the ă and ŏ in the form of a Ḥaṭeph (נַֽעֲרִי, פָּֽעֳלִי); before a following Šewâ this Ḥaṭeph passes into a simple helping vowel (ă, ŏ), according to §28c; hence נַֽעַרְךָ, &c.
93d In the plural an a-sound almost always appears before the tone-bearing affix ־ִים (on the analogy of forms with original a in the second syllable; cf. §84aa), in the form of a pretonic Qameṣ, whilst the short vowel of the first syllable becomes vocal Šewâ. The original a of the 2nd syllable is elided in the construct state, so that the short vowel under the first radical then stands in a closed syllable. The omission of Dageš in a following Begadkephath (מַלְכֵי, not מַלְכֵּי, &c.) is due to the loss of a vowel between ל and כ. On the other hand, the pretonic Qames of the absolute state is retained before the light plural suffixes, whilst the grave suffixes are added to the form of the construct state.—The ending of the absolute state of the dual is added, as a rule, to the ground-form (so in a–d and h, but cf. k). The construct state of the dual is generally the same as that of the plural, except, of course, in cases like m.
93e Paradigms g and h exhibit forms with middle u and i (§84ac, γ and δ); the ground forms maut and zait are always contracted to môth, zêth, except in the absol. sing., where u and i are changed into the corresponding consonants ו and י.
Paradigm i exhibits one of the numerous forms in which the contraction of a middle u or i has already taken place in the absol. sing. (ground-form šauṭ).
Paradigm k is a formation from a stem ל״ה (§84ac, ε).
93f Paradigms l, m, n are forms from stems ע״ע, and hence (see §67a) originally biliteral, yam, ʾim, ḥuq, with the regular lengthening to יָם, אֵם, חֹק. Before formative additions a sharpening, as in the inflexion of verbs ע״ע, takes place in the second radical, e.g. אִמִּי, יַמִּים, &c. (see §84ac, β).
93g 1. A. On I. a and d (ground-form qaṭl). In pause the full lengthening to ā generally takes place, thus כָּ֫רֶם vineyard, נָ֫עַר, זָ֫רַע seed (from זֶ֫רַע), and so always (except ψ 4811), in אֶ֫רֶץ earth with the article, הָאָ֫רֶץ, according to §35o (cf. also in the LXX the forms Ἀβέλ, Ἰαφέθ for הֶ֫בֶל, יֶ֫פֶת). However, the form with è is also sometimes found in pause, along with that in ā, e.g. חֶ֫סֶד together with חָ֫סֶד; and very frequently only the form with Seghôl, e.g. מֶ֫לֶךְ, דֶּ֫שֶׁא grass, נֶ֫צַח perpetuity, פֶּ֫לֶא a wonder, צֶ֫דֶק righteousness, קֶ֫דֶם the East, יֶ֫שַׁע help, &c.—With two Seghôls, although with a middle guttural, we find לֶ֫חֶם bread (in pause לָ֫חֶם) and רֶ֫חֶם womb (in pause רָ֫חֶם), besides רַ֫חַם Ju 530 (in pause רָ֫חַם). A helping Seghôl always stands before a final א, as דֶּ֫שֶׁא, טֶ֫נֶא (with suff. טַנְאֲךָ), פֶּ֫לֶא, פֶּ֫רֶא (also written פֶּ֫רֶה), except in גַּיְא, see v.
93hB. The constr. st. is almost always the same as the absolute. Sometimes, however, under the influence of a final guttural or ר, Pathaḥ appears in the second syllable as the principal vowel (see below, §93s), e.g. גְּבַר ψ 1826; זְרַע
Paradigms of Masculine Nouns.
|"||with light suff.||מַלְכִּי||סִפְרִי||קָדְשִׁי||נַֽעֲרִי||נִצְתִי||פָּֽעֳלִי|
|"||with grave suff.||מַלְכְּכֶם||סִפְרְכֶם||קָדְשְׁכֶם||נַעַֽרְכֶם||נִצְחֲכֶם||פָּֽעָלְכֶם|
|"||with light suff.||מְלָכַי||סְפָרַי||[קֳדָשַׁי]||נְעָרַי||נְצָחַי||פְּעָלַי|
|"||with grave suff.||מַלְכֵיכֶם||סִפְרֵיכֶם||קָדְשֵׁיכֶם||נַֽעֲרֵיכֶם||נִצְחֵיכֶם||פָּֽעֳלֵיכֶם|
|(word)||(wise)||(an old man)||(shoulder)||(court)||(field)|
|"||with light suff.||דְּבָרִי||חֲכָמִי||זְקֵנִי||כְּתֵפִי||חֲצֵרִי||שָׂדִי|
|"||with grave suff.||דְּבַרְכֶם||חֲכַמְכֶם|
|"||with light suff.||דְּבָרַי||חֲכָמַי||זְקֵנַי||חֲצֵרַי||פָּנַי|
|"||with grave suff.||דִּבְרֵיכֶם||חַכְמֵיכֶם||זִקְנֵיכֶם||חַצְרֵיכֶם||פְּנֵיכֶם|
|"||with light suff.||מוֹתִי||זֵיתִי||שׁוֹטִי||פִּרְיִי||יַמִּי||אִמִּי||חֻקִּי|
|"||with grave suff.||מֽוֹתְכֶם||זֵֽיתְכֶם||שֽׁוֹטְכֶם||פֶּרְיְכֶם||יַמְּכֶם||אִמְּכֶם||חָקְכֶם|
|"||with light suff.||זֵיתַי||שׁוֹטַי||(kids)||יַמַּי||אִמּוֹתַי||חֻקַּי|
|"||with grave suff.||זֵיֽתֵיכֶם||שֽׁוֹטֵיכֶם||יַמֵּיכֶם||אִמּֽוֹתֵיכֶם||חֻקֵּיכֶם|
|"||with light suff.||עֽוֹלָמִי||אֹֽיְבִי||חֹזִי||פְּקִידִי||כְּתָבִי|
|"||with grave suff.||עֽוֹלַמְכֶם||אֹֽיִבְכֶם||חֹֽזְכֶם||פְּקִֽירְכֶם||כְּתָֽבְכֶם|
|"||with light suff.||עֽוֹלָמַי||אֹֽיְבַי||חֹזַי||פְּקִידַי||[כְּתָבַי]|
|"||with grave suff.||עֽוֹלְמֵיכֶם||אֹֽיְבֵיכֶם||חֹֽזֵיכֶם||פְּקִֽידֵיכֶם||עֲנִיֵּיכֶם||[כְּתָֽבֵיכֶם]|
|(pair of tongs)||(balance)||(two weeks)|
93i C. The ־ָה locale is, according to §90i, regularly added to the already developed form, e.g. נֶ֫גְדָה ψ 11614.18: הַפֶּ֑תְחָה Gn 196, to the door; but also with a firmly closed syllable נֶ֫גְבָּה Ex 4024; under the influence of a guttural or ר, חַ֫דְרָה, אַ֫רְצָה, in pause אָ֫רְצָה (cf. גָּ֫זְרָה 1 Ch 1416, from גֶּ֫זֶר).
