Modern Hebrew

CHAPTER 22 MODERN HEBREW PHILIP ZHAKEVICH AND BENJAMIN KANTORMODERN HEBREW Philip Zhakevich and Benjamin Kantor 1 INTRODUCTION Modern Hebrew, spoken by about six million people primarily in the state of Israel (Map 22.1), belongs to the Canaanite subgroup within the branch of Northwest Semitic languages. Of the Canaanite languages, Modern Hebrew is the only language spoken today. With respect to the historical development of the language, Modern Hebrew fol- lows Biblical Hebrew (12th–2nd c. bce), Post-Biblical Hebrew (2nd c. bce–2nd c. ce), Rabbinic Hebrew (2nd–5th c. ce) and Medieval Hebrew (6th–15th c. ce) as well as the Hebrew of the early modern period (16th–18th c. ce). Hebrew ceased to be a spoken language around the 3rd century ce, but it remained a liturgical language until the early modern period. In the 18th century, during the Jewish Enlightenment in Europe, Modern Hebrew emerged as a medium for writing secular academic works as well as non-reli- gious literature, both prose and poetry. It was in the 19th century, however, that Hebrew was revived as a spoken language by Eastern European immigrants in Palestine. Today, Modern Hebrew is used for day-to-day communication in all registers and environments (e.g., media, education, medicine). MAP 22.1  THE MODERN HEBREW SPEECH AREA 572  Philip Zhakevich and Benjamin Kantor Some researchers cast doubt on the Semitic character of Modern Hebrew on the basis of non-Semitic elements present in the language (Zuckerman 2008, Wexler 1990). Foreign influence on Modern Hebrew is clearly exemplified in loan words, calque translations, semantic borrowing, borrowed lexical suffixes and borrowed syntax. Indeed, Hebrew has a unique history in that it was revived for speaking purposes after almost two millennia of dormancy. This required its initial speakers to fill in the gaps of the language with lexical items from foreign languages. Nevertheless, in terms of its morphological, phonological and syntactic features, Modern Hebrew retains a strong and discernible Semitic character (Goldenberg 1996, Shlesinger 2013: 375–81). Because Hebrew served as a literary language from the 3rd until the 19th century ce, it experienced a much slower rate of change in its morphology than many other languages. For this reason, the morphology of Modern Hebrew closely resembles the morphology of earlier stages of the language. This, of course, means that earlier texts of Hebrew (e.g., Bible, Mishnah, medieval commentaries) remain accessible to an educated speaker of Modern Hebrew. While ancient Hebrew had regional dialects, Modern Hebrew has been relatively homogenous. From the 1890s to the 1920s, however, a unique dialect created by Yitzḥak Epstein existed in the Galilee. Called the ‘Galilean Dialect’, it was based on a tradi- tional pronunciation of Hebrew that was customary among Jews from Arabic-speaking countries. The dialect included the pronunciation of guttural and velarized (“emphatic”) consonants as well as geminated consonants (Bar-Adon 1975). Today, however, linguists only speak of sociolects, ethnolects, and religiolects (Schwarzwald 2013e, Bar-Asher 2010). Modern Hebrew superficially appears to be a continuation of earlier stages of Hebrew. This resemblance, however, masks the fact that the language has undergone sig- nificant changes that distinguish it from its earlier attestations. Phonologically, Modern Hebrew is a combination of two pronunciation traditions, the Sephardic and Ashkenazic; the former is the source of the vocalic pronunciation of the language, whereas the latter is the source of the consonantal pronunciation. Modern Hebrew, then, composed of these two traditions, does not reflect the phonology of ancient Hebrew but the fusion of two particular descendant phonologies of the ancient pronunciation. Morphologically, Modern Hebrew generally resembles Biblical Hebrew in verbal and nominal inflections. However, its TAM reflects the Rabbinic system rather than bibli- cal. Rabbinic forms, both nominal and verbal, served to fill gaps within the lexicon; for instance, verbal nouns (e.g., ktiva ‘writing’), habitual past tense forms (i.e., haja holeχ be.pst.3msg go.prs.msg ‘he used to go’), and certain verbal patterns (i.e., ʃiχtev rewrite. pst.3msg ‘he rewrote’) can be traced to Rabbinic Hebrew. Syntactically, Modern Hebrew is similar to Rabbinic Hebrew in its sentence structure, but it resembles Biblical Hebrew in its phrase structure. The influence of European languages has also left its mark on the syntax of Modern Hebrew, especially with respect to its macrostructures such as compli- cated sentences and paragraphs (Reshef 2013a). 2 WRITING SYSTEM Modern Hebrew is written right to left in an alphabet that consists of 22 consonants, three of which may also be marked by an apostrophe to signify foreign sounds. There are essentially two different scripts used for writing Modern Hebrew today, a square script and a curisve script. The square script, whose roots are ancient, is used in printed text,  Modern Hebrew 573 TABLE 22.1  HEBREW ALPHABET square script ‫א‬ ‫ּב‬ ‫ב‬ ‫ג‬ ‫ד‬ ‫ה‬ ‫ו‬ ‫ז‬ ‫ח‬ ‫ט‬ ‫י‬ ‫ּכ‬ ‫)ך) כ‬ cursive script ‫א‬ ‫ּב‬ ‫ב‬ ‫ג‬ ‫ד‬ ‫ה‬ ‫ו‬ ‫ז‬ ‫ח‬ ‫ט‬ ‫י‬ ‫ּכ‬ ‫)ך( כ‬ ipa ʔ b v g d h v z χ t j k χ square script ‫ל‬ ‫)ם( מ‬ ‫)ן( נ‬ ‫ס‬ ‫ע‬ ‫ּפ‬ ‫)ף) פ‬ ‫)ץ( צ‬ ‫ק‬ ‫ר‬ ‫ׁש‬ ‫ׂש‬ ‫ת‬ cursive script ‫ל‬ ‫)ם( מ‬ ‫)ן( נ‬ ‫ס‬ ‫ע‬ ‫ּפ‬ ‫)ף) פ‬ ‫)ץ) צ‬ ‫ק‬ ‫ר‬ ‫ׁש‬ ‫ׂש‬ ‫ת‬ ipa l m n s ʔ (ʕ) p f ʦ k r ʃ s t square script '‫ג‬ '‫ז‬ '‫צ‬ cursive script '‫ג‬ '‫ז‬ '‫צ‬ ipa ʤ ʒ ʧ Note: Signs appearing in parentheses are final forms of the corresponding signs not in parentheses. whereas the cursive script, whose emergence dates to the Middle Ages, is utilized for text written by hand (Engel 2013: 492–4, 501). See Table 22.1 for the Hebrew alphabet. The Hebrew square script derives from ancient Aramaic script, which utilized 22 dis- tinct graphemes to represent the consonants of the language. In addition to representing consonants, Aramaic script also utilized certain graphemes to indicate certain vowels, namely, h (final a), w (u, o), j (i) and ʔ (a). These “vowel letters” are referred to as matres lectionis ‘mothers of reading’, and their use for writing vowels was adopted into Hebrew already in the biblical period and has continued into Modern Hebrew (Daniels 2013: 89–90). Besides the matres lectionis, Hebrew also has a diacritic system, niqqud ‘pointing’, to mark vowels, geminated consonants, as well as other aspects of pronunciation. This point- ing system consists of dots and lines that appear underneath, beside, within and above consonants. Established by the Tiberian Masoretes around the 8th century ce, niqqud is used today as a system that is superimposed upon the historical matres lectionis system. While Modern Hebrew has inherited the diacritics of Tiberian Hebrew, these diacrit- ics generally do not reflect the same sounds in Modern Hebrew as they did in Tiberian Hebrew. For instance, today, there are only five vowels: a, e, i, o, u (Bolozky 2013f: 985), whereas the Tiberian system specified seven distinct vowel qualities that could occur as long or short vowels (Khan 2013: 774). Also, the function of a diacritic called dagesh – a dot in the middle of a letter (e.g., ‫ – )ּג‬is different in Modern Hebrew when compared to Tiberian Hebrew. While one of its functions in Tiberian Hebrew was to mark gemination, it remains only a spelling convention in many cases in Modern Hebrew due to the lack of phonemic gemination in the language. Nevertheless, its presence in certain letters still serves to mark a plosive, rather than fricative, consonant when both the plosive and fric- ative are represented by the same basic letter: i.e., ‫ּב‬ = /b/ and ‫ב‬ = /v/, ‫ּפ‬ = /p/ and ‫פ‬ = /f/, ‫ּכ‬ = /k/ and ‫כ‬ = /χ/. Throughout its history, Hebrew writing conventions relating to the matres lectionis were applied inconsistently, which has resulted in fluid writing conventions. The intro- duction of the Tiberian system helped establish a correct way of reading the text, but also complicated spelling conventions. Official rules for spelling were only created in the 20th century (Barak 2013: 956–64) and resulted in two official writing orthographies: pointed 574  Philip Zhakevich and Benjamin Kantor and unpointed. For instance, in pointed orthography, the word ʦohorajim ‘noon’ is writ- ten as ‫( ָצה ֳַרי ִם‬consonantal text = ʦhrjm), whereas in unpointed form it is written as ‫צוהריים‬ (consonantal text = ʦwhrjjm). Pointed orthography appears in children’s books and poetry, while unpointed orthography appears in almost all other printed texts and public signs. Today, a word can be spelled in different ways and different publishers may display different spellings of the same word. 3 PHONOLOGY As mentioned earlier, the phonology of Modern Hebrew is a hybrid of the Ashkenazi and Sephardi pronunciation traditions. While the early modern Hebrew speech community was made up mostly of Sephardi Jews, later immigration waves of European Jews to Palestine resulted in a heavy Ashkenazi influence on pronunciation. The earliest wave of Ashkenazi migrants attempted to adopt the Sephardi pronunciation and distance them- selves from certain distinctive features of their own tradition (due to its association with the Diaspora), but much of their pronunciation remained. It was predominantly in the realm of vowels and word stress that Sephardi phonology was adopted, whereas the con- sonantal system remained largely Ashkenazi (Morag 1980: 86, Reshef 2013a: 399–400, Reshef 2013b). Modern Hebrew phonemes not original to either the Ashkenazi or Sep- hardi pronunciation traditions mainly occur in loan words or are the result of natural speech development. 