Gesenius' Hebrew Grammar/96. Nouns of Peculiar Formation

§96. Nouns of Peculiar Formation.
In the following Paradigms,[1] pp. 282 to 284, a number of frequently used nouns are arranged, whose flexion presents more or less striking peculiarities. These peculiarities, however, are almost always subordinate to the usual phonetic laws, and the usual designation of the nouns as irregular is, therefore, not justified, when once the ground-forms are properly recognized on which the present forms are based.
Sing. absolute אָב אָח אָחוֹת אִישׁ אִשָּׁה
(father) (brother) (sister) (man) (woman)
" construct אֲבִי אֲחִי אֲחוֹת אִישׁ אֵ֫שֶׁת
" with suff. of 1 sing. אָבִי אָחִי אֲחֹתִי אִישִׁי אִשְׁתִּי
" 2 masc. אָבִ֫יךָ אָחִ֫יךָ אֲחֽוֹתְךָ אִשְׁתְּךָ
" 2 fem. אָבִיךְ אָחִיךְ אֲחוֹתֵךְ אִישֵׁךְ
" 3 masc. אָבִיו (אָבִ֫יהוּ) אָחִיו (אָחִ֫יהוּ) אֲחֹתוֹ אִישׁוֹ אִשְׁתּוֹ
" 3 fem. אָבִ֫יהָ אָחִ֫יהָ אֲחֹתָהּ אִישָׁהּ
" 1 Pl. אָבִ֫ינוּ אָחִ֫ינוּ אֲחֹתֵ֫נוּ
" 2 masc. אֲבִיכֶם אֲחִיכֶם [אֲחֽוֹתְכֶם]
" 2 fem. אֲבִיכֶן
" 3 masc. אֲבִיהֶם אֲחִיהֶם אֲחֹתָם
" 3 fem. אֲבִיהֶן
Plur. absolute. אָבוֹת אַחִים אֲנָשִׁים נָשִׁים
" construct אֲבוֹת אֲחֵי אַנְשֵׁי נְשֵׁי
" with suff. of 1 sing. אֲבֹתַי אַחַי, pause אֶחָי אַחְיוֹתַי אֲנָשַׁי נָשַׁי
" 2 masc. אֲבֹתֶ֫יךָ אַחֶ֫יךָ אֲנָשֶׁ֫יךָ נָשֶׁ֫יךָ
" 2 fem. אַחַ֫יִךְ אֲחוֹתַ֫יִךְ
" 3 masc. אֲבֹתָיו אֶחָיו אַחְיֹתָיו אֲנָשָׁיו נָשָׁיו
" 3 fem. אַחֶ֫יהָ אֲנָשֶׁ֫יהָ
" 1 Pl. אֲבֹתֵ֫ינוּ אַחֵ֫ינוּ אֲנָשֵׁ֫ינוּ נָשֵׁ֫ינוּ
" 2 masc. אֲבֹֽתֵיכֶם אֲחֵיכֶם אֲחֽוֹתֵיכֶם נְשֵׁיכֶם
" 3 masc. אֲבֹתָם (אֲבֹֽתֵיהֶם) אֲחֵיהֶם אַחְיֹֽתֵיהֶם אַנְשֵׁיהֶם נְשֵׁיהֶם
" 3 fem. אַנְשֵׁיהֶן


אָב father; the constr. אֲבִי, like אֲחִי and בְּנִי (which occurs once), belongs to the connective forms discussed in §90k, which serve as the model for the Ḥireq compaginis. However, אַב also occurs in compound proper names, e.g. אַבְשָׁלוֹם, beside אֲבִֽישָׁלוֹם, &c.; also Gn 174 f. אַב־הֲמוֹן for the purpose of explaining the name אב[ר]הם. On the plur. אָבוֹת see §87p.

אָח brother. The plur. absol. אַחִים has Dageš forte implicitum (§22c); אֶחָיו stands for אַחָיו according to the phonetic law stated in §27q, and so also אֶחָֽי in pause for אַחָֽי. The sharpening of the ח merely serves to keep the preceding Pathaḥ short, as in גְּמַלִּים, &c. (§93ee).