93k D. The suffixes of the singular are likewise added to the ground-form, but forms with middle guttural take Ḥaṭeph-Pathaḥ instead of the Šewâ quiescens; נַֽעֲרִי, &c. (but also לַחְמִי, זַעְמִי, &c.). In a rather large number of qăṭl-forms, however, before suffixes in the sing., as well as in the constr. st. plur. and dual, the ă of the first syllable is attenuated to ĭ, thus בִּטְנִי my womb, יִתְרוֹ; so in בֶּ֫גֶד, בֶּ֫צַע, גֶּ֫זַע, זֶ֫בַח, טֶ֫בַח, פֶּ֫שַׁע, פֶּ֫תַח, צֶ֫דֶק, קֶ֫בֶר, קֶ֫רֶב, רֶ֫שַׁע, שֶׁ֫מֶשׁ, and many others. In some cases of this kind besides the form with ă there most probably existed another with original ĭ in the first syllable; thus certainly with יֵ֫שַׁע beside יֶ֫שַׁע, נֵ֫צַח beside נֶ֫צַ֫ח, &c. (According to the Diqduqe ha-ṭeamim, § 36, the absolute st. in such cases takes è, the constr. ē; cf. נֶ֫דֶר Nu 304 (absol.) and נֵ֫דֶר 3010 (constr.); שֶׁ֫בֶר Lv 2420 (absol.) and שֵׁ֫בֶר Am 66 (constr.). According to this theory פֵּ֫לֶא (so the best authorities) Is 95 would be the constr. st., although the accentuation requires an absol. st.)—A weakening of the firmly closed syllable occurs in בִּגְדִי, &c. from בֶּ֫גֶד and יִקְבֶ֫ךָ Dt 1514, 1613, in both cases evidently owing to the influence of the palatal in the middle of the stem. With Seghôl for ĭ: הֶבְלִי, יֶשְׁעֲךָ, נֶגְדִּי, &c.
93l E. In the plural the termination וֹת is found as well as ־ִים, e.g. נְפָשׁוֹת, עֲצָמוֹת together with נְפָשִׁים (Ez 1320 [but read חָפְשִׁים; see comm.]), &c., constr. st. נַפְשׁוֹת. Other nouns have only the ending וֹת, e.g. אֲרָצוֹת, constr. אַרְצוֹת from אֶ֫רֶץ. Without Qaṃeṣ before the ending ־ִים we find רַֽחֲמִים '(bowels) mercy. On the numerals עֶשְׂרִים twenty, &c., cf. §97f, note 2. Moreover a is not inserted before plural suffixes with the tone on the penultima in אַשְׁרֶ֫יךָ, &c., properly thy happiness! (a word which is only used in the constr. st. pl. and at an early period became stereotyped as a kind of interjection).
93m F. In the constr. st. plural a firmly closed syllable is sometimes found, contrary to the rule, e.g. כַּסְפֵּיהֶם Gn 4225.35; רִשְׁפֵּי Ct 86 (רִשְׁפֵי ψ 764); טַרְפֵּי Ez 179; צִמְדֵּי Is 510, and so always in נִסְכֵּיכֶם Nu 2939, נִסְכֵּיהֶם ψ 164, &c. (on the other hand, according to the best authorities not in חַסְדֵי Is 553, &c., though in ψ 10743 Ginsburg reads חַסְדֵּי); cf. §46d. Even with a middle guttural בַּעְלֵיהֶן Est 117.20.—The attenuation of ă to ĭ also occurs sometimes in this form (see above, k), e.g. זִבְחֵי, &c., even יִלְדֵי Is 574 beside יַלְדֵי Ho 12, &c.
93n G. In the dual absol. beside forms like רַגְלַ֫יִם feet, with suff. רַגְלֶ֫יךָ, רַגְלָיו, &c. אַלְפַּ֫יִם two thousand, נַֽעֲלַ֫יִם sandals, בִּרְכַּ֫יִם knees (ă attenuated to ĭ, constr. st. בִּרְכֵּי with a firmly closed syllable), with suffixes בִּרְכַּי, &c. (cf., however, בִּרְכֵיהֶם Ju 76), forms with pretonic Qameṣ are also found (in consequence of the tendency to assimilate the dual to the plural in form: so König, Lehrgeb., ii. 17), as קְרָנַ֫יִם horns, with suff. קְרָנָיו (Dn 83 ff.; elsewhere always קַרְנַ֫יִם, קַרְנָיו, &c.), and so always דְּלָתַ֫יִם, constr. st. דַּלְתֵי folding-doors, דְּרָכַ֫יִם (?) double way.
93o 2. On Paradigms b and e. With a final א rejected (but retained orthographically) we find חֵטְא sin. An initial guttural before suffixes generally receives Seghôl instead of the original ĭ, e.g. חֶלְקִי, עֶזְרִי, &c., so in the constr. st. plur. עֶגְלֵי, &c.; חֵטְא forms חֲטָאֵי 2 K 1029, &c., retaining the Qameṣ of חֲטָאִים before the weak א.—The pausal forms סָ֫תֶר and שָׁ֫בֶט (out of pause always סֵ֫תֶר, שֵׁ֫בֶט) go back to by-forms סֶ֫תֶר, שֶׁ֫בֶט.—On עִשְּׂבוֹת (constr. st. plur. of עֵ֫שֶׂב) Pr 2725, cf. §20h; שִׁקְמִים sycamores, without Qameṣ before the termination ־ִים (see above, l), is probably from the sing. שִׁקְמָה found in the Mišna.