3.1 Consonants The consonantal inventory of Modern Hebrew, as given in Table 22.2, is nearly identical to that of the Ashkenazi tradition with the exception that the reflex of the post-vocalic allophone of /t/ is realized as [t] instead of [s] as it is in Ashkenazi Hebrew (see (8) in Table 22.3) (Bolozky 1997, 2013c, Reshef 2013a: 399). (Parentheses enclose phonemes particular to loan words and angled brackets enclose phonemes present among a minority of speakers.) This consonantal system reflects a number of mergers from Biblical Hebrew (see Table 22.3; see also Chapter 21). In addition to these mergers, a few further points are noteworthy: Gemination: Historical gemination (i.e., consonantal length) has been simplified and is thus no longer phonemic: e.g., [diber] (< *dibber) ‘speak.pst.3msg’ (but see §3.4). TABLE 22.2  HEBREW CONSONANTS Bilabial Labio- Alveolar Palato- Palatal Velar Uvular Pharyngeal Glottal dental alveolar Stop pb td kg ʔ Affricate ʦ (ʧ) (ʤ) Fricative fv sz ʃ (ʒ) χ <ʁ> <ħ> <ʕ> h Nasal m n Lateral l Trill <r> <ʀ> Approximant (w) <ɹ> j ʁ  Modern Hebrew 575 TABLE 22.3  CONSONANTAL MERGERS (1): /w/, /v/ > /v/ (2): /ħ/, /χ/ > /χ/ (3): /q/ (or /kˀ/), /k/ > /k/ (4): /tˁ/ (or /tˀ/), /t/ > /t/ (5): /ʕ/, /ʔ/, (/h/) > /ʔ/ > Ø (6): /g/, /γ/ > /g/ (7): /d/, /ð/ > /d/ (8): /t/, /θ/ > /t/1 (9): /ɬ/, /s/ > /s/ (1) The approximant /w/ ‫ ו‬has merged with the post-vocalic allophone of /b/ ‫[( ּב‬v] ‫)ֿב‬.2 (2) The voiced pharyngeal fricative /ħ/ ‫ ח‬has merged with the post-vocalic allophone of /k/ ‫[( ּכ‬χ] ‫)ֿכ‬. (3–4)  The “emphatics” /q/ ‫( ק‬or /kˀ/) and /tˁ/ ‫( ט‬or /tˀ/) have merged with their non-emphatic counterparts /k/ ‫ ּכ‬and /t/ ‫ּת‬.3 (5)  For most speakers, the voiced pharyngeal /ʕ/ ‫ ע‬has merged with /ʔ/ ‫א‬, which itself has come to be realized as zero (Ø). The glottal fricative /h/ ‫ ה‬is also, though less frequently, realized as Ø. (6–8)  The post-vocalic allophones of /g/ ‫[( ּג‬γ] ‫)ֿג‬, /d/ ‫[( ּד‬ð] ‫ )ֿד‬and /t/ ‫[( ּת‬θ] ‫ )ֿת‬have merged with their plosive counterparts. (9)  Finally, the lateral fricative /ɬ/ ‫ ׂש‬and /s/ ‫ ס‬have merged (i.e., /ɬ/, /s/ > /s/), but this merger likely began during the biblical period. Spirantization: Historically, the stops /b/, /p/, /g/, /k/, /d/, /t/ each had a fricativized allophone in post-vocalic environments: /b/~[v], /p/~[f], /g/~[γ], /k/~[χ], /d/~[ð], /t/~[θ]. In Modern Hebrew, only the fricative allophones of /b/~[v], /p/~[f] and /k/~[χ] remain, and they have come to attain phonemic status in and of themselves. The phonemic sta- tus of the fricatives /v/, /f/ and /χ/ has arisen as a consequence of a number of historical processes that have resulted in violations of the rule C[+plosive] > C[+fricative] /V_. Examples of such processes include vowel syncope (e.g., [malχe(j)] < *malakē king.mpl.bnd ‘kings of’), the simplification of gemination (e.g., [kibes] < *kibbes ‘wash.pst.3msg’), the merg- ers of /w/ > /v/, /q/ > /k/, /ħ/ > /χ/ (e.g., [viter] < *witter ‘give.up.pst.3msg’; [bakar] < *baqar ‘cattle; beef’; [χadal] < *ħadal ‘cease.pst.3msg’), loan words (e.g., [mikroskop] ‘microscope’) and analogy (e.g., [kiven] ‘aim.pst.3msg’ > [jekaven] ‘3msg.aim.fut’, instead of expected [jeχaven]; [tafar] ‘sew.pst.3msg’ > [jitfor] ‘3msg.sew.fut’, instead of expected [jitpor]). Numerous minimal pairs are created as a result of these processes, with some even exhibiting morphophonemic contrasts: e.g., [χiber] ‘connect.pst.3msg’ vs. [χiver] ‘pale.adj’; [sapa] ‘sofa’ vs. [safa] ‘language’; and [lefaχot] ‘at.least.adv’ vs. [le=paχot] for=less.adv ‘for less . . . ’ (Schwarzwald 2011: 526, Bolozky 2013a, 2013c). Gutturals: The pharyngeals /ħ/ and /ʕ/ may be preserved in limited contexts. In a high register of the language reserved for television and radio broadcast, both /ħ/ and /ʕ/ may be pronounced, though such a practice is not as common as it used to be. Also, in the speech of Israelis from North African or Middle Eastern backgrounds, /ħ/ is typically preserved (Shatil 2013).4 /r/: The realization of /r/ exhibits a considerable amount of variation. It is most com- monly realized as a voiced uvular approximant [ʁ], though some realize it as a uvu- lar fricative [ʀ]. The uvular approximant has a tendency to weaken or elide entirely in 576  Philip Zhakevich and Benjamin Kantor common vocabulary or in fast speech: e.g., [ʦariχ laleχet] > [ʦəχ laleχet] need.prs.msg go. inf ‘need(s) to go’. A minority of speakers, mostly from North African or Middle Eastern backgrounds, realize /r/ as an alveolar trill [r], but this pronunciation is also common to the elevated register of television and radio broadcast. Other speakers realize /r/ as a dento-alveolar approximant [ɹ] or uvular trill [ʀ] (Bolozky 2013e). /ʦ/: It is likely that the realization of /ʦ/ as an alveolar affricate [ts] reflects an ancient pronunciation minus glottalization/ejection ([ʦˀ]) (Steiner 1982: 11–44). This is as opposed to the non-affricate, pharyngealized realization of /ʦ/ as [sˁ] in Tiberian Hebrew of the Middle Ages, which likely developed as a result of contact with Arabic. Phonemes in loan words: The phonemes /w/, /ʒ/, /ʧ/ and /ʤ/ occur in loan words: e.g., /ˈwiski/ ‘whisky’, /ʒaˈket/ ‘jacket’, /ʧips/ ‘chips’, and /ʤiˈrafa/ ‘giraffe’. Voicing assimilation: Obstruents often assimilate in voicing to a following obstruent: e.g., /jisgor/ [jizˈgor] 3msg.close.fut ‘he will close’ and /zkenim/ [skeˈnim] old.adj.mpl ‘old people’ (Bolozky 1997, 2013c). 3.2 Vowels The vocalic inventory of Modern Hebrew (see Table 22.4) varies only slightly from the Sephardi pronunciation tradition of five vocalic phonemes from which it derives. Only vowel quality is phonemic, with variations in length being realized phonetically (Bolozky 1997, 2013c). This vocalic system reflects a number of mergers from the Tiberian seven-vowel sys- tem (see Table 22.5; see also Chapter 21). A few further points are worthy of mention: /e/ vowels: In some environments, there may be a distinction between two types of /e/ vowels, namely, the /e/ vowel indicated by the sign ‫( ֶא‬segol) ([e]) and a sort of offglided ֶ [moˈre] ‘teacher.msg’ vs. ‫מֹורי‬ diphthong [ej] indicated by the sign ‫( ֵא‬tsere): e.g., ‫מֹורה‬ ֵ TABLE 22.4  VOWELS Front Back high i u e o low a TABLE 22.5 MODERN REFLEXES OF TIBERIAN VOWELS (EXEMPLIFIED WITH DUMMY CONSONANT ‫)א‬ (1): /e/ (‫)ֵא‬, /ε/ (ֶ‫ א‬, ‫)א‬, ֱ ə5 (‫ > )ְא‬/e/ (2): /a/ (‫ַא‬, ‫)ֲא‬, /ɔ/ ‫ > ָא‬/a/ (3): /o/ (ֹ ‫א‬, ‫)אֹו‬, /ɔ/ (‫ָא‬, ‫ > ) ֳא‬/o/ (1) The vowels represented by the signs ‫( ֵא‬tsere), ‫ ֶא‬, (segol), ‫( ֱא‬xataf segol) and ‫( ְא‬the shewa sign) merge to a front-mid vowel /e/ (lower than tsere). (2) The vowels represented by the signs ‫( ָא‬qamats), ‫( ַא‬patax) and ‫( ֲא‬xataf patax) merge to a low vowel /a/. (3) The vowels represented by the signs ‫( ֳא‬xataf qamats) and ‫( ָא‬qamats qatan [in closed unstressed sylla- ble]) are realized as /o/.  Modern Hebrew 577 [moˈrej] teacher.mpl.bnd ‘teachers of’.6 This likely has roots in the Ashkenazi pronuncia- tion of tsere (Berman 1997: 314, Schwarzwald 2011: 526). Phonetic length: While vowel length is not phonemic, phonetic long vowels may be produced according to prosodic factors. For example, similar to the ancient develop- ment of pretonic lengthening in Biblical Hebrew, though far more limited in scope, pre- tonic vowels in unstressed open syllables sometimes lengthen phonetically: e.g., /ʔaˈni roˈʦa/ [ʔaˈni roːˈʦa] sbj.1csg want.prs.fsg ‘I want’ and /ˈʔima ʃeˈli/ [ˈʔima ʃeːˈli] mom. fsg gen=poss.1csg ‘my mom’ (Bolozky 2013d). Phonetic length may also arise as the result of the deletion of /ʕ/, /h/ or /ʔ/, which results in length being a “semi-distinctive feature” (Bolozky 2013f): e.g., [taːˈvod] (< *taʕavod) 2msg.work.fut ‘you will work’ and [maːpeˈχa] (< *mahapeχa) ‘revolution’. In some cases, such length distinctions can actu- ally result in minimal pairs or morphological distinctions: e.g., [naˈtati] (< *natatti) give. pst.1csg ‘I gave’ vs. [naˈtaːti] (< *nataʕti) plant.pst.1csg ‘I planted’, [taˈvi] (< *taviʔ) 2msg.bring.fut ‘you will bring’ vs. [taˈviː] (< *taviʔi) 2fsg.bring.fut ‘you will bring’, and [jaˈrim] (< *jarim) 3msg.raise.fut ‘he will raise’ vs. [jaːˈrim] (< *jaʕarim) 3msg.deceive. fut ‘he will deceive’ (Schwarzwald 2011: 526; Bolozky 1997, 2013f). 3.3  Diphthongs and triphthongs The three historical diphthongs are /uj/, /oj/ and /aj/: e.g., /kanuj/ buy.pass.ptcp.msg ‘bought’, /goj/ ‘gentile’, /ʔelaj/ ‘to me’. The offglided pronunciation of tsere (ֵ‫[( )י‬ej]) in certain contexts would also qualify as a diphthong (see §3.2). In fact, certain minimal pairs with this diphthong such as [be(j)n] ‘between’ vs. [ben] ‘son’ indicate that the diphthongs are phonemic in Modern Hebrew (Berman 1997: 314, Schwarzwald 2011: 526, 2013b). Diphthongs are often found in loan words and exclamations: e.g., [miki maws] ‘Mickey Mouse’, [plejstejʃen] ‘play station’ and [ʔaw] ‘ouch!’. The rare triphthong is also found in exclamations: e.g., [waj] ‘Oh no!’ and [waw] ‘wow!’ (Schwarzwald 2013b). 3.4  Syllabification and phonotactics The following four syllable types are permissible in Modern Hebrew: CV, CVC, CːV, CːVC. In loan words, these rules are not always followed and additional syllable types are attested: e.g., [ʃtrudel] ‘strudel; at (@)’ and [spring] ‘(brand name)’ (Cohen-Gross 2013). Consonant clusters: Word-initial clusters (#CC) or word-medial clusters (C.CC) fol- low the Sonority Sequencing Principle (SSP), namely, sonority increases from the onset to the nucleus and then decreases from the nucleus to the coda: e.g., [ʃlita] ‘control’, [gmul] ‘reward’, and [dvar-im] thing-mpl ‘things’. When the SSP would be violated, an epenthetic [e] is inserted to resolve the cluster: e.g., [levan-im] ‘white.adj-mpl’, [mesu- dar] ‘tidy.adj.msg’, and [jesod] ‘foundation’. Word-medial strings of three consonants (C.CC) are typically limited to loans or compound words: e.g., [ʔabstrakti] ‘abstract. adj.msg’ and [pesekzman] ‘time-out’. As in Biblical Hebrew, final consonant clusters are permissible in the 2fsg past tense verbal form [katav-t] write.pst-2fsg ‘you wrote’ as well as in some loan words, provided that the final consonant is an obstruent and it obeys the SSP: e.g., [bank] ‘bank’ and [terapist] ‘therapist’ (cf. [ʒaner] ‘genre’ and [faʃizem] ‘Facism’) (Cohen-Gross 2003, 2013). Phonetic gemination and the obligatory contour principle: When the normal forma- tion of a pattern would result in a geminate consonant, an epenthetic [e] is inserted: e.g., 578  Philip Zhakevich and Benjamin Kantor [zalelan] ‘glutton’ (cf. [kamʦan] ‘miser’), [noχeχut] ‘presence’ (cf. [rokχut] ‘pharma- cology’), and [χageg-a] celebrate.pst-3fsg ‘she celebrated’ (cf. [katv-a] write.pst-3fsg ‘she wrote’). This rule, an expression of the Obligatory Contour Principle (OCP), is not always maintained in fast speech: e.g., [hi ˌχageˈga] > [ˌiχagˈga] ‘she celebrated’. In some cases, this rule does not apply and gemination may be realized secondarily across morpheme boundaries: e.g., /ʃavat + ti/ [ʃavat-ti] strike.pst-1csg ‘I was on strike’ and /natan + nu/ [natan-nu] give.pst-1cpl ‘we gave’. It is more common, however, for identi- cal and homorganic consonants to be broken up by an epenthetic vowel or for gemination to be simplified: e.g., [ʃavateti] ‘I was on strike’, [avad-eti] work.pst-1csg ‘I worked’ and [natanu] ‘we gave’ (Bolozky 1997, 2013c: 116–17). 