אֶחָד one (for אַחָד, likewise with Dageš forte implicitum, §22c, cf. §27q), constr. and otherwise in close connexion, אַחַד, Gn 4822, 2 S 1722, Is 2712,
אָמָה בַּ֫יִת בֵּן בַּת יוֹם כְּלִי
(handmaid) (house) (son) (daughter) (day) (vessel)
בֵּית בֶּן־ בַּת יוֹם כְּלִי
אֲמָתִי בֵּיתִי בְּנִי בִּתִּי
אֲמָֽתְךָ בֵּֽיתְךָ בִּנְךָ, pause בִּנֶ֫ךָ בִּתְּךָ, pause בִּתֶּ֫ךָ כֶּלְיְךָ
בֵּיתֵךְ בְּנֵךְ
אֲמָתוֹ בֵּיתוֹ בְּנוֹ בִּתּוֹ יוֹמוֹ
אֲמָתָהּ בֵּיתָהּ בְּנָהּ בִּתָּהּ
בֵּֽיתְכֶם בִּתְּכֶם
בֵּיתָם יוֹמָם
אֲמָהוֹת בָּֽתִּים בָּנִים בָּנוֹת יָמִים כֵּלִים
אַמְהוֹת בָּֽתֵּי בְּנֵי בְּנוֹת יְמֵי כְּלֵי
אַמְהֹתַי בָּנַי בְּנֹתַי יָמַי כֵּלַי
בָּֽתֶּ֫יךָ בָּנֶ֫יךָ בְּנֹתֶ֫יךָ יָמֶ֫יךָ כֵּלֶ֫יךָ
בָּֽתַּ֫יִךְ בָּנַ֫יִךְ בְּנֹתַ֫יִךְ יָמַ֫יִךְ
אַמְהֹתָיו בָּנָיו בְּנֹתָיו יָמָיו כֵּלָיו
אַמְהֹתֶ֫יהָ בָּנֶ֫יהָ בְּנֹתֶ֫יהָ יָמֶ֫יהָ כֵּלֶ֫יהָ
בָּֽתֵּ֫ינוּ בָּנֵ֫ינוּ בְּנֹתֵ֫ינוּ יָמֵ֫ינוּ כֵּלֵ֫ינוּ
אַמְהֹֽתֵיכֶם בָּֽתֵּיכֶם בְּנֵיכֶם בְּנֹֽתֵיכֶם יְמֵיכֶם כְּלֵיכֶם
בָּֽתֵּיהֶם בְּנֵיהֶם בְּנֹֽתֵיהֶם יְמֵיהֶם כְּלֵיהֶם
אַמְהֹֽתֵיהֶן בָּֽתֵּיהֶן בְּנֵיהֶן

Zc 117; and especially before מִן (מֵ‍) Gn 322, Ex 3014, Nu 1615, Ju 175, 1 S 93, Ez 1810; fem. אַחַת una (for אַחַדְתְּ, according to §19d), in pause אֶחָֽת. Once חַד masc. (by aphaeresis, §19h), Ez 3330, as in Aramaic; plur. אֲחָדִים some, but also iidem.