93p 3. On Paradigms c and f. קשְׁטְ occurs in Pr 2221 without a helping vowel; with a middle guttural פֹּ֫עַל, &c., but with ה also אֹ֫הֶל, בֹּ֫הֶן; with a final guttural גֹּ֫בַהּ, רֹ֫בַע &c., but with א, גֹּ֫מֶא; with a firmly closed syllable אָסְפֵּי Mi 71.
93q Before suffixes the original ŭ sometimes reappears in the sing., e.g. גֻּדְלוֹ (ψ 1502) beside גָּדְלוֹ, from גֹּ֫דֶל greatness; סֻבֳּלוֹ (with Dageš forte dirimens, and the ŭ repeated in the form of a Ḥaṭeph-Qameṣ, cf. §10h) Is 93, &c.; גֻּשְׁמָהּ Ez 2224.—Corresponding to the form פָּֽעָלְכֶם pŏʿŏlekhèm we find קָֽטָבְךָ Ho 1314, even without a middle guttural; similarly קָֽטֳנִי (so Jablonski and Opitius) 1 K 1210, 2 Ch 1010, from קֹ֫טֶן little finger; but the better reading is, no doubt, קָֽטָנִּי (so ed. Mant., ‘the ק proleptically assuming the vowel of the following syllable’; König, Lehrgeb., ii. 69), and the form is to be derived, with König, from קְטֹן, not qŭtŭn, as Brockelmann quotes him, in Grundriss, p. 103. The reading קָֽטֳנִּי (Baer and Ginsburg) is probably not due to a confusion of the above two readings, but ־ֳ is merely intended to mark the vowel expressly as ŏ. In the forms פֹּֽעֲלוֹ Is 131 (for פָּֽעֳלוֹ) and תֹּֽאֲרוֹ Is 5214 (for תָּֽאֳרוֹ 1 S 2814), the lengthening of the original ŭ to ō has been retained even before the suffix; cf. §63p and §74h (בְּמֹצַֽאֲכֶם Gn 3220).—In the same way ō remains before ־ָה locale, e.g. גּ֫ׄרְנָה, הָאֹהֱלָה Gn 186, 2467, &c. Dissimilation of the vowel (or a by-form נֶ֫כַח?) seems to occur in נִכְחוֹ Ex 142, Ez 469, for נָכְחוֹ.
93r In the absol. st. plur. the original ŭ generally becomes Šewâ before the Qameṣ, e.g. בְּקָרִים from בֹּ֫קֶר morning, פְּעָלִים works, רְמָחִים lances, שְׁעָלִים handfuls (constr. st. שַֽׁעֲלֵי Ez 1319); on the other hand, with an initial guttural the ŭ-sound reappears as Ḥaṭeph Qameṣ, e.g. חֳדָשִׁים months, עֳפָרִים gazelles, אֳרָחוֹת ways; and so even without an initial guttural, הַגֳּרָנוֹת the threshing-floors, 1 S 231, Jo 224; קָֽדָשִׁים sanctuaries, and שָֽׁרָשִׁים roots (qŏdhāšîm, &c., with ŏ for ־ֳ); also קָֽדָשַׁי [but קֳדָשֶׁ֫יךָ, קֳדָשָׁיו, once קָֽ׳], where, however, the reading frequently fluctuates between קָֽ׳ and קֳ׳; with the article הַקֳּ׳, בַּקֳּ׳, לַקֳּ׳, according to Baer and Ginsburg. On these forms cf. especially §9v. From אֹ֫הֶל tent, both בָּֽאֳהָלִים and אֹֽהָלִים (cf. §23h and פֹּֽעֲלוֹ above) are found; with light suffixes אֹֽהָלַי, &c.; so from אֹרַח way, אֹֽרְחֹתָיו (also אָרְחֹתַי)—hence only with initial א, ‘on account of its weak articulation’ (König, Lehrgeb., ii. 45). It seems that by these different ways of writing a distinction was intended between the plural of אֹרְחָה caravan, and of אֹרַח way; however, אָרְחוֹת is also found in the former sense (in constr. st. Jb 619) and אֹֽרְחוֹת in the latter (e.g. Jb 1327 according to the reading of Ben Naphtali and Qimḥi); cf. also אֽוֹנִיּוֹת 2 Ch 818 Keth. (אֳנ׳ Qerê).—The constr. st. plural of בֹּ֫הֶן thumb is בְּהֹנוֹת Ju 16 f., as if from a sing. בְּהֹן: of נׄ֫גַהּ brightness, Is 599 נְגׄהוֹת (on these qeṭōl-forms, cf. t).—If אָפְנָיו Pr 2511 is not dual but plural (see the Lexicon) it is then analogous to the examples, given in l and 0, of plurals without a pretonic Qameṣ; cf. בָּטְנִים pistachio nuts, probably from a sing. בָּטְנָה. According to Barth, ZDMG. xlii, 345 f. אָפְנָיו is a sing. (אָפְנַי, the ground-form of אָפְנֶה, with suffix).
93s 4. Besides the forms treated hitherto we have to consider also a series of formations, which have their characteristic vowel under the second radical, as is ordinarily the case in Aramaic (on the origin of these forms see further, §84ae). Thus (a) of the form קְטַל; דְּבַשׁ honey, מְעַט little; in pause, דְּבָשׁ, מְעָט; גְּבַר man (as constr. st., see above, h), ψ 1826 (elsewhere always גֶּ֫בֶר), and infinitives like שְׁכַב (§45c; on קְחַת, see above, h); שְׁכֶם shoulder, ă being modified to è (but in pause שֶׁ֫כֶם); locative שְׁכֶ֫מָה, also שֶׁ֫כְמָה Ho 69. With suffixes in the usual manner שִׁכְמִי, שִׁכְבָהּ Gn 1933.35 (an infin. with suffix, therefore not שִׁכְבָּהּ). On the other hand, the ă is retained in the plur. absol. by sharpening the final consonant: אֲגַמִּים (constr. אַגְמֵי) marshes, הֲדַסִּים myrtles, מְעַטִּים few.