3.5 Stress Modern Hebrew generally reflects the stress system of the Sephardi tradition rather than the Ashkenazi tradition (Reshef 2013b). As in Biblical Hebrew, stress is usually on the ultima, though there are a number of exceptional categories due to historical developments (e.g., segolate nouns, i.e., nouns from the historical pattern *CVCC, such as [ˈsefer] ‘book’ and II–weak verbs such as [ˈkam-u] get.up.pst-3cpl ‘they got up’ [see §4.4.4]). While stress has become entirely predictable in the verbal system in Modern Hebrew, stress in the nominal system has become even more complex (Cohen and Ussishkin 2013). Verbs are normally stressed on the ultima (e.g., [kaˈtav] write.pst.3msg ‘he wrote’; [hiχtiv] dictate.pst.3msg ‘he dictated’; [hitkatev] correspond.pst.3msg ‘he corresponded’) with three exceptions: when a consonant-initial suffix is added (e.g., [kaˈtav-ti] write. pst-1csg ‘I wrote’),7 when a vocalic suffix is added to a verb in the hifʕil stem (e.g., [hiχˈtiv-u] dictate.pst-3cpl ‘they dictated’), and when the verb has a word-final epenthetic vowel, which is ignored in stress assignment (e.g., [hivˈtiaχ] promise.pst.3msg ‘he prom- ised’) (Cohen and Ussishkin 2013). In the nominal system, there is a distinction between “accented” and “unaccented” forms. Most nominal forms in Modern Hebrew are “unaccented” and thus have mobile stress, according to which the stress always falls on the ultima: e.g., [daˈvar] ‘thing’ and [dvaˈr-im] thing-mpl ‘things’. The only native Hebrew category without ultimate stress is the class of segolate forms (e.g., [ˈsefer] ‘book’ > [sfaˈr-im] book-mpl ‘books’). In “accented” nominal forms, stress is fixed on a particular syllable and does not shift if a suffix is added, though longer words can sometimes exhibit a shift toward (but not to) the end of the word: e.g., [ˈtiras] ‘corn’ > [ˈtiras-im] corn-mpl and [ˈtelefon] ‘telephone’ > [ˈtelefon-im]/[teleˈfon-im] telephone-mpl. Instances of such “lexical stress” are particu- larly common in loan words, acronyms (e.g., ‫[ ַמנְּכָ”ל‬manˈkal] ‘CEO’ > ‫ַמנְּכָ”לִים‬ [manˈkal-im] CEO-mpl), and other categories. Function words also behave similarly to nominal forms, though they may undergo a significant degree of reduction and/or neutral- ization of stress (Melčuk and Podolsky 1996, Cohen and Ussishkin 2013). 4 MORPHOLOGY 4.1 Pronouns 4.1.1  Independent pronouns The independent pronominal system in Modern Hebrew differentiates person, number and gender (except in 1csg and 1cpl) (see Table 22.6). The only independent pronoun  Modern Hebrew 579 TABLE 22.6  NOMINATIVE PRONOUNS Singular Plural 1 common ʔani ʔanaχnu/ʔanu 2 masculine ʔata ʔatem feminine ʔat ʔaten 3 masculine hu hem feminine hi hen TABLE 22.7  PRONOMINAL SUFFIXES Group 1 Group 2 1csg =i =aj 2msg =χa =e(j)χa 2fsg =aχ =ajiχ 3msg =o =av 3fsg =a =e(j)ha 1cpl =anu =e(j)nu 2mpl =χem =e(j)χem 2fpl =χen =e(j)χen 3mpl =ahem =e(j)hem 3fpl =ahen =e(j)hen with variant forms is the 1cpl; the common form is ʔanaχnu, while ʔanu belongs to a higher register. 4.1.2  Pronominal suffixes While nominative pronouns appear as independent forms, genitive and accusative forms occur as suffixes on nouns, prepositions and other lexicalized forms. The suffixal pro- nouns appear in two forms. The first set of suffixes (Group 1) is attached to singular nouns and a closed set of prepositions, whereas the second set of suffixes (Group 2) is attached to plural nouns and a second closed set of prepositions (see Table 22.7). Examples of lexemes with suffixes of both groups are given below (see Table 22.8). Suffixes attached to nouns (singular or plural) mark the genitive (e.g., sifr=i book. msg=poss.1csg ‘my book’; sfar=aj book.mpl=poss.1csg ‘my books’), while suffixes attached to prepositions signify the object of the preposition (e.g, ʕal=aj on.prep=poss. 1csg ‘on, about me’). Possessive pronouns are created by appending suffixes to the prep- osition ʃel- ‘of’, while objective pronouns are formed by attaching suffixes to the gram- maticalized morpheme ʔot- obj (Coffin and Bolozky 2005: 168–70). 4.1.3  Demonstrative pronouns Demonstrative pronouns in Modern Hebrew are marked for number and gender. The common set of demonstratives includes ze dem.msg ‘this’, zot dem.fsg ‘this’ and ʔele dem. 580  Philip Zhakevich and Benjamin Kantor TABLE 22.8  PRONOMINAL SUFFIXES Group 1 Group 2 sg nouns gen=suffix acc=suffix pl nouns prep sifr- ‘book’ ʃel- ‘of’ ʔot- sfar- ‘books’ ʕal ‘on’ 1csg sifr=i ʃel=i ʔot=i sfar=aj ʕal=aj 2msg sifr=eχa ʃel=χa ʔot=χa sfar=e(j)χa ʕal=e(j)χa 2fsg sifr=eχ ʃel=aχ ʔot=aχ sfar=ajiχ ʕal=ajiχ 3msg sifr=o ʃel=o ʔot=o sfar=av ʕal=av 3fsg sifr=a ʃel=a ʔot=a sfar=e(j)ha ʕal=e(j)ha 1cpl sifr=enu ʃel=anu ʔot=anu sfar=e(j)nu ʕal=e(j)nu 2mpl sifr=eχem ʃel=aχem ʔet=χem (colloquial form: sifr=e(j)χem ʕal=e(j)χem ʔot=χem) 2fpl sifr=eχen ʃel=aχen ʔet=χen (colloquial form: sifr=e(j)χen ʕal=e(j)χen ʔot=χen) 3mpl sifr=ehem ʃel=ahem ʔot=am sifr=e(j)hem ʕal=e(j)hem 3fpl sifr=ehen ʃel=ahen ʔot=an sifr=e(j)hen ʕal=e(j)hen TABLE 22.9  DISTAL DEMONSTRATIVES Masculine Feminine singular ha-hu ha-hi plural ha-hem ha-hen cpl ‘these’; belonging to a higher register are the pronouns zo dem.fsg ‘this’ and ʔelu dem. cpl (for definiteness and demonstratives, see §5.4). In Modern Hebrew, the forms of distal pronouns, which always occur in post-nominal position, are identical to the 3rd person independent pronouns. Unlike the use of prox- imal demonstrative pronouns, which can occur with or without the definite article, the utilization of distal demonstratives always includes the definite article on both elements of the noun phrase (e.g., ha-sefer ha-hu def-book def-3msg ‘that book’) (see Table 22.9). 4.1.4  Relative pronouns There are two relative pronouns in Hebrew, ʃe- and ʔaʃer. The former, a proclitic lexeme, is used commonly, while the latter, a free form, belongs to a higher register (Zewi 2013: 360). Etymologically, ʃe- most likely derives from ʔaʃer (Huehnergard 2006). See §5.6 for the syntax of relative sentences. 4.1.5  Interrogative pronouns Hebrew interrogatives can be divided into interrogative pronouns and interrogative adverbs. Pronominal interrogatives are the following: mi int ‘who’, ma int ‘what’, ʔe(j)ze int.adj.msg ‘which’, ʔe(j)zo int.adj.fsg ‘which’ and ʔe(j)lu int.adj.cpl ‘which’.  Modern Hebrew 581 Interrogative adverbs, on the other hand, are as follows: ʔe(j)fo int.adv ‘where’, mataj int.adv ‘when’, lama int.adv ‘why’, ʔe(j)χ int.adv ‘how’ and kama int.adv ‘how much, how many’. There are other interrogative adverbs that belong to a higher register (e.g., ke(j)ʦad int.adv ‘how’, he(j)χan int.adv ‘where’, maduaʕ int.adv ‘why’). Another set of interrogative adverbs, which is used only in higher register, introduces yes-no questions: ha- int.adv, ha-ʔim int.adv, klum int.adv, and ve-χi int.adv (Burstein 2013: 316–20). 4.1.6  Indefinite pronouns Modern Hebrew has few independent indefinite pronouns. Words such as ha-kol ‘every- thing’, kulam ‘everyone’, maʃehu ‘something’, and miʃehu ‘someone’, as well as klum ‘nothing’ and meʔuma ‘nothing’, are examples of such lexemes. Other indefinite pro- nouns occur as a combination of a determiner and noun, creating an indefinite phrase. For instance, kol ‘each, every’ can occur with any singular noun (e.g., kol jeled every.det child.indf ‘every child’) as well as in fixed phrases (kol davar every.det thing.indf ‘every thing’, kol ʔeχad every.det one ‘everyone’); when appearing before plural nouns, kol is always followed by the definite article ha- and carries the meaning ‘all of the’ (e.g., kol ha-jelad-im all.det def-child-mpl ‘all of the children’). The pronouns ʃum neg.det ‘no; any’ and ʔaf neg.det ‘no; any’ can occur before any singular noun (e.g., ʃum jeled neg.det child ‘no child’ or ʔaf jeled neg.det child ‘no child’) or in fixed phrases to designate neg- ative indefinite pronouns (e.g., ʃum davar neg.det thing ‘nothing’, ʔaf ʔeχad neg.det one ‘no-one’). Certain quantifier words function as indefinite adverbs (e.g., harbe ‘many.det’, kʦat ‘a.bit.det’, meʕat ‘a.little.det’, hamon ‘a.great.deal.of.det’). The quantifier kama ‘some.det’ always stands before plural nouns (e.g., kama jelad-im some.det child-mpl ‘some children’). The word ‘one’, ʔeχad one.adj.msg, which stands after nouns, can carry the semantics of an indefinite article (e.g., jeled ʔeχad child.msg one.adj.msg ‘one child; a child’); and the plural form of ‘one’ means ‘some’ (e.g., jelad-im ʔaχad-im child-mpl one. adj-mpl ‘some children’) (Fruchtman 2013: 257–60, Coffin and Bolozky 2005: 173–5). 