אָחוֹת sister, from ʾăḥăwăt or ʾăḥăyăt, with elision of the ו or י, and with the â, which has arisen from ăă, obscured to ô.[2] In Nu 67 אַחֹתוֹ stands for אֲחֹתוֹ (with virtual sharpening of the ח). The plur. absol. (אֲחָיוֹת) does not happen
[מַי] עִיר פֶּה רֹאשׁ שֵׁם [שָׁמַי]
(water) (city) (mouth) (head) (name) (heaven)
עִיר פִּי רֹאשׁ שֵׁם, שֶׁם־
עִירִי פִּי רֹאשִׁי שְׁמִי
עִֽירְךָ פִּ֫יךָ רֹֽאשְׁךָ שִׁמְךָ, pause שְׁמֶ֫ךָ
רֹאשֵׁךְ שְׁמֵךְ
עִירוֹ פִּיו, פִּ֫יהוּ רֹאשׁוֹ שְׁמוֹ
עִירָהּ פִּ֫יהָ רֹאשָׁהּ שְׁמָהּ
פִּ֫ינוּ רֹאשֵׁ֫נוּ שְׁמֵ֫נוּ
פִּיכֶם רֹֽאשְׁכֶם שִׁמְכֶם
עִירָם פִּיהֶם רֹאשָׁם שְׁמָם
פִּיהֶן רֹאשָׁן
מַ֫יִם עָרִים פִּיּוֹת רָאשִׁים שֵׁמוֹת שָׁמַ֫יִם
מֵי, מֵימֵי עָרֵי רָאשֵׁי שְׁמוֹת שְׁמֵי
מֵימַי עָרַי
מֵימֶ֫יךָ עָרֶ֫יךָ שָׁמֶ֫יךָ
מֵימָיו עָרָיו רָאשָׁיו שָׁמָיו
מֵימֶ֫יהָ עָרֶ֫יהָ רָאשֶׁ֫יהָ
מֵימֵ֫ינוּ עָרֵ֫ינוּ רָאשֵׁ֫ינוּ
עָֽרֵיכֶם רָֽאשֵׁיכֶם שְׁמֵיכֶם
מֵֽימֵיהֶם עָֽרֵיהֶם רָֽאשֵׁיהֶם שְׁמוֹתָם
רָֽאשֵׁיהֶן שְׁמוֹתָן

to occur. In Ez 1652 אַֽחֲיוֹתֵךְ occurs (for אַחְיֹתַ֫יִךְ). In the forms אֲחוֹתַי Jos 213 Keth., אֲחוֹתַ֫יִךְ Ez 1651.55.61 (to be read also in verse 45 for אֲחוֹתֵךְ, which has been erroneously assimilated to the singular occurring in vv. 48, 49, 56), and אֲחֽוֹתֵיכֶם Ho 23 (for which, however, read אֲחֽוֹתְכֶם), the third radical has been entirely lost.

אִישׁ man, according to the common opinion either incorrectly lengthened for אֵשׁ (from ʾišš, with assimilation of the Nûn of the ground-form ʾinš, which again has been attenuated from ʾanš from the stem אָנַשׁ), or softened directly from ʾinš. It is, however, probable that a separate stem (אישׁ to be strong?) is to be assumed for the singular[3]; consequently the stem אָנַשׁ to be sociable, would be connected only with the plur. אֲנָשִׁים (אִישִׁים is found only in Is 533, ψ 1414, Pr 84).

אָמָה slave, handmaid; with the plur. אֲמָהוֹת, with consonantal ה, cf. in Aram. אֲבָהָן fathers, and similarly in Phoen. דלהת from דלת, also Arab. ʾabahât (fathers), ʾummahât (mothers), with an artificial expansion into a triliteral stem.

אִשָּׂה woman, probably for אִנְשָׁה; from אָנַשׁ i.e. not (as Aram. אִתְּתָא shows) אָנַשׁ to be sociable (see above, on אִישׁ) but אָנַשׁ to be weak (Arab. ʾănŭṯă). So De Lagarde, Uebersicht, p. 68; König, Lehrgeb., ii. 159 f. The form אֵ֫שֶׁת (for ʾišt, with ת fem., from ʾišš, after rejection of the doubling and lengthening of the ĭ to ē) occurs in Dt 2111, 1 S 287, ψ 589, even in absol. st. [cf., however, below, §130. 4, 5].—In ψ 1283 אֶשְׁתְּךָ is found for אִשְׁתְּךָ. Instead of the plur. נָשִׁים, we find in Ez 2344 אִשֹּׁת.[4]