93t (b) Of the form קְטֵל: בְּאֵר a well, זְאֵב wolf, &c.; locative בְּאֵרָה, with suff. בְּאֵרִי, plur. זְאֵבִים, זְאֵבֵי; but בְּאֵרוֹת, constr. בֶּֽאֱרוֹת; on the infin. constr. שְׂאֵת, cf. §76b.
(c) of the form קְטֹל: בְּאשׁ stench (with suff. בָּאְשׁוֹ, just as סֻבְּכוֹ occurs in Jer 47 along with the constr. st. סְבָךְ־ ψ 745; cf. for the Dageš, §20h), perhaps also לְאֹם nation, pl. לְאֻמִּים.
93u 5. Paradigms g–i comprise the segholate forms with middle ו or י: (a) of the form qăṭl with Wāw as a strong consonant, in which cases the original ă is almost always lengthened to ā (Paradigm g), thus מָ֫וֶת, אָ֫וֶן vanity, עָ֫וֶל iniquity, תָּ֫וֶךְ midst; with final א, שָׁוְא falsehood; cf. however, also רֶ֫וַח space. In the constr. st. contraction always occurs, מוֹת, &c. (from original maut), and likewise before suffixes מוֹתוֹ, &c. Exception, עָ֫וֶל as constr. st. Ez 2818 (according to Qimḥi) and with suff. עַוְלוֹ. The contraction remains also in all cases in the plural (but see below, w).
93v (b) Of the form qăṭl with consonantal Yôdh (Paradigm h). With final א, גַּיְא (also גַּי), in Is 404 גֶּיא, in the constr. st. (also absol. Zc 144) גֵּיא (also גֵּי); plur. 2 K 216 and Ez 63 Keth. according to Baer גאות, i.e. doubtless גֵּאוֹת (cf. גֵּֽיאוֹתֶיךָ Ez 358; according to another reading [and so Ginsburg] גיאות, i.e. doubtless גְּיָאוֹת), but in Qerê, and all other passages, נֵּֽאָיוֹת. The uncontracted form (in the absol. st. with helping Ḥireq) remains also before ־ָה locale, e.g. בַּ֫יְתָה (but in the constr. st. e.g. בֵּ֫יתָה יוֹסֵף).—עִירֹה (from עַ֫יִר) Gn 4911 is peculiar, so also שִׁיתוֹ Is 1017 (from שַׁ֫יִת).—In the plural absol. uncontracted forms occur, like חֲיָלִים hosts, עֲיָנוֹת springs, עֲיָרִים young asses, תְּיָשִׁים he-goats, &c.; as constr. st. Pr 828 עִינוֹת for עֵינוֹת.
93w (c) With the contraction of the ו and י even in the absol. st. sing. (Paradigm i). In this way there arise formations which are unchangeable throughout; thus from the ground-form qăṭl: יוֹם (cf., however, §96), סוֹף, שׁוֹר, &c.; with middle Yôdh, חֵיל 1 Ch 913 (elsewhere חַ֫יִל), לֵיל Is 2111 (elsewhere לַ֫יִל, in prose לַ֫יְלָה, see above, §90f); from the ground-form qĭṭl, דִּין, שִׁיר, עִיר (see, however, §96); from the ground-form qŭṭl, גּוּר, רוּחַ &c. The plurals דְּוָדִים pots, שְׁוָקִים streets, שְׁוָרִים oxen, have a strong formation (but for חֲוָחִים 1 S 136 read חוֹרִים as in 1411). Finally, forms with a quiescent middle א also belong to this class, such as רֹאשׁ head (obscured from רָאשׁ=raʾš, see §96) and צֹאן sheep.
93x 6. On Paradigm k: segholate forms from ל״ה stems. Besides the formations mentioned in §84ac, ε, like בֶּ֫כֶה, &c., and שָׂ֫חוּ Ez 475, with the original ו resolved, according to §24d (cf. the constr. plur. חַגְוֵי clefts, Ob 313, &c., and קַצְוֵי ends, ψ 4811, &c., where the ו becomes again a strong consonant, from חֶ֫גֶו and קֶ֫צֶו or חָ֫גוּ and קָ֫צוּ), there occur also (a) commonly, of the ground-form qaṭl, forms like פְּרִי, בְּכִי, גְּדִי, לְחִי, צְבִי, אֲרִי, &c.; in pause פֶּ֫רִי, בֶּ֫כִי, לֶ֫חִי, צֶ֫בִי (cf. §29m), but אֲרִ֑י Ju 1418; with suffixes פִּרְיוֹ (attenuated from păryô), בִּכְיִי ψ 69, but also פֶּרְיְךָ, לֶחְיוֹ &c.; before a grave suffix פְּרִיהֶם, but also פֶּרְיְכֶם. Plur. גְּדָיִים (constr. גְּדָיֵי, see above, o, חֲטָאֵי), אֲרָיִים and אֲרָיוֹת; with softening of the י to א (as elsewhere in בְּלוֹאֵי Jer 3812 for which there is בְּלוֹיֵ in verse 11, according to §8k; עַרְבִיאִים 2 Ch 1711, cf. 267 Keth.; probably in דּֽוּדָאִים, לֻֽלָאוֹת from דּוּדַי and לוּלַי; also חלכאים ψ 1010 Keth., divided into two words by the Masora, is to be referred to a sing. חֶלְכַּי hapless): חֲלָאִים jewels, Ct 72 (from חֲלִי), טְלָאִים lambs, Is 4011 (from טְלִי); but instead of פְּתָאִים and צְבָאִים (from פֶּ֫תִי and צְבִי) the Masora requires פְּתָאיִם and צְבָאיִם; dual: לְחָיַ֫יִם, constr. st. לְחָיֵי, with suff. לְחָיַי, &c. On דַּל door, cf. §95f, and on such formations generally, see Barth on biliteral nouns in ZDMG. 1887, p. 603 ff., and Nominal-bildung (isolated nouns), p. 1 ff.