4.2  Nouns and adjectives 4.2.1 Nouns Nouns in Hebrew are marked for gender and number (see Table 22.10) as well as bound or nonbound state (= dependent or independent state; see §5.3). The masculine singular is unmarked, while the feminine singular is marked by a suffixed -a, -at, -it, -et or -ut. Certain feminine nouns are not marked as feminine (ʕir ‘city.fsg’, sakin ‘knife.fsg’). Masculine plural nouns take the suffix -im, while feminine plurals are marked by the suffix -ot8 (see Table 22.10). Certain masculine nouns, however, take the feminine TABLE 22.10  NOMINAL INFLECTION (JELED ‘CHILD’, STUDENT ‘STUDENT’) jeled- student- masculine feminine masculine feminine singular jeled jald-a student student-it plural jelad-im jelad-ot student-im student-ijot 582  Philip Zhakevich and Benjamin Kantor pluralizing marker (e.g., χalon-ot window.m-pl ‘windows’, kir-ot wall.m-pl ‘walls’), while a number of feminine nouns take the masculine pluralizing marker (e.g., naʃ-im woman.f-pl ‘women’, nemal-im ant.f-pl ‘ants’). A number of nouns that occur in pairs take the dual suffix -ajim. This ending occurs on dual body parts (ʕe(j)n-ajim eye.f-du ‘eyes’), on certain temporal nouns (ʃnat-ajim year.f-du ‘two years’, χodʃ-ajim month.m-du ‘two months’) and on certain articles of clothing (garb-ajim sock.m-du ‘socks’). The gender of a noun with the dual ending ­corresponds to the gender of the same noun in the singular form. Thus, feminine singu- lar nouns will remain feminine when appearing with a dual ending (e.g., ʕajin ‘eye.fsg’ ~ ʕe(j)n-ajim eye.f-du ‘eyes’; ʃana ‘year.fsg’ ~ ʃnat-ajim year.f-du ‘two years’), while masculine singular nouns will remain masculine when taking a dual ending (e.g., ʃad ‘breast.msg’ ~ ʃad-ajim breast.m-du ‘breasts’; χodeʃ ‘month.msg’ ~ χodʃ-ajim month.m-du ‘two months’). By default, nouns are indefinite (e.g., bajit ‘a house’). A noun can be marked as definite by prepending the definite article ha- to the noun (e.g., ha-bajit def-house ‘the house’). A noun is also definite when it takes a pronominal possessive suffix (e.g., sifr=i book. msg=poss.1csg ‘my book’). 4.2.2 Adjectives Adjectives, like nouns, are marked for gender and number, and they always stand after the head noun. There are two types of adjectives. In the first type, feminine and plural suffixes are attached directly to the base of the adjective (see tov in Table 22.11). The second type is the result of a substantive being modified with the morpheme -i to form an adjective (see prati in Table 22.11), after which feminine and plural suffixes are attached. The morpheme -i is commonly used to derive adjectives from substantives (e.g., prati ‘private.adj.msg’ < prat ‘detail.msg’; χodʃi ‘monthly.adj.msg’ < χodeʃ ‘month.msg’). This morpheme is also used to derive ordinal numbers (e.g., ʃiʃi ‘sixth.adj.msg’ < ʃeʃ ‘six.f’; see §4.3.2). Adjectival modifiers inflect according to the gender of nouns irrespective of the plural suffix (e.g., χalon-ot gdol-im window.m-pl big.adj-mpl ‘large windows’; ʕar-im gdol-ot city.f-pl big.adj-fpl ‘large cities’). Adjectives modifying nouns with the dual ending take a pluralizing suffix that corresponds to the gender of the noun (garb-ajim χum-im sock.m-du brown.adj-mpl ‘brown socks’; ʕe(j)n-ajim jaf-ot eye.f-du beautiful.adj-fpl ‘beautiful eyes’). Lastly, adjectives modifying a definite noun must take a definite article (e.g., ha-bajit ha-gadol def-house.msg def-big.adj.msg ‘the big house’; sifr=i ha-katan book=poss.1csg def-small.adj.msg ‘my small book’) (Danon 2001: 1073–82, 2013c: 684–8). TABLE 22.11  ADJECTIVAL INFLECTION (TOV ‘GOOD’, PRATI ‘PRIVATE’) tov- prati- masculine feminine masculine feminine singular tov tov-a prati prati-t plural tov-im tov-ot prati-jim prati-jot  Modern Hebrew 583 4.2.3  Bound and nonbound state (dependent and independent state) See §5.3. 4.2.4  Noun patterns As in other Semitic languages, Hebrew nouns and adjectives can be categorized accord- ing to fixed nominal patterns. Some estimate that Modern Hebrew attests to around 300 distinct patterns (Schwarzwald 2013d: 646–67; Even-Shoshan 2003: 2169–78). Such patterns, whose shape is determined by particular vowel sequences and affixes, are super- imposed upon a triconsonantal root in order to form words. While most nominal patterns have been inherited from earlier stages of the language, Modern Hebrew has innovated new patterns as well. Table 22.12 illustrates how different nominal patterns may be super- imposed on the etymological root *k-t-b, which is semantically related to writing (note that etymological *k may be realized as χ and *b as v). 4.3 Numerals 4.3.1  Cardinal numerals Cardinal numbers 1–10, as given in Table 22.13, are marked for gender. In regard to gender marking, numbers 1 and 2 behave as expected; that is, the unmarked base forms are the masculine forms, which can be marked by the feminine suffix/affix -t to form the corresponding feminine forms. The gender marking of numbers 3–10, however, behaves in a manner that is opposite to what is expected: the unmarked base forms are actually feminine forms, which, when marked by the common feminine suffix -a, become mas- culine forms. The number 1 functions adjectively, always standing after the head noun, while num- bers 2–10 behave as quantifiers, positioned before the head noun. Numbers 3–10 are also marked for nonbound or bound (independent or dependent) state; nonbound numerals are used before indefinite nouns, and bound-form numerals occur before definite nouns. Numbers 11–19 are marked for gender. These numbers consist of two elements, the singular masculine or feminine number followed by the element ʕasar for masculine TABLE 22.12  NOMINAL PATTERNS (SAMPLE) Noun Pattern Hebrew Term Gloss CCaC ktav ‘script’ CCiC ktiv ‘spelling’ CCiCa ktiva ‘writing’ CCuCa ktuba ‘marriage contract’ CCuCit ktuvit ‘subtitle’ CaCaCa katava ‘newspaper article’ CCoCet ktovet ‘address’ miCCaC miχtav ‘letter’ miCCaCa miχtava ‘writing desk’ taCCiC taχtiv ‘edict’ 584  Philip Zhakevich and Benjamin Kantor TABLE 22.13  CARDINAL NUMERALS Masculine Feminine Nonbound Bound Nonbound Bound 1 ʔeχad ʔaχad ʔaχat ʔaχat 2 ʃnajim ʃne(j) ʃtajim ʃte(j) 3 ʃloʃ-a ʃloʃ-et ʃaloʃ ʃloʃ 4 ʔarbaʕ-a ʔarbaʕ-at ʔarbaʕ ʔarbaʕ 5 χamiʃ-a χameʃ-et χameʃ χameʃ 6 ʃiʃ-a ʃeʃ-et ʃeʃ ʃeʃ 7 ʃivʕ-a ʃivʕ-at ʃevaʕ ʃevaʕ 8 ʃmon-a ʃmon-at ʃmone ʃmone 9 tiʃʕ-a tiʃʕ-at teʃaʕ teʃaʕ 10 ʕasar-a ʕaser-et ʕeser ʕeser numbers (e.g., ʔarbaʕa ʕasar four.m ten/-teen.m ‘fourteen’) or ʕesre for feminine num- bers (e.g., ʔarbaʕ ʕesre four.f ten/-teen.f ‘fourteen’). The masculine and feminine forms of the number 11 are comprised of the dependent form of the number one followed by ʕasar in the masculine (i.e., ʔaχad ʕasar one.m.bnd ten/-teen.m) and ʕesre in the femi- nine (i.e., ʔaχat ʕesre one.f.bnd ten/-teen.f). Both genders of the number 12 consist of a variant of the number 2, followed by ʕasar and ʕesre (i.e., ʃnem ʕasar two.m ten/-teen.m ‘twelve’, ʃtem ʕesre two.f ten/-teen.f ‘twelve’). Finally, for numbers 13, 17 and 19, the feminine numerals consist of variant forms of the numbers 3, 7 and 9 followed by the expected ʕesre element (i.e., ʃloʃ ʕesre three.f.bnd ten/-teen.f ‘thirteen’, ʃvaʕ ʕesre sev- en.f.bnd ten/-teen.f ‘seventeen’ and tʃaʕ ʕesre nine.f.bnd ten/-teen.f ‘nineteen’), whereas the corresponding masculine numbers consist of the expected forms (i.e., ʃloʃa ʕasar three.m.bnd ten/-teen.m ‘thirteen’, ʃivʕa ʕasar seven.m.bnd ten/-teen.m ‘seventeen’ and tiʃʕa ʕasar nine.m.bnd ten/-teen.m ‘nineteen’). Speakers of Hebrew often confuse the numerals, using forms of the gender that is grammatically incongruent with the gender of the head noun. Numbers 20, 30, 40, 50, 60, 70, 80 and 90 occur only as plural masculine forms and are composed of variants of numbers 3–10. So, 20 is a plural of 10 (i.e., ʕesr-im ten-pl), while 30 and 40 are plurals of 3 and 4 (i.e., ʃloʃ-im three-pl ‘thirty’ and ʔarbaʕ-im four-pl ‘forty’). The number 100 is comprised of the word meʔa hundred.fsg, while 200 is mat-ajim hundred.f-du, a dual form of meʔa. The numbers 300–900 consist of the plural form of meʔa (i.e., meʔ-ot hundred.f-pl), which is preceded by the dependent feminine forms of numbers 3–9 (e.g., ʃloʃ meʔ-ot three.f.bnd hundred.f-pl ‘three hundred’). The thousands are constructed in similar manner as the hundreds. Thus, 1,000 is ren- dered with the lexeme ʔelef thousand.msg, while 2,000 with the dual of ʔelef (i.e., ʔalp-ajim thousand.m-du); 3,000–9,000 are constructed with the plural of ʔelef (i.e., ʔalaf-im thou- sand.m-pl), which is preceded by the dependent masculine forms of the numbers 3–9 (e.g., ʃloʃet ʔalaf-im three.m.bnd thousand.m-pl ‘three thousand’) (Meir 2013: 903–6; Coffin and Bolozky 2005: 177–87).  Modern Hebrew 585 4.3.2  Ordinal numerals Ordinal numbers, as given in Table 22.14, are adjectives and stand after the noun that they modify. Ordinals display agreement in number, gender and definiteness with the noun that they modify (e.g., ha-jom ha-riʃon def-day.msg def-first.msg ‘the first day’) (Coffin and Bolozky 2005: 187–8). The ordinals 2–10 are constructed by modifying the relevant base form with the adjectival morpheme -i, which in the feminine forms precedes the feminine suffixes -a and -t. 4.4 Verbs 4.4.1  Verb forms, tense and mood Hebrew verbs are based on triconsonantal roots (and less frequently on quadriconso- nantal roots). In this section, the root *k-t-b, which is associated with the semantics of writing, will be used to demonstrate the standard conjugation of verbs. In Modern Hebrew, verbs are marked for gender, number, person, tense and mood. The indicative present tense, which is technically a nominal/adjectival/participial form, is not marked for person. A unique verbal noun and infinitive forms are also part of the system. The three tenses of Modern Hebrew, namely, past, present and future, are marked by prefixes, suffixes and vowel sequence patterns. The indicative mood in Modern Hebrew is unmarked. The past tense is distinguished by suffixes that mark gender, number and person (see Table 22.15), and refers to past events, both episodic and habitual. It can also denote the past perfect and present perfect when used with specific adverbs (e.g., kvar ‘already’). The future tense takes prefixes which, as in the past tense, mark gender, number and person (see Table 22.16). The future tense denotes future events, and is also associated with mood and modality, for instance, in requests and in low-frequency regularities. The present tense is inflected only for number and gender (see Table 22.17), and can be considered morphologically a participial form, although it fulfills the duty of both the present tense and the participle. The present tense form is used primarily to express pres- ent tense, including habitual actions. This form can also denote the future tense (maχar TABLE 22.14  ORDINAL NUMERALS Masculine Feminine 1 riʃon riʃon-a 2 ʃeni ʃni-ja 3 ʃliʃi ʃliʃi-t 4 reviʕi reviʕi-t 5 χamiʃi χamiʃi-t 6 ʃiʃi ʃiʃi-t 7 ʃviʕi ʃviʕi-t 8 ʃmini ʃmini-t 9 tʃiʕi tʃiʕi-t 10 ʕasiri ʕasiri-t 586  Philip Zhakevich and Benjamin Kantor TABLE 22.15  PAST TENSE *k-t-b 1csg katav-ti 1cpl katav-nu 2msg katav-ta 2mpl ktav-tem (colloquial form: katav-tem) 2fsg katav-t 2fpl ktav-ten (colloquial form: katav-ten) 3msg katav-∅ 3mpl katv-u 3fsg katv-a 3fpl TABLE 22.16  FUTURE TENSE *k-t-b 1csg ʔe-χtov 1cpl ni-χtov 2msg ti-χtov 2mpl ti-χtev-u 2fsg ti-χtev-i 2fpl 3msg ji-χtov 3mpl ji-χtev-u 3fsg ti-χtov 3fpl TABLE 22.17  PRESENT TENSE *k-t-b msg kotev mpl kotv-im fsg kotev-et fpl kotv-ot TABLE 22.18  IMPERATIVE *k-t-b 2msg ktov 2mpl kitv-u 2fsg kitv-i 2fpl ani noseaʕ le=tel aviv tomorrow.adv sbj.1csg travel.prs.msg to=Tel Aviv ‘tomorrow, I am travelling to Tel Aviv’). The present tense form can also carry a present perfect connota- tion (see §4.4.2). In direct speech narration, the present tense form may also refer to the past (ʔetmol, ʔani metajel, ve=pitʔom ʔani roʔe ʔet david yesterday.adv sbj.1csg stroll. prs.msg and.conj=suddenly.adv sbj.1csg see.prs.msg obj David ‘yesterday, I’m taking a stroll and suddenly I see David’) (Boneh 2013c; Hatav 2010). The imperative mood is used to express commands and requests. An imperative form, marked for gender and number, exists in Modern Hebrew (see Table 22.18), but it is primarily used in weak roots (bo come.imp.msg ‘come!’ < weak root *b-w-ʔ), whereas the use of the imperative in strong roots is generally limited to an elevated register. In common speech, the 2nd person future forms of verbs, as given in Table 22.13, are used most frequently as command forms for strong roots. The verbal noun (see Table 22.19) is a nominal form and can therefore take pronominal possessive suffixes as any other noun. The infinitive can take pronominal possessive or objective suffixes in higher registers.  Modern Hebrew 587 TABLE 22.19  INFINITIVE AND VERBAL NOUN *k-t-b infinitive liχtov verbal noun ktiva The subjunctive/optative mood is marked by the subordinating marker ʃe-, which pre- cedes a verb in the future tense (e.g., ʃe=je-χake that.sbrd=3msg-wait.fut ‘let him wait’). It is also possible to express the subjunctive mood by utilizing the lexeme halevaj ‘let it be’ along with ʃe- ‘that.sbrd’ and a verb in any tense (halevaj ʃe=ʔu-χal lavo let.it.be that.sbrd=1csg-be.able.fut come.inf ‘May it be that I will be able to come’). Finally, the conditional mood is expressed in two ways in Hebrew. For potential events, past, present, and future tense forms are used. Hypothetical or counterfactual events are expressed by using a conjugated past tense of the verb ‘to be’ (i.e., haya ‘be. pst.3msg’) along with the present tense of any verb. Such formations are introduced with certain subordinating elements such as ʔim ‘if’ and lu ‘if’ (e.g., ʔim hu haja ba b=a-zman, hu haja roʔe ʔot=i if.conj sbj.3msg be.pst.3msg come.prs.msg in=def-time.msg, sbj.3msg be.pst.3msg see.prs.msg obj=poss.1csg ‘If he were to come on time, he would have seen me!’) (Boneh 2013b: 696–7). 4.4.2 Aspect While Modern Hebrew verbs are not morphologically marked for viewpoint aspect, verbs may carry an aspectual sense in different contexts. In the past tense, dynamic predicates function perfectively by default, whereas stative predicates function imperfectively. Exceptions to the default state exist, however. Specif- ically, in punctual when-clauses and in clauses modified by time-frame adverbs (e.g., ‘on that night’), dynamic predicates can function imperfectively whereas stative predicates can function perfectively (Boneh 2013a: 211). Contrary to the default state in past tense, in the present tense verbs of all aspec- tual classes (i.e., state, activity, accomplishment, achievement) function imperfectively (Boneh 2013a: 212). The future tense, on the other hand, primarily expresses mood and modality (Boneh 2013c: 748). Aspectual values can also be expressed in a periphrastic manner. The most common of these, briefly mentioned earlier, consists of the past tense of the verb ‘to be’ (haja ‘be. pst.3msg’) and the present tense of any verb; such a structure expresses past habitual action and counterfactual modality (e.g., haji-ti ʕose be.pst-1csg do.prs.msg ‘I used to do; I would do’). Certain auxiliary verbs with infinitives can denote actions that are about to happen (e.g., ʔani ʕomed lesajem sbj.1csg stand.prs.msg finish.inf ‘I am about to finish’) and frequent actions (hu marbe liʃtot sbj.3msg increase.prs.msg drink.inf ‘he drinks a lot’). Yet another periphrastic type utilizes a bleached auxiliary verb combined with a fully lexical verb by means of the conjunction ‘and’ (e.g., ʔani χozer ve=ʔomer sbj.1csg return.prs.msg and.conj=say.prs.msg ‘I repeat’) (Boneh 2013a: 213–15). A perfect reading of a verb may arise with the presence of adverbs such as kvar ‘already’, ʕaχʃav ‘now’, and mi-ze ‘for’, which must come with a temporal noun phrase. The adverb mi-ze ‘for’ can also occur in the present tense, creating a perfect meaning (Boneh 2013a: 215–17). 588  Philip Zhakevich and Benjamin Kantor The derived stems, which are different verbal templates (see §4.4.3), may express var- ious aspectual values. For instance, a root in one verbal template may carry a durative meaning, whereas the same root will carry a punctual meaning in a different template. Such formations, however, are rare (Boneh 2013a: 213, Laks 2013: 31, Doron 2013: 370). 4.4.3  Derived stems Modern Hebrew has seven templates, referred to as stems or binyanim ‘structures’, that are used to derive verbs by means of consonantal patterns, vowel sequences and affixes. The basic template, known as qal or paʕal, has traditionally been viewed as an unmarked template, whereas the other templates derive from qal and are marked with affixes and vowel sequences (Doron 2013: 364–71, Sadan 2013: 925–9). As Table 22.20 displays, Modern Hebrew has three active stems (paʕal, piʕel, hifʕil) that differ in degree of action and three corresponding passive stems (nifʕal, puʕal, hufʕal). Historically, puʕal and hufʕal were mere subsets of the piʕel and hifʕil stems. In theory, there should also be three corresponding templates in the middle voice. In practice, only piʕel has a corresponding stem carrying a middle voice meaning (i.e., hit- paʕel), whereas nifʕal functions as both the passive voice and the middle voice of paʕal; and hifʕil has no corresponding stem with a middle voice. See Table 22.20 for a general picture of the semantics of the different verbal patterns (examples are in the participle/ present tense form). Theoretically, a root can occur in all stems. In such a case, the meaning of the root in each template would generally fall in line with the semantics of each template. Roots are typically restricted to only a few stems, however, and do not always carry the expected meaning of a particular stem. It is also worth noting that the template system of Modern Hebrew is very productive, which allows speakers to innovate new verbs by using roots in new templates. The paʕal stem is a simple active stem and the piʕel and hifʕil generally serve their respective functions in relation to the qal stem. While the piʕel stem can carry an intensive meaning in relation to qal (e.g., qal: ʃavar break.pst.3msg ‘he broke’ ~ piʕel: ʃiber shatter. pst.3msg ‘he shattered’), it can also carry functions such as causative (e.g., qal: lamad study.pst.3msg ‘he studied’ ~ piʕel: limed teach.pst.3msg ‘he taught’), factitive (e.g., qal: gadal grow.up.pst.3msg ‘he grew up’ ~ piʕel: gidel make.grow.pst.3msg ‘he made grow; he raised’), and denominative (e.g., bijet domesticate.pst.3msg ‘he domesticated’, derived from bajit ‘house’). The hifʕil stem is usually causative (e.g., qal: lavaʃ wear.pst.3msg ‘he wore’ ~ hifʕil: hilbiʃ dress.pst.3msg ‘he dressed (s.o.)’) but can also function factitively TABLE 22.20  DERIVED STEMS (msg FORMS) simple intensive causative active paʕal: kotev piʕel: meχatev hifʕil: maχtiv ‘write(s)’ ‘CC(’s), address(es), inscribe(s)’ ‘dictate(s)’ passive puʕal: meχutav hufʕal: muχtav nifʕal: niχtav ‘is CC’d, is addressed, is inscribed’ ‘is dictated’ middle voice ‘is written’ hitpaʕel: mitkatev Ø ‘correspond(s) by writing’  Modern Hebrew 589 (e.g., heʔedim turn.red.pst.3msg ‘he turned red’, factitive of ʔadom ‘red’) and denomina- tively (hifʦiʦ bomb.pst.3msg ‘he bombed’, derived from pʦaʦa ‘bomb’). The nifʕal stem fulfills the passive voice of paʕal. However, it can also express mid- dle voice (ha-χalon niʃbar def-window.msg break.pst.3msg ‘the window broke’), active meaning (niχnas enter.pst.3msg ‘he entered’, nilχam war.pst.3msg ‘he warred’), or incho- ative sense (nizkar remember.pst.3msg ‘he remembered’). The stems puʕal and hufʕal function as passives of piʕel and hifʕil, respectively. Finally, the hitpaʕel stem functions reflexively (hitraχeʦ wash.oneself.pst.3msg ‘he washed himself’; hitlabeʃ get.dressed. pst.3msg ‘he got dressed’), reciprocally (hitkatev correspond.pst.3msg ‘corresponded with’), passively (hitkabel be.accepted.pst.3msg ‘he was accepted’), and inchoatively (hitjaʃev sit.down.pst.3msg ‘he sat down’). Metathesis occurs in the hitpaʕel stem when the first radical of the root is s or ʃ: e.g., hiʃtameʃ use.pst.3msg ‘he used’ < *hitʃameʃ. When the first radical is z, in addi- tion to metathesis, partial assimilation also occurs: e.g., hizdaken grow.old.pst.3msg (< *hiztaken < *hitzaken) ‘he grew old’. Metathesis also occurs in the case of tsade: e.g., hiʦtaʕer be.sorry.pst.3msg ‘he was sorry’. Historically, a first radical tsade would also have brought about partial assimilation (*hitṣaʕer > *hiṣṭaʕer (ṣ = [ʦˀ], [ʦˁ], [sˀ], or [sˁ]; ṭ = [tˀ] or [tˁ]) > hiʦtaʕer be.sorry.pst.3msg ‘he was sorry’), but due to the merger of his- torical tˀ (or tˁ), t > t, the assimilation is no longer realized in speech but is still reflected in the orthography (Schwarzwald 2013a). The piʕel template, as well as the related templates puʕal and hitpaʕel, were histor- ically marked by a geminated middle consonant (i.e., piʕel < piʕʕel; puʕal < puʕʕal; hitpaʕel < hitpaʕʕel). For example, the common piʕel verb diber speak.pst.3msg ‘he spoke’ historically derives from dibber (with a geminated middle consonant), which was orthographically marked by a dot (dagesh) in the middle consonant to signify gemination. Because these three templates geminated the middle root consonant, it was possible for quadriconsonantal or reduplicated biconsonantal roots to occur in these templates (e.g., piʕel: gilgel roll.pst.3msg ‘he rolled’ < g-l-g-l). This phenomenon continues to exist in Modern Hebrew (e.g., piʕel: tirgem translate.pst.3msg ‘he translated’ < t-r-g-m). These three stems are highly productive in forming denominative roots (both triconsonantal as well as quadriconsonantal) from foreign words. For instance, the 3msg past tense piʕel and puʕal forms, fikes ‘he focused’ and fukas ‘it was focused’, are built upon the newly created triconsonantal root f-k-s, which was extracted from the English word ‘focus’ (i.e., f-k-s < focus); similarly, the 3msg past tense piʕel and puʕal forms, fikses ‘he faxed’ and fuksas ‘it was faxed’, are built upon the newly shaped quadriconsonantal root f-k-s-s, which derives from the English word ‘fax’ (i.e., f-k-s-s < fax) (Bat-El 2013: 704–9). Formally, the seven verbal templates have the following distinguishing marks (see Table 22.21): 1 The qal stem is marked by an a–a vowel sequence in the past, an o–e vowel sequence in the present (as well as a–e for stative verbs), and an i–o vowel sequence in the future, with variations in roots containing gutturals and glides. 