בַּ֫יִת house, locative בַּ֫יְתָה, הַבַּ֫יְתָה, in pause בָּ֫יְתָה, הַבָּ֫יְתָה, constr. בֵּ֫יתָה, plur. בָּֽתִּים (but in Dt 611, 1 Ch 2811 בָּתִּים without Metheg), pronounced bâttîm. The explanation of the Dageš in the ת is still a matter of dispute. The Syriac bâttîn, however, shows that the Dageš is original, and belongs to the character of the form.[5] According to Wright, Comparative Grammar, p. 88, בָּֽתִּים is simply contracted from bai-tîm (as אָן from אַ֫יִן, עֵינָם from עֵינָיִם, &c.), and the Dageš, therefore, is lene; König, Lehrgeb., ii. 56, proposes the name Dageš forte orthoconsonanticum; on the other hand Rahlfs, ThLZ. 1896, col. 587, suggests that the י is assimilated to the ת, while Philippi, ZDMG. xlix, p. 206, assumes for the plural a stem distinct from that of the singular. A definite solution is at present impossible. The incorrectness of the formerly common pronunciation bottîm is sufficiently shown by the Babylonian punctuation (see §8g, note 3), which leaves no doubt as to the â.

בֵּן son (Gn 3019 בֵּֽן־שִׁשִּׁי) constr. usually בֶּן־ (also with a conjunctive accent as an equivalent for Maqqeph, Gn 1717, Is 82, &c., 1 Ch 921; even with smaller disjunctives, especially in the combination מִבֶּן, Ex 3014, Lv 273, &c. [מִבֶּן־ only after וְאִם and before חֹ֫דֶשׁ, also in Is 5112; see Strack on Ex 3014]), rarely בִּן־ (Dt 252, Jon 410 twice, Pr 301, and so always in the combination בִּן־נוּן, and in the proper names בִּנְיָמִין [but בֶּן־יִמִינִי Benjamite] and בִּן־יָקֶה Pr 301), once בְּנִי (cf. §90l) Gn 4911, and בְּנוֹ (§90o) Nu 2318, 243.15.—In Gn 4922 בֵּן, for which בֶּן־ ought to be read, is intended by the Masora for the absol. st., not the constr. בַּת daughter (from bant, and this again, according to the law stated in §69c, for bint, fem. of בֵּן), with suff. בִּתִּי for בִּנְתִּי. Plur. בָּנוֹת, from the sing. בָּנָה, comp. בָּנִים sons.

חָם husband’s father, only with suff. חָמִיךְ, חָמִ֫יהָ; and חָמוֹת husband’s mother, only with suff. חֲמוֹתֵךְ, חֲמוֹתָהּ. Cf. אָב, אָח, and especially אָחוֹת.

יוֹם day (Arab. yaum),[6] dual יוֹמַ֫יִם; the plur. יָמִים is probably from a different sing. (יָם yām), constr. יְמֵי and (poetically) יְמוֹת, Dt 327, ψ 9015.

כְּלִי vessel, in pause כֶּ֫לִי (with suff. כֶּלְיְךָ Dt 2325) from כָּלָה to contain, plur. כֵּלִים (as if from כֵּל, כֵּ֫לֶה; according to König, ii. 63, shortened from kilyîm).

מַ֫יִם water; on the plur. cf. §88d.

עִיר city. The plur. עָרִים is scarcely syncopated from עֲיָרִים, as it is pointed in Ju 104 (no doubt erroneously, in imitation of the preceding עֲיָרִים ass colts), but from a kindred sing. עָר, which still occurs in proper names.

פֶּה mouth, constr, st. פִּי (for original פֵּי = פֵּה?). Its origin is still disputed. According to Gesenius and König (ii. 103), פֶּה stands for פֵּאֶה (ground-form piʿay) from פָּאָה to breathe, to blow; according to Olshausen, for פַּי, from a stem פָּיָה or פָּוָה. But parallel with the Hebrew פֶּה are Assyr. , Arab. fû, fam, famm, fumm, bibl. Aram. פֻּם, פֻּמָּא, Syr. pûm, pûmā, so that Barth, ZDMG. xli, p. 634, assumes two forms of development from the same stem (פמו), viz. fm and fw. פִּי my mouth, from pi-y; for פִּיהֶם we find in ψ 1710, 587, 5913 פִּ֫ימוֹ. The supposed plur. פִּים 1 S 1321 is generally explained as a contraction from פִּיִּים, but the text is altogether corrupt. The plur. פִּיּוֹת, for the edges of a sword, occurs in Pr 54; reduplicated פִּֽיפִיוֹת Is 4115, ψ 1496.