93y (b) From the ground-form qiṭl, חֲצִי half, in pause חֵ֫צִי, with suff. חֶצְיוֹ, &c.—From stems with middle Wāw arise such forms as אִי (from ʾiwy), עִי, צִי ship, plur. אִיִּים, צִיִּים, &c.; instead of the extraordinary plur. צִים Nu 2424 read with the Samaritan יֽוֹצְאִים, and for בַּצִּים Ez 309 read probably with Cornill אָצִים.
93z (c) From the ground-form qŭṭl sometimes forms like תֹּ֫הוּ, בֹּ֫הוּ (from tŭhw, bŭhw), sometimes like חֳלִי, עֳנִי, and even without an initial guttural דֳּמִי, יֳפִי, צֳרִי (also יְפִי דְּמִי, צְרִי), רָאִי, &c; in pause חֹ֫לִי, &c., with suff. חָלְיוֹ, plur. חֳלָיִים. From עֳפִי branch, there occurs in ψ 10412 the plur. עֳפָאיִם (analogous to פְּתָאיִם, &c., see above, x); the Keeth. evidently intends עֳפָאִים (so Opitius and others). Dual, with suff. דָּלְיָו Nu 247, bucket (from דְּלִי, for דֳּלִי), more correctly, with the Masora, דָּ֣לְיָו with Munaḥ for Metheg. This unusual Metheg is to be treated as following the analogy of the cases mentioned in §9v.
93aa 7. On Paradigms l–n: segholate forms from stems ע״ע (see §84ac, β).
(a) In the qaṭl-form the ă of the contracted formation is sometimes lengthened in the absol. st., sing. as in יָם (so also in the constr. st., except in the combination יַם־סוּף the Red sea; and even before Maqqeph, יָֽם־הַמֶּ֫לַח the salt sea), sometimes it remains short, e.g. פַּת morsel, עַם people, but even these formations generally have Qameṣ in pause, as well as after the article (e.g. הָעָם). Adjectives under the influence of a guttural either have forms like לַחִים, צַחִים, or, with compensatory lengthening, רָעִים, רָעֵי. In the constr. st. חַי living (in the plural חַיִּים also a substantive, life), and דַּי sufficiency, are contracted to חֵי and דֵּי. As a locative form notice הֶ֫רָה to the mountain, Gn 1410 (see §27q) beside הָהָ֫רָה. The stem is expanded to a triliteral form in הֲרָרִי (unless it is simply derived from a by-form הָרָר on the analogy of qătăl-forms) Jer 173 (but in ψ 308 for הַֽרֲרִי read הַֽרֲרֵי) and הַֽרֲרָם Gn 146; plur. constr. הַֽרֲרֵי Nu 237, &c. (but only in poetical passages), with suffix, הֲרָרֶ֫יהָ Dt 89; עֲמָמִים Ju 514 (where, however, read probably בְּעַמֶּ֫ךָ), Neh 922; עַֽמֲמֵי Neh 924: elsewhere עַמִּים, עַמֵּי.—Before suffixes and in the plur. ă is sometimes attenuated to ĭ, e.g. פִּתִּי, פִּתִּים, from פַּת; סִפִּים and סִפּוֹת (also סַפּוֹת 2 S 1728) from סַף. Before ח ă is retained in a virtually sharpened syllable, e.g. פַּחִים traps.
(c) Qŭṭl-forms: חֹק, כֹּל totality, before Maqqeph חָק־, כָּל־, with suff. חֻקִּי, &c., with omission of Dageš forte (according to §20m) always חָקְךָ, חָקְכֶם, but from עֹז, עֻזִּי, עֻזְּךָ, עֻזְּכֶם, for which עָזִּי and עָזְּךָ are also found. חִקְקֵי, expanded to a triliteral form, Ju 515 and Is 101, generally explained as a secondary form of חֻקְקֵי with abnormal weakening of the ŭ to ĭ, is more probably to be referred to a qiṭl-form=Arabic ḥiqq.
93cc The forms with assimilated middle Nûn likewise follow the analogy of Paradigms l–n, e.g. אַף nose, anger (אַפִּי, dual אַפַּ֫יִם, also face) for ʾanp; חֵךְ palate for ḥink, זִקִּים fetters, עֵז goat, plur. עִזִּים, for ʿinz, probably also אֵב green herb, for ʾinb.
93dd 2. Paradigm II comprises all formations with original short vowels, whether in the first or second syllable; cf. §84af–f, and the general laws of formation, §92b–g. Rem. 1. On Paradigms a and b: ground-form qătăl. The lengthening of the second ă to ā is maintained in the constr. st. sing. only in ל״א-forms, e.g. צָבָא army, צְבָא. For the construct forms חֲלֵב milk, לְבֶן־ white, Gn 4912, instead of the ordinary absolutes חָלָב, לָבָן, a secondary form חָלֵב, לָבֵן must be assumed; from עָשָׁן smoke, the constr. st. עֶ֫שֶׁן occurs once, Ex 1918, beside עֲשַׁן, from הָדָר ornament the constr. st. הֶ֫דֶר Dn 1120, beside the common form הֲדַר.—The plur. פָּֽרָשִׁים horses, Is 217 (instead of פְּרָשִׁים, ground-form părăš) is no doubt due to a confusion with the qaṭṭâl-form פָּרָשׁ horseman.
93ee A. Sometimes a sharpening of the third radical takes place, in order to keep the preceding vowel short, e.g. גְּמַלִּים camels, קְטַנִּים small ones, פְּלַגּוֹת brooks (see §20a).—The attenuation of the ă of the first syllable to ĭ does not take place in the constr. st. plur. as a rule after an initial guttural, as חַכְמֵי, עַנְוַי, but חִזְקֵי, and never before a middle guttural, e.g. נַֽהֲרֵי; nor (according to König, owing to the influence of the nasal) in the non-guttural forms זַנְבוֹת tails, כַּנְפוֹת, and (in the dual) כַּנְפֵי wings, from זָנָב, כָּנָף.—The dual נַֽהֲרַ֫יִם from נָהָר river, shows an abnormal omission of the lengthening of the ă before a tone-bearing termination, but cf. §88c.