2 The piʕel stem is marked by an i–e vowel sequence in the past, a preformative me- and an a–e vowel sequence in the present, and an e–a–e vowel sequence in the future. 3 The hifʕil stem is marked by a preformative hi- and an i theme vowel in the past, a preformative ma- and an i theme vowel in the present, and an a–i vowel sequence in the future. 590  Philip Zhakevich and Benjamin Kantor TABLE 22.21 BASIC FORMS OF THE DERIVED STEMS (PAST AND FUTURE FORMS ARE 3msg, WHILE PRESENT AND IMPERATIVE FORMS ARE msg) qal piʕel hifʕil nifʕal puʕal hufʕal hitpaʕel past katav kitev hi-χtiv ni-χtav kutav hu-χtav hit-katev present kotev me-χatev ma-χtiv ni-χtav me-χutav mu-χtav mit-katev future ji-χtov je-χatev ja-χtiv ji-katev ju-χtav ju-χtav jit-katev imperative ktov katev ha-χtev hi-katev Ø Ø hit-katev infinitive liχtov leχatev lehaχtiv lehikatev Ø Ø lehitkatev verbal noun ktiva kituv haχtava hikatvut Ø Ø hitkatvut 4 The nifʕal stem is marked by a preformative ni- and an a theme vowel in the past and present, and an i–a–e vowel sequence in the future. 5 The puʕal is marked by a u–a vowel sequence in the past, a preformative me- and a u–a vowel sequence in the present, and a u–a vowel sequence in the future. 6 The hufʕal stem is marked by a preformative hu- and an a theme vowel in the past, a preformative mu- and an a theme vowel in the present, and a u–a vowel sequence in the future. 7 The hitpaʕel stem is marked by a preformative hit- and an a–e vowel sequence in the past, a preformative mit- and an a–e vowel sequence in the present, and a prefix t- and an i–a–e vowel sequence in the future. 4.4.4  Root and weak verbs The verbal paradigms given previously display the standard conjugation patterns of strong roots, whereas weak roots deviate from the standard pattern in their vowel sequences and consonantal patterns. Strong roots consist of any consonant excluding a guttural, a glide, n or r, whereas roots defined as weak are those that contain any of the aforementioned consonants. As a result of historical processes, Modern Hebrew has inherited from ancient Hebrew varying vowel patterns in verbal forms whose roots originally contained the gutturals *h, *ħ (χ in Modern Hebrew), *ʔ, or *ʕ (generally ʔ in Modern Hebrew) (as well as r): e.g., laʔavod [laʔavod]/[laːvod] work.inf ‘to work’ (< *laʕavod [root: *ʕ-b-d]; cf. liχtov write. inf ‘to write’ [root: *k-t-b]). When the gutturals h, χ (< *ħ), or ʔ (< *ʕ) occur word-finally, an epenthetic a vowel is added just before the close of the syllable (e.g., liʃmoaʔ listen.inf ‘to listen’ [< *liʃmoʕ]). Also, as discussed in §3.1, the voiced pharyngeal ʕ, the glottal stop ʔ, and the glottal fricative h have come to be realized as zero (Ø) in most environments (e.g., koret read.prs.fsg ‘write(s)’ < *q-r-ʔ [cf. kotevet write.prs.fsg ‘write(s)’ < *k-t-b]). There is an important distinction to be made regarding verbal forms that are spelled with a final -h in the pst.3msg form. Such verbal forms fall into two categories with respect to their triconsonantal root: those that derive from a root with an etymological h as the third radical (e.g., tama [‫=תמה‬tmh] be.amazed.pst.3msg ‘he was amazed’ < *t-m-h) and those that derive from a root with an etymological j or w as the third radical (e.g., raʔa [‫=ראה‬rʔh] see.pst.3msg ‘he saw’ < *r-ʔ-j). Roots with an etymological h display the h orthographically in the verbal paradigm, although it is usually deleted in pronunciation, resulting in the lengthening of the adjacent vowel (e.g., [taˈmaːti] (< *tamah-ti) be.amazed. pst-1csg ‘I was amazed’); as discussed in §3.2, this also occurs in roots containing other  Modern Hebrew 591 gutturals as the final consonant. Similarly, in verbal and nominal patterns of roots with original j (less frequently w), the j has generally not been preserved in pronunciation. It is, however, preserved in verbal nouns (e.g., reʔija ‘vision’ < *r-ʔ-j); and although not usually pronounced, an original j is often still reflected in the spelling of certain past tense forms (e.g., raʔi-ta [‫ראית‬ = rʔjt] see.pst-2msg ‘you saw’ < *r-ʔ-j). Roots with initial j or n display certain deviations in the infinitive and imperative forms. Specifically, in infinitives and imperatives of such roots, the first syllable begin- ning with j or n is dropped entirely (e.g., ʃev sit.imp.msg ‘sit!’ < j-ʃ-b [cf. ktov write.imp. msg ‘write!’ < *k-t-b]). Also, when n appears as the first phoneme in a consonant clus- ter, it usually assimilates to the second consonant (e.g., ji-pol 3msg-fall.fut ‘he will fall’ < *jippol < *jinpol < *n-p-l). In j-initial roots that historically derive from w-initial roots, an original w is preserved as an o vowel in post-vocalic environments (e.g., word-initial y (< *w): jald-a give. birth.pst-fsg ‘she gave birth’; cf. post-vocalic w: nold-a be.born.pst-fsg ‘she was born’ < *nawladat < *w-l-d). Also, in such roots an original w is preserved as v if the origi- nal w was historically geminated (e.g., lehivaled be.born.inf ‘to be born’ < *lVhiwwaled < *w-l-d). In roots with a glide as their second consonant, the glide falls out entirely in most of the verbal paradigm (e.g., kam arise.prs.msg ‘arise(s)’; kam-ti arise.pst-1csg ‘I arose’; ʔa-kum 1csg-arise.fut ‘I will arise’ < *q-w-m). Lastly, in discussing the root and weak verbs, it is noteworthy that roots with an iden- tical second and third radical often drop the latter (e.g., nifʕal past tense: namas melt. pst.3msg ‘it melted’ < *m-s-s). 4.5  Prepositions, phrasal verbs, adverbs and conjunctions 4.5.1  Prepositions Prepositions stand before nominals in Modern Hebrew. Certain prepositions consist of individual morphemes, a small number of which are proclitics (e.g., le- ‘to’, be- ‘in’, mi-/me- ‘from’), while others are free forms (e.g., ʕal ‘on, about’, min ‘from’, ʔel ‘to, toward’). Some are comprised of two morphemes (e.g., meʕal ‘above; from on top of’ [lit: me=ʕal from.prep=on.prep ‘from on/above’]) and many derive etymologically from nouns (e.g., lejad ‘next to’ [lit: le=jad to.prep=hand.fsg.bnd ‘to/at the hand of’] and biʃvil ‘for the sake of’ [lit: bi=ʃvil in.prep=path.msg.bnd ‘in the path of’]) (Botwinik 2013b). Certain verbs appear to take a prepositional phrase rather than a direct object as a complement. Such phrasal verbs often take the preposition be- ‘in’ (e.g., hu bataχ b=a sbj.3msg trust.pst.3msg in.prep=poss.3fsg ‘he trusted her’), while other verbs occur with ʕal ‘on, about’ (e.g., hu hitχaret ʕal ze sbj.3msg regret.pst.3msg about.prep dem.msg ‘he regretted this’) or le- ‘to’ (e.g., hu hifriaʕ l=a sbj.3msg interrupt.pst.3msg to.prep= poss.3fsg ‘he interrupted her’). Some verbs can occur with either an object or a preposi- tion (e.g., hu baʕat b=o sbj.3msg kick.pst.3msg in.prep=poss.3msg ‘he kicked him’ or hu baʕat ʔot=o sbj.3msg kick.pst.3msg obj=poss.3msg ‘he kicked him’) (Botwinik 2013a). 4.5.2  Adverbs Adverbs, which usually occur post-verbally, are formed in various ways in Modern Hebrew. The most common method is to prefix a preposition to a noun or adjective, creating a prepositional phrase that can function adverbially (e.g., bi=mhirut in.prep=quickness.fsg 592  Philip Zhakevich and Benjamin Kantor ‘quickly’, b=a-riʃon-a in.prep=def-first.adj-fsg ‘at first’, k=a-rega as.prep=def-moment ‘now’); many of these forms have become lexicalized (e.g., kimʕat [ki=mʕat as.prep= little.msg] ‘almost’, mijad [mi=jad from.prep=hand.fsg] ‘immediately’). The preposi- tional phrase be=ʔofen in.prep=manner.msg ‘in a specific manner’ is used with various adjectives to produce adverbial phrases (e.g., be=ʔofen klali in.prep=manner.msg general. adj.msg ‘generally, in a general manner’, be=ʔofen kriti in.prep=manner.msg critical.adj. msg ‘critically, in a critical manner’). It is also possible to form an adverbial by suffixing the feminine plural marker -ot to adjectives (gluj-ot revealed.adj-fpl ‘transparently’) or by suffixing the feminine singular marker -t to the adjectival suffix -i (e.g., ʔiʃi-t personal.adj-fsg ‘personally’). Addition- ally, one can attach the suffix -a to the four cardinal directions or to place names in order to create directional adverbs (e.g., ʦafon-a north-dir ‘northward’, tel aviv-a Tel Aviv-dir ‘Tel Aviv-ward’). The masculine singular forms of most adjectives can be used as adverbs (e.g., ha-mivχan kaʃe def-test.msg difficult.adj.msg ‘the exam is hard’; hem lomd-im kaʃe sbj.3mpl study.prs-mpl difficult.adj.msg ‘they study hard’). The definite article can be prefixed to terms related to time to create temporal adverbs (e.g., ha-jom def-day ‘today; the day’, ha-ʃavuaʕ def-week ‘this week; the week’). There are also lexicalized terms that function as adverbs (e.g., meʔod ‘very’, harbe ‘a lot’) (Oren 2013: 52, Mor 2013: 44–9, Ravid and Shlesinger 2000: 342–4). 