רֹאשׁ head (obscured from רָאשׁ=răʾš); plur. רָאשִׁים (for רְאָשִׁים, §23c); רֹאשָׁיו only in Is 152.

שֶׂה a head of small cattle (sheep or goat), constr. st. שֵׂה, with suff. שְׂיֵ֫הוּ 1 S 1434 and שֵׂיוֹ Dt 221, according to König, ii. 131, from a ground-form siʾay, but according to De Lagarde, Uebersicht, 81 f., from a stem ושי (שֶׂה=say=wĭsay).

שֵׁם name, constr. generally שֵׁם (only six times שֶׁם־); cf. בֵּן.

שָׁמַ֫יִם heaven (§88d).

  1. The only omissions from these Paradigms are אֶחָד, חָם, and חָמוֹת (on which see the remarks), and all forms which are not found in the O.T.
  2. This explanation of אָחוֹת (and חָמוֹת q. v.) still seems to us more probable than the assumption that the fem. ending ăth is lengthened to compensate for the loss of the 3rd radical (so Wellhausen, Skizzen, vi. 258), or that the form is derived from ʾaḥâ, the old-semitic constr. st. of the accusative, with ת feminine (so Barth, ZDMG. 1899, p. 598).
  3. So already Gesenius in his Thes. Linguae Hebr., i. 83 f., and recently again Friedr. Delitzsch, Prolegg., p. 160 ff., Praetorius in Kuhn’s Orient. L.-B., 1884, p. 196; König, Lehrgeb., ii. 38; while Nöldeke (ZDMG. 1886, p. 739 f.), against Delitzsch, would connect both אִישׁ and נָשִׁים with the stem אנשׁ.
  4. Friedr. Delitzsch (in his Babylonian glosses to Baer’s text of Ezekiel, p. xi) on Ez 2344 remarks that in Assyro-Babylonian the plur. of aššatu (woman) is aššâti, corresponding, therefore, to אִשּׁוֹת, not to the ordinary plur. נָשִׁים. The a of נָשִׁים (instead of i as in Arab. or e as in Syr.) is to be explained with Barth (Orient. Studien zu Ehren Th. Nöldekes, Giessen, 1906, p. 792) from the natural connexion of the ideas ‘men’ and ‘women’, נָשִׁים and אֲנָשִׁים.
  5. This disposes of the traditional view that the Dageš (after a firm Metheg, see §16f ζ) only serves to distinguish if from בָּתִים passing the night, ptcp. Qal of בּוּת, a stem which never occurs in the O.T. According to P. Haupt the stem is בא to go in, ת therefore being the feminine termination, as in bint daughter, and the original form baʾtu, bātu (entrance) is preserved in the plural bāttim where the tt is to be explained as due to the analogy of trisyllabic stems. In the singular bāt passed into bēt (?), and this was resolved into bait, as Yerūšālēm into Yerūšālayim.
  6. Cf. Nöldeke, Beitröge, p. 58, yaum, probably an extension of a biliteral word which has survived in יָמִים, יְמֵי. Barth, however, Orient. Studien, p. 791 (see above on אִשָּׂה), sees in יָמִים, יְמֵי, יְמוֹת new formations in Hebrew, caused by the naturally close connexion and association of these plurals with שְׁנִים, שְׁנֵי, שְׁנוֹת years, to which they became assimilated in form. The view that יוֹם is merely an incorrect obscuring of יָם, and therefore distinct from the Arab. yaum, is contradicted by the invariable spelling יוֹם, &c., notwithstanding the spelling ובים (=וּבְיֹם ?) in the Siloam inscription, line 3 (cf. §7f), and מִיֹּמָ֑יִם Ho 62. Cf. also the note on §100g.