93ff B. From ע״ע stems, forms like חָלָל, עָנָן, &c., belong to this class.
93gg C. The few nouns of the ground-form qĭṭăl follow the same analogy, such as לֵבָב heart, שֵׁכָר strong drink, עֵנָב grape, &c. From שֵׂעָר hair, in the constr. st. besides שְׂעַר the form שַׂ֫עַר is also found (perhaps a survival of a secondary form like those in Paradigm I, d); so from צֵלָע rib, צֶ֫לַע and even צֵ֫לַע 2 S 1613 (so ed. Mant., Ginsb.; but Baer צֶ֫לַע), both, probably, old secondary forms (also used for the absol. st.) of צֵלָע; cf. also צַלְעִי and צַלְעוֹ, as well as the constr. st. plur. צַלְעוֹת; also from נֵכָר strangeness, the constr. st. נֵ֫כַר־ is found, Dt 3116.
93hh 2. On Paradigms c–e: ground-form qăṭĭl, developed to qāṭēl; with a final guttural, e.g. שָׂבֵעַ satisfied. In the constr. st. the original ĭ of the second syllable, probably on the analogy of the forms discussed in §69c, becomes ă, e.g. זְקַן, חֲדַל, חֲסַר, &c., but not before suffixes, כְּתֵפִי, &c., nor in forms from ל״א stems, e.g. מָלֵא full, מְלֵא; cf., moreover, עֲקֵב Gn 2526 from עָקֵב heel, and אֲבֶל־ ψ 3514, mourning. Paradigm d represents forms which in the constr. st. instead of the ordinary כְּתַף, &c., have a segholate form, as אֶ֫רֶךְ, גֶּ֫דֶר, יֶ֫רֶךְ, גֵּ֫זֶל, עֶ֫רֶל (Ez 449), constr. st. of אָרֵךְ long, גָּדֵר wall, יָרֵךְ thigh, גָּזֵל robbery, עָרֵל uncircumcised. In Is 1114 בְּכָתֵף would be altogether without precedent as a constr. st. (for בְּכֶ֫תֶף); most probably the absol. st. is intended by the Masora (according to Nöldeke, Gött. Gel. Anzeigen, 1871, No. 23 [p. 896] for בכ׳ אֶחָד with one shoulder, i.e. shoulder to shoulder); [cf. Driver, Tenses, § 190, Obs.].
93ii In the plur. constr. the ē lengthened from ĭ is frequently retained in verbal adjectives of this formation, e.g. שְׁכֵחֵי, שְׂמֵחֵי, אֲבֵלֵי, יְשֵׁנֵי, חֲפֵצֵי; cf. also יְתֵֽדֹתָיו (with ē under the protection of the secondary tone) from יָתֵד tent-peg. On the other hand from יָרֵא fearing, always יִרְאֵי; cf. also רִגְעֵי ψ 3520 from רָגֵעַ.—With ă retained in the initial syllable cf. אַחֵר alius (with a virtual sharpening of the ח).—From ע״וּ stems come forms like מֵת dead person, גֵּר resident stranger, עֵד witness, with unchangeable Qameṣ; hence מֵתִים, מֵתֵי, &c.
93kk Kindred in character are the formations from the ground-form qăṭŭl. This ground-form is regularly lengthened to qāṭōl, e.g. עָנֹל round, עָמֹק deep, אָדֹם red; but before formative additions the short ŭ returns, protected by the sharpening of the following consonant (see ee above), as עֲגֻלִּים, &c. (but in stems with a third guttural or ר, גְּבֹהָה, שְׁחֹרִים). The form עָגוֹל, 1 K 1019, is abnormal; likewise עֲמוּקָה Pr 2327, Jablonski (ed. Mant. עֲמֻקָה, Baer and Ginsburg עֲמֻקָּה).
93ll 3. On Paradigm f: ground-form qăṭăl from ל״ה stems. As in verbs ל״ה §75h, the general rule is that before the terminations of the plur. and dual and before suffixes beginning with a vowel, the third radical is usually elided altogether. But besides שָׂדֶה the form שָׂדַי, with the final Yôdh retained, is also found in poetry (cf. also the singulars with suffixes, like מִשְׁתֵּיהֶם, in ss); in the same way final ו is retained in עֲנָוִים the poor, constr. עַנְוֵי. The plur. of שָׂדֶה is שָׂדוֹת, constr. שְׂדוֹת (also שְׂדֵי, unless this is a sing., contracted from שְׂדַי; so Barth, ZDMG. xlii, p. 351). The qĭṭăl-form (see §84ai) רֵעֶה 2 S 1537, 1616, 1 K 45 is remarkable as a constr. st. (the reading רֵעֵה of Opitius and others is opposed to the express statement of the Masora). To the category of these forms also belongs without doubt פָּנִים face (only in plur.), פְנֵי, פָּנַי, פְּנֵיכֶם, &c.
93mm In a few formations of this kind the vowel of the second syllable appears to have been already lost in the absol. st. sing.; so according to the ordinary view, in יָד hand, constr. יַד, with suff. יָדוֹ, but יֶדְכֶם; plur. יָדוֹת, constr. יְדוֹת, dual יָדַ֫יִם, יְדֵי, with suff. יָדַי, יְדֵיכֶם, &c., and in דָּם blood, constr. דַּם, with suff. דָּמִי, but דִּמְכֶם (ă attenuated to ĭ), plur. דָּמִים, דְּמֵי. But perhaps both these nouns are to be regarded as primitive (§81), and as original monosyllabic formations.
93nn 3. Paradigm III comprises forms with an unchangeable vowel in the first syllable, whilst the vowel of the second syllable has been lengthened from an original short vowel, and is therefore changeable. The special cases are to be distinguished in which the original short vowel is lengthened both in and before the tone, but in an open syllable becomes Šewâ (Paradigm a, but cf. also examples like אֽוֹפַנִּים wheels, for אֽוֹפָנִים, and אֻֽלַמִּים porches), secondly, the cases in which the vowel becomes Šewâ even before the tone (Paradigm b), and finally, those in which the termination of ל״ה formations is entirely lost (Paradigm c).
93oo Rem. 1. On the model of עוֹלָם (which, moreover, is obscured from ʿâlăm), the following forms also are inflected: מִקְמָל (§85h), in some cases with virtual sharpening of the third radical (see §20a), as מִבְטַחוֹ Jer 177, ψ 405, Jb 814, &c.; ל״א nouns of this form maintain the Qameṣ in the constr. st. plur., e.g. מִקְרָאֵי from מִקְרָא; on the other hand, in the plur. of the participles Niph. (§85n) of verbs ל״אֹ (which likewise belong to this class), are found not only regular forms like נִקְרָאִים but also נֶחְבְּאִים Jos 1017, נִטְמְאִים Ez 2030 f., and so always נִבְּאִים (except Ez 132 הַנִּבָּאִ֑ים) and נִמְצְאִים 1 S 1315, 2 K 1414, &c. (except Ezr 825 הַנִּמְצָאִֽים in pause).
93oo Moreover, the other participles in ā also follow the analogy of עוֹלָם as regards the final syllable (מְקֻטָּל, מָקְטָל; cf., however, הַמּוּשַׁב Gn 4312 in close connexion; see the analogous cases in §65d); also שֻׁלְחָן table (§85u; plur. שֻׁלְחָנוֹת, constr. שֻׁלְחֲנוֹת), קָרְבָּן, constr. קָרְבַּן, hence in plur. constr. with suff. קָרְבְּנֵיהֶם Lv 738; עַקְרָב (§85w), plur. עַקְרַבִּים (with sharpening of the final consonant for עַקְרָבִים, cf. also עֵירֹם naked, plur. עֵֽירֻמִּים Gn 37 [but in 225 עֲרוּמִּים, according to §9o an orthographic licence for עֲרֻמִּים from עָרֹם, מַֽעֲרֻמִּים nakedness, 2 Ch 2815; קַרְדֹּם, קַרְדֻּמּוֹ; מַֽעֲמַקֵּי Is 5110; נִכְבַּדֵּי Is 238 f.; מִשְׂגַּבִּי ψ 183; even with attenuation of the ă to ĭ, מֽוֹרִגִּים threshing instruments, 2 S 2422, 1 Ch 2123, from מוֹרָג), מַתָּן (§85g), מָגֵן (§85i), מָעֹז (§85k), inasmuch as they retain the ā of the first syllable, contrary to rule, even when not pretonic, e.g. מָֽגִנִּי, מָֽעֻזִּי; מוֹשָׁב (§85g); תּוֹשָׁב (§85p), constr. st. plur. תּֽשָׁבֵי 1 K 171; also isolated forms according to §84at, and §84bb, c, k, m, n, o. Cf. finally, צַוָּאר neck (from ṣăwʾăr), constr. st. צַוַּאר Jer 2810 ff., constr. st. plur. צַוְּארֵי Gn 4514, &c.
93qq 2. (Paradigm b; cf. §84as.) Instead of the original ĭ in such forms as אֹֽיִבְכֶם (cf. 2 K 2229), the second syllable more frequently has ĕ, e.g. יֽׄצֶרְךָ thy creator; with a closing guttural (according to §91d; but cf. also אֹבַד Dt 3228) forms are found sometimes like שֹׁלֵֽחֲךָ, sometimes like בֹּרַֽאֲךָ; constr. st. without suff. נֹטַ֫ע ψ 949 (according to §65d); with a middle guttural גּֽׄאַלְךָ Is 4817; cf. 4314.—The same analogy also is followed in the flexion of the other participles which have ē in the final syllable (מְקַטֵּל, מִתְקַטֵּל, &c.), see further, in §84bd, גִּבֵּן, &c. (but with exceptions, as שִׁלֵּשִׁים, רִבֵּעִים), and ibid. l, p; §85i, k (מִזְבֵּחַ altar, constr. st. מִזְבַּח, plur. מִזְבְּחוֹת), and ibid. q , but here also there are exceptions like מַקְהֵלִים ψ 2612, מוֹסֵרוֹת Jer 55, רִבֵּעִים, שִׁלֵּשִׁים Ex 205, שֹׁמֵמוֹת Is 498, שֹׁמֵמִים La 116 (cf. König, ii. 109).
93rr 3. (Paradigm c: part. Qal of verbs ל״ה, differing from Paradigm II, f in the unchangeableness of the vowel of the first syllable.) In Ez 1715 ē in the absol. st. is abnormal, and Seghôl in the constr. st. in 2 S 2411 (so Opitius, Ginsburg; but Baer חֹזֵה), Ec 215 (according to Baer, but not the Mantua ed.; מִקְרֶה Ec 319 is in the absol. st.). To this class belong, as regards their formation, the ל״ה-forms mentioned in §84ar, §85g (with suff., e.g. הַמַּֽעַלְךָ Dt 201, which brought thee up), and h.
93ss In a few instances, before a suffix beginning with a consonant, the original ăy of the termination has been contracted to ê, and thus there arise forms which have apparently plural suffixes; as מִשְׁתֵּיהֶם Is 512, Dn 110.16; מַרְאֵיהֶם their appearance, Dn 115, Gn 4121, cf. Na 25; נֽוֹטֵיהֶם who stretched them forth, Is 425; defectively אֹֽפֵהֶם Ho 75 (cf. נְוֵהֶם Ez 3414); on the other hand, the examples in Is 1411, Gn 4717, which were formerly classed with the above, are really plurals. But מַֽחֲנֶ֫יךָ thy camp, Dt 2315 (מַֽחֲנֶ֫ךָ occurs just before), מִקְנֶ֫יךָ thy cattle, Is 3023 (probably also שָׂדֶ֫יךָ 1 K 226), מַרְאַ֫יִךְ Ct 214, and מַרְאָיו the sight of him, Jb 411 (with the י here retained orthographically), מַֽעֲלָיו Ez 4031, &c., are still to be explained as singulars.—On a few other examples which may perhaps be thus explained, see §124k. Before the plural ending the original termination ay reappears in מְמֻֽחָיִם Is 256 (part. Pu. from מָחָה).
93tt 4. Paradigm IV comprises the forms with a changeable vowel (a, b), or a vowel which has already become Šewâ (c), in the first syllable, and an unchangeable vowel in the second. With Paradigm c (which, however, for the most part consists merely of forms based on analogy, without biblical parallels) are also connected all the forms which have unchangeable vowels in both syllables, and therefore (like כְּתָב) cannot undergo any vowel changes.
93uu Rem. 1. Analogous to פָּקִיד (ground-form păqîd) are §84ak, גָּדוֹל, &c. (with ô, not changeable ô for ŭ); in substantives like שָׁלוֹם, this ̂ is demonstrably obscured from â (Arab. sălâm); ibid. k, m, אָסוּר, אָסִיר, &c.; §85u, זִכָּרוֹן, constr. זִכְרוֹן; חִזָּיוֹן, constr. חֶזְיוֹן; כִּלָּיוֹן, constr. כִּלְיוֹן (cf., however, the forms in the constr. st. עִצְּבוֹן, קִנְּמוֹן, and with the plural suffix עִזְּבוֹנַ֫יִךְ Ez 2712 ff.); §85w, חַלָּמִישׁ, constr. חַלְמִישׁ; §85l, מָקוֹם, &c.
93uu 2. עָנִי (ground-form ʿănîy, stem עָנָה) represents forms in which a final Yôdh has been resolved into î; before formative additions the original Yôdh under the protection of a Dageš forte again becomes audible as a firm consonant, whilst the (originally short) vowel of the first Syllable becomes Šewâ; cf. §84al, נָקִי, plur. נְקִיִּים, and §87a.
93ww 3. כְּתָב with unchangeable â in the second syllable, whilst the Šewâ is weakened from a short vowel (Arab. kĭtâb); constr. st. כְּתָֽב־ Est 48 (readings like כְּתַב 2 Ch 354 are incorrect, although יְקַר Est 14 and כְּתַב־ 48 are supported by fairly good authority; however, these qeṭâl-forms in Hebrew are probably all loan-words from the Aramaic). The only plural form found in the O.T. is עֲבָֽדֵיהֶם their deeds, Ec 91. In a narrower sense the forms enumerated in §84an–p belong to this class; in a wider sense all those which have unchangeable vowels throughout, thus §84au, §84be (קַטָּל, cf., however, the anomalous forms mentioned there), ibid. f–i, m (No. 34 f.), n (No. 39), p (No. 44), also partly §85b–w (especially l and r).
93xx In opposition to the anomalous shortening of the form קַטָּל (see above), cases are also found where pretonic vowels are retained even in the antepenultima (with the secondary tone); cf. above, ii and pp, also of the form קָטִיל (properly qăṭîl) the examples סָֽרִיסִים, פָּֽרִיצִים, שָֽׁלִישִׁים, whilst the constr. st. sing. according to the rule, changes the ā into Sewâ (סְרִיס, פְּרִיץ). (These are not to be confounded with forms like עָרִיץ tyrant, which is for עַרִּיץ, and consequently has an unchangeable Qameṣ.) Of the form קָטוּל (qăṭûl) in this class are שָׁבוּעַ week, plur. שָֽׁבֻעִים and שָֽׁבֻעוֹת, constr. שְׁבֻעוֹת, but with Metheg of the secondary tone in the fifth syllable from the end, שָֽׁבֻעֹֽתֵיבֶם.—On מָעוֹז, מָֽעֻזִּי, &c., cf. §85k.
- A sort of detailed commentary on the following scheme of Hebrew declensions is supplied by E. König in his Hist.-krit. Lehrgeb. der hebr. Spr., it. 1, p. 1 ff.
- According to P. Haupt ‘The book of Nahum’ in the Journ. of bibl. Lit., 1907, p. 29, the e in סֵ֫פֶר and the o in קֹ֫דֶשׁ are not long but accented, and hence to be pronounced σέφρ, ὄζν (אֹ֫זֶן), a theory unknown at any rate to the Jewish grammarians.
- According to M. Lambert, REJ. 1896, p. 21, a tends to remain with labials; so in 14 cases out of 22 masculines, and in 3 out of 6 feminines.
- Probably only a theory of one particular school and not generally accepted, or at any rate not consistently carried out; cf. König, Lehrgeb., ii. 22.
- The proposal of Haupt (SBOT. ‘Proverbs’, p. 34, l. 44 ff.) to read בֵּאר, זֵאב, &c., does not seem to be warranted. The case here is quite different from that in Pr 122 where the Masora requires תְּאֵֽהֲבוּ, no doubt on the analogy of בְּאֵר, &c., for תֵּֽאהֲבוּ, which was probably intended, see §63m.
- Nöldeke, Beiträge, p. 58: the direct or indirect retention of this ו is hardly a feature of early Hebrew. The true Hebrew forms from קָצֶה would be קְצֵה, קָצוֹת, קְצוֹת, the aramaizing forms קָצָה, קְצָת, קְצָווֹת.
- חֵי only in Dn 127 as constr. st., since in the asseverative formulae (cf. §149) חֵי פַרְעֹה, חֵי נַפְשְׁךָ (otherwise only in 2 S 1521, after חַי יהוה, and Amos 814), חֵי is a contracted form of the absol. st. (prop. living is Pharaoh! &c.). It is evidently only a rabbinical refinement which makes the pronunciation חַי distinctive of an oath by God (or of God by himself), as in the regular formulae חַי אָ֫נִי (חַי אָֽנֹכִי Dt 3240) and חַי יְהֹוָה (=חַי אֲדֹנָי).
- מְקַדְשֵׁיהֶם Ez 724 for מִקְדְּשׁ׳ (from מִקְדָּשׁ) is wholly irregular; perhaps, however, the part. Piʿēl is intended, without Dageš in the ד (according to §20m).
- Brockelmann, Grundriss, p. 659, observes that except in 2 Ch 511, 3517 הַנִּמְצְאִים is always followed by a preposition governing a word, so that the punctuators perhaps intended to indicate a sort of constr. st.