4.5.3  Conjunctions Coordination is usually expressed with the conjunctions ve- ‘and’, ʔo ‘or’, ʔaval ‘but’, and ʔela ‘(but) rather’. Subordination is marked by the conjunctions ʃe- and ʔaʃer, both of which mean ‘that, which, who(m)’, ki ‘that; because’ and ʔim ‘if, whether’. There are many more subordinating conjunctions, most of which require a following ʃe- (e.g., lif- ne(j) ʃe- ‘before’; ʔaχare(j) ʃe- ‘after’; biglal ʃe- ‘because’). Conjunctions always precede the element that they are conjoining (Glinert 2013a: 566–9). 4.5.4  Negation markers The three common sentence negation markers are lo, ʔe(j)n and ʔal. For most cases, in statements and questions, lo is used as the negator; ʔe(j)n fulfills an identical function but it belongs to a higher register. The negator ʔal with the future tense functions as the negation of the imperative, and, in a higher register, it can negate 1st and 3rd person commands. There are many other negative words that can occur along with the aforementioned neg- ative markers. A selection of such words are klum ‘nothing’, ʃum ‘any’, ʔaf ‘not a sin- gle, even’, ʔaf paʕam ‘not even once’, leʕolam ‘never’ and bli ‘without’ (Glinert 2013b: 811–14). See §5.7 for syntax of negation markers. 5 SYNTAX 5.1  Word order 5.1.1  Sentential word order The typical word order for sentences with an overt subject is SVO (possibly followed by an adverbial) (Giora 1982, Ilani, Shlomo, and Goldberg 2013, Halevy 2013b):  Modern Hebrew 593 ha-ʕitonaj hikʃiv l=a-sipur be=ʕinjan rav def-journalist.msg listen.pst.3msg to=def-story.msg in=interest.msg great.msg ‘The journalist listened to the story with great interest’. There are, however, a number of factors that give rise to variants in word order (e.g., disambiguation, focus, length, legal language). One clear example is when a sentence begins with a complement (especially in written Hebrew), in which case the basic SVO word order normally changes to VSO as long as the verb is in the past or future tense: be=ʕinjan rav hikʃiv ha-ʕitonaj l=a-sipur in=interest.msg great.msg listen.pst.3msg def-journalist.msg to=def-story.msg ‘With great interest the journalist listened to the story’. If the verb is in the present, both SVO and VSO are acceptable after a complement: be=ʕinjan rav ha-ʕitonaj makʃiv l=a-sipur in=interest.msg great.msg def-journalist.msg listen.prs.msg to=def-story.msg or be=ʕinjan rav makʃiv ha-ʕitonaj l=a-sipur in=interest.msg great.msg listen.prs.msg def-journalist.msg to=def-story.msg ‘With great interest the journalist listens to the story’. This variation does not apply, however, when the subject is a pronoun: maχar hu ja-vo ʔel=aj tomorrow.adv sbj.3msg 3msg-come.fut to=poss.1csg ‘Tomorrow he will come to me’. In existential sentences, the VS word order occurs: jored geʃem rain.prs.msg rain.msg ‘It is raining’. Similarly, in statements of existential possession which utilize the existential particles jeʃ ‘there is/are’ and ʔe(j)n ‘there is/are not’, the VS word order is preferred (see also §5.2): jeʃ l=i meχonit exist to=poss.1csg car ‘I have a car’. 5.1.2  Phrasal word order Phrasal word order is Head-Dependent (Borochovsky 1986, Berman 2013, Danon 2013b, Faust 2013, Halevy 2013a): P-P: ʕal ha-ʃulχan on def-table ‘on the table’ 594  Philip Zhakevich and Benjamin Kantor N-N: miʦ tapuz-im juice oranges-mpl ‘orange juice’ N-Adj: χalon-ot gdol-im window.m-pl big.adj-mpl ‘big windows’ N-Rel: ha-jeled ʃe=kana ʔet ha-matana ʃe=bikaʃnu lekabel def-boy rel=buy.pst.3msg obj def-gift rel=ask.pst.1cpl receive.inf ‘the boy who bought the gift that we asked to receive’ N-Dem: ha-makom ha-ze def-place def-dem.msg ‘this place’ Most modifiers follow the noun in the following order: genitive attribute, adjective, gen- itive exponent (ʃel), prepositional phrase and attributive clause: ʃulχan ha-ktiva ha-jaʃan ʃel=i table.msg.bnd def-writing.fsg def-old.msg gen=poss.1csg ‘my old desk’ There are, however, a select number of modifiers that precede the noun (e.g., quantifiers, prenominal determiner ʔoto/ʔota/ʔotam/ʔotan ‘that. . . ; the same. . .’): ʃloʃa ʔanaʃ-im three.m people-mpl ‘three people’ kol ha-ʔanaʃ-im all def-people-mpl ‘all the people’ ʔot-o ha-ʔiʃ dem-msg def-man.msg ‘that man, the same man’ 5.2  Types of predication Modern Hebrew has three types of predication: verbal, nominal and existential. Verbal predication is achieved by means of a finite verb phrase. Nominal predication is achieved by means of a noun phrase, adjective phrase, prepositional phrase or adverbial phrase. In the case of nominal predication, the addition of a copula, which is identical with the 3rd person independent or demonstrative pronouns, may be implemented for clarity (Ornan 1979, Danon 2013b, Uziel-Karl 2013): Verbal predication: Dan noten neʃika le=Rina pn give.prs.msg kiss.fsg to=pn ‘Dan gives a kiss to Rina’.  Modern Hebrew 595 Nominal predication without copula: ha-jeled b=a-bajit def-boy in=def-house.msg ‘The boy is at home’. ha-sefer ha-ze meʔod populari ʕaχʃav def-book.msg def-dem.msg very popular.adj.msg now ‘This book is very popular now’. Nominal predication with copula: ha-sefer ha-ze hu meʔod populari ʕaχʃav def-book.msg def-dem.msg sbj.3msg very popular.adj.msg now ‘This book is very popular now’. Existential predication is achieved by means of the existential particle jeʃ exist ‘there is/ are’ or ʔe(j)n ‘there is/are not’. jeʃ harbe ʔanaʃ-im ʃe=lo meχabd-im ʔeχad ʔet ha-ʃeni exist many.adv people-mpl rel=neg respect.prs-mpl one.msg obj def-second.msg ‘There are many people who do not respect one another.’ ʔe(j)n ʦedek b=a-ʕolam neg.exist justice.msg in=def-world.msg ‘There is no justice in the world’. Hebrew is a “non-habere” language, and a combination of the existential particles and the preposition le- ‘to’ with suffixes is used to indicate possession. In less formal language, the direct object marker can even precede the grammatical subject. This construction is referred to in the literature as a verboid: jeʃ l=i sefer exist to=poss.1csg book ‘I have a book’. jeʃ l=i ʔet ha-sefer ve=jeʃ ʔot-o b=a-sifrijah exist to=poss.1csg obj def-book conj=exist obj=3msg in=def-library.fsg ‘I have the book and it is also in the library’. With respect to predication, there are a number of ways to formulate impersonal state- ments in Hebrew (Halevy 2013b). It is possible, for instance, to use one of the passive stems, as in the following example with the nifʕal stem: ha-bajit neheras def-house.msg be.destroyed.pst.3msg ‘the house was destroyed’ (see §4.4.3). One can also employ the 3rd person masculine plural form of an active stem to denote a depersonalized discourse stance (e.g., harsu ʔet ha-bajit destroy.pst.3cpl obj def-house ‘they destroyed the house’). Yet another way to create impersonal statements in Hebrew is to use the 3rd person mas- culine singular form of a passive verb. This structure is called in Hebrew by the acronym χagam (i.e., χaser guf ve-min ‘lacking person and gender/number’): huχlat leʃanot ʔet ha-χok be.decided.pst.3msg change.inf obj def-law ‘It was decided to change the law’. 596  Philip Zhakevich and Benjamin Kantor The χagam paradigm is in fact very dynamic and heterogeneous. In addition to occurring with 3rd person forms of passive stems, the structure also commonly occurs with modals: mumlaʦ lifnot l=a-rofe recommended.adj.msg turn.inf to.prep=def-doctor ‘. . . it is recommended to turn to a doctor’. χaʃuv l=i lehagid ʃe important.adj.msg to=poss.1csg say.inf comp ‘It is important for me to say that . . .’ 5.3 Synthetic/analytic A genitival relationship between nouns in Modern Hebrew can be achieved either syn- thetically or analytically. In the case of nouns with the feminine singular ending (-a) or with the masculine plural ending (-im), the genitival relationship is indicated by direct apposition of the nouns (i.e., the head noun is immediately followed by the dependent noun(s) without the genitive exponent ʃel intervening) as well as an overt bound form (bnd) of the fsg/mpl morphemes (fsg -at; mpl -e(j); these are also called construct forms). In the case of unmarked nouns (-Ø) or nouns with feminine plural morphological endings (-ot), the dependent genitival relationship is indicated by direct apposition of the nouns and possibly also by a variant vowel pattern. Other morphosyntactic features, such as agreement and definiteness in adjectives modifying the genitive chain, can also indicate that a sequence of nouns is in the bound (construct) state. Gender is determined by the head noun and definiteness is marked only on the final element of the chain (Glinert 1989: 33–49, Doron and Meir 2013, Edzard 2013b, Ilani, Shlomo, and Goldberg 2013): roʃ memʃala head.msg.bnd government ‘prime minister’ be(j)t ha-miʃpat (cf. bajit) house.msg.bnd def-judgment (house.msg.nbnd) ‘courthouse’ dira-t Moʃe (cf. dira) apartment-fsg.bnd pn (apartment.fsg.nbnd) ‘Moshe’s apartment’ The same genitival relationship can be expressed analytically in three different ways. First, the nonconstruct form may be used with the genitive exponent ʃel: ha-dira ʃel Moʃe def-apartment gen pn In mostly formal and written registers, the head noun may be combined with a cataphoric pronoun and followed by the genitive exponent ʃel: dira-t=o ʃel Moʃe apartment-bnd=poss.3msg gen pn  Modern Hebrew 597 Finally, the genitive exponent may be combined with a preposition indicating authorship: maʔamar mi=ʃel Moʃe article from=gen pn ‘an article by Moses’ 5.4  Definiteness Formally, definiteness in Modern Hebrew is most commonly expressed either by pre- fixing the definite article ha- to a noun or by adding a possessive suffix to a noun: e.g., ha-sefer def-book ‘the book’ and sifr=o book=poss.3msg ‘his book’. Other determiners such as demonstratives and the prenominal determiner ʔoto ‘that . . . ; the same . . .’ may also indicate definiteness: e.g., ha-sefer ha-ze def-book.msg def-dem.msg ‘this book’ and ʔoto ha-sefer same/that.det.msg def-book.msg ‘that/the same book’. In each of these examples, the article is actually optional: e.g., sefer ze book.msg dem.msg ‘this book’ and ʔoto sefer same/that.det.msg book.msg ‘that/the same book’. The demonstrative without the article in the example sefer ze ‘this book’ is reflective of a formal register. Both demonstratives and adjectives must agree in definiteness with the head noun: e.g., ha-sfar-im ha-ʔele def-book-mpl def- dem.cpl ‘these books’, sifrija zo library.fsg dem.fsg ‘this library’, sefer ʔadom book.msg red.adj.msg ‘(a) red book’, and ha-sifrija ha-ʔaduma def-bookcase.fsg def-red.adj.fsg ‘the red bookcase’. When used pronominally, a demon- strative is inherently definite and thus preceded by the obj when used as an object (e.g., ze sefer dem.msg book.msg ‘this is a book’; kara-ti ʔet ze read.pst-1csg obj dem.msg ‘I read this’) (Fruchtman 1982, Glinert 1989: 12–23, 91–101; Wintner 2000, Coffin and Bolozky 2005: 170–2, Danon 2001, 2013c). Nouns in the construct state are marked as definite by virtue of the last element of the construct state being marked as definite: e.g., χalon ha-bajit window.msg.bnd def-house. msg ‘the window of the house’ and ʕekron χofeʃ ha-bituj principle.msg.bnd freedom.msg. bnd def-expression.msg ‘the principle of freedom of speech’. In colloquial speech, it is not uncommon for the definite article to be attached to the initial element of the con- struct chain rather than the final element: e.g., ha-ʕoreχ din def-editor/arranger.msg.bnd judgment.msg ‘the lawyer’ (cf. ʕoreχ ha-din editor/arranger.msg.bnd def-judgment.msg), ha-be(j)t sefer def-house.msg.bnd book.msg ‘the school’ (cf. be(j)t ha-sefer house.msg. bnd def-book.msg) and ha-ʕuga-t tapuχ-im def-cake-fsg.bnd apple.m-pl ‘the apple cake’ (cf. ʕuga-t ha-tapuχim cake-fsg.bnd def-apple.m-pl). This is especially common with lex- icalized compounds and construct pairs in which the second element is a non-referential modifier (Doron and Meir 2013, Edzard 2013a). There is no explicit indefinite marker in Modern Hebrew, though the adjectival numeral ʔeχad/ʔaχat ‘one’ is occasionally implemented to express the idea of ‘a cer- tain’: e.g., ʔiʃ ʔeχad man.msg one.adj.msg ‘a certain man’. The interrogative ʔe(j)ze ‘which?’ is also utilized for a similar purpose, though with a narrower meaning of ‘some (kind of a)’: e.g., hu halaχ le=ʔe(j)ze mesiba sbj.3msg go.pst.3msg to=which. int (to=which.indf) party.fsg ‘he went to a/some party’. A similar use has also been identified for the noun min ‘kind’, though its meaning is more along the lines of ‘a kind of/some sort of’ (with the optional addition of ka=ze as.prep=dem.msg, ka=zot as. prep=dem.fsg ‘like this’): e.g., jeʃ l=a min marʔe klasi (ka=ze) exist to=poss.3fsg kind. msg.bnd appearance.msg classic.adj.msg (as.prep=dem.msg) ‘she has a kind of classic look’ (Rubin 2013). 598  Philip Zhakevich and Benjamin Kantor 5.5 Case Modern Hebrew does not exhibit a full-fledged system of morphologically marked case. Pronouns, however, are indeed marked for accusative case (see ʔot- obj in Table 22.8 of §4.1.2). Likewise, directional case occurs to a limited degree in the language (e.g., ʔani noseaʕ darom-a/tel aviv-a sbj.1csg travel.prs.msg south-dir/Tel Aviv-dir ‘I am traveling south(ward)/to Tel Aviv’) (Danon 2013a: 397). There are two environments in which case is observable in Modern Hebrew. First, structural case is apparent in genitival constructions formed synthetically (see earlier). In such constructions, head nouns (albeit only fsg and mpl nouns) undergo morphological changes and appear adjacent to genitival nouns, thereby assigning the genitive case in a specific structural configuration (Danon 2013a: 398). In such structures, adjectives and prepositions can also function as the head (e.g., tov-e(j) lev good.adj-mpl.bnd heart.msg ‘good of heart, good-hearted’; lifne(j) ha-ʃulχan before.prep def-table.msg ‘in front of the table’). Second, case is observable in the use of the lexicalized element ʔet obj, which appears before all definite direct objects (e.g., ʔani kore ʔet ha-sefer sbj.1csg read.prs.msg obj def-book.msg ‘I am reading the book’). Traditionally, ʔet has been interpreted as marking the accusative because it appears before direct objects, albeit definite direct objects. More recently, though, ʔet has been viewed as a preposition-like element functioning as the head of definite direct objects, thereby assigning structural genitive case to direct objects (Danon 2013a: 398–9). 5.6 Subordination Relative clauses in Modern Hebrew are usually introduced by the particles ʔaʃer, which was the common relative particle of Biblical Hebrew, and ʃe-, which was the common relative particle of Rabbinic Hebrew. In addition to ʔaʃer and ʃe-, the definite article ha- can also be utilized in place of a relative before a participle or adjective. This usage is characteristic of formal register (Glinert 1989: 361–75, Halevy 2013b, Kotek 2013, Zewi 2013): Relative particle ʔaʃer (example from the internet): hem ha-manhig-im ʔaʃer jaʦ-u me=ha-tnuʕa sbj.3mpl def-leaders-mpl rel come.out.pst-3cpl from=def-movement ‘They are the leaders who left the movement’. Relative particle ʃe-: ʔe(j)fo ha-brag-im ʃe=jaʦ-u where.int def-screws-mpl rel=come.out.pst-3cpl ‘Where are the screws that came out?’ Definite article ha- used as a relative: ʃel ʔanaʃ-im ha-makir-im ze ʔet ze he(j)tev rel people-mpl def-know.prs-mpl dem.msg obj dem.msg well.adv ‘. . . of people who know each other well’.  Modern Hebrew 599 When the antecedent is also the direct object or prepositional object within the relative clause, a resumptive pronoun corresponding to the antecedent may be used: ha-ʦijur ʃe=hitbonan-ta b=o def-painting.msg rel=look.pst-2msg in=poss.3msg ‘the painting at which you looked’ ha-ʃir ʃe=ʔani ʔohev ʔot=o def-song.msg rel=sbj.1csg like.prs-msg obj=poss.3msg ‘the song that I like’ Resumptive pronouns may also be omitted, especially in the case of the direct object resumptive pronoun: ha-ʃita ʃe=hiʃtamaʃ-nu (b=a) lo matʔim-a def-system.fsg rel=use.pst-1cpl (in=poss.3fsg) neg appropriate.adj-fsg ‘The system that we used (it) isn’t appropriate’. ha-ʃir ʃe=ʔani ʔohev def-song.msg rel=sbj.1csg like.prs-msg ‘the song that I like’ Finally, while some sort of relative particle is obligatory in common register, asyn- detic relative clauses are found in highly formal or poetic registers. Such a con- struction is often expressed by means of a preposed prepositional phrase with a resumptive pronoun in the absence of the relative particle (Halevy 2013b; see also Cohen 2016): ha-ʦijur b=o hitbonan-ta def-painting.msg in=poss.3msg look.pst-2msg ‘The painting at which you looked’. ha-bajit b=o gar-ti def-house.msg in=poss.3msg reside.pst-1csg ‘the house in which I lived’. ha-ʔiʃa ʔot=a pagaʃ-ti def-woman.fsg obj=3fsg meet.pst-1csg ‘the woman whom I met’. 5.7 Negation There are three primary negative particles in Modern Hebrew: lo, ʔe(j)n and ʔal. The various usages of these particles are mainly determined syntactically, though they may also vary according to register. The main negative particle, lo, is used for negation in both main clauses and subordinate clauses in most contexts, such as negating verbs in all forms (but see comments on the participle and ʔe(j)n later) and negating adjectives 600  Philip Zhakevich and Benjamin Kantor (Glinert 1982, 1989: 293–307, Berman 1997: 327, Schwarzwald 2011: 532, Agmon 2013, Bolozky 2013b, Glinert 2013b, Halevy 2013b): hu be=vadaj lo jadaʕ sbj.3msg in=certainty neg know.pst.3msg ‘He certainly did not know’. hem lo joʃv-im sbj.3mpl neg sit.prs-mpl ‘They do not sit; They are not sitting’. ʔata lo ta-zuz sbj.2msg neg 2msg-move.fut ‘You won’t move’. sipur=enu matχil be=makom lo raχok story=poss.1pl begin.prs.msg in=place.msg neg far.adj.msg ‘Our story begins in a place not far away (lit. “a not-far place”)’. (example from the internet) While ʔal is used formally to express negative commands (see later), lo following the complementizer ʃe- may also be used to express a prohibitive or negative volitional: ʃe=lo te-ʦe me=ha-bajit rel=neg 2msg-leave.fut from=def-house ‘don’t you dare leave the house!’ As is common in Semitic, the imperative form cannot be negated. Instead, to express a negative command, the particle ʔal is normally utilized, before a 2nd person form of the “future” conjugation: ʔal ta-zuz neg 2msg-move.fut ‘don’t move’ In formal register, the particle ʔal may also precede the 1st and 3rd person verbal forms of the prefix conjugation to indicate a negative volitional: e.g., ʔal ni-ʃkaχ neg 1cpl-forget. fut ‘let us not forget’. The particle ʔe(j)n is regarded as the more formal or “correct” negative particle for negating the present tense (cf. lo). When following the subject (or in lieu of an overt sub- ject), the particle ʔe(j)n inflects with suffixes. When preceding an explicit subject, ʔe(j)n does not inflect: ha-ʦav ʔe(j)n=o zaz def-turtle.msg neg.exist=poss.3msg move.prs.msg ‘The turtle is not moving’. hem ʔe(j)n=am joʃv-im sbj.3mpl neg.exist=poss.3mpl sit.prs-mpl ‘They do not sit; they are not sitting’.  Modern Hebrew 601 ʔe(j)n ha-ʦav zaz neg.exist def-turtle.msg move.prs.msg ‘The turtle is not moving’. k=a-regaʕ ʔe(j)n Sara levad at=def-moment neg.exist pn alone.adv ‘At the moment Sara is not alone’. Such usages are common in formal register. In colloquial speech, however, the particle lo typically negates the present tense (see earlier). The particle ʔe(j)n may also be used to negate the infinitive to express a prohibitive in certain restricted contexts and/or registers (example from the internet): ʔe(j)n lehaχnot ʔo laχsom maʕavar neg.exist park.inf or.conj block.inf passage.msg ‘no parking or blocking of passage’ It is also common for the negative lo to be used in similar contexts: lo lidroχ ʕal ha-deʃe neg step.inf on def-grass ‘Do not step on the grass!’ For the use of ʔe(j)n as an existential particle, which is common in all registers, see §5.2. There are also a number of other nominal negators used in Modern Hebrew with far more restricted usage. The particles bli conj and le-lo conj (< to.prep=neg) are both used to express the idea of ‘without’, whether before a nominal or verbal form: e.g., bli kesef without.conj money.msg ‘without money’, gamr-u bli leʃanot klum finish.pst-mpl with- out.conj change.inf anything ‘they finished without changing anything’ and le-lo hefsek without.conj (< to.prep=neg) ceasing.msg ‘without ceasing’. The particle bilti may negate adjectives and participles, roughly approximating English un- or in-/im-: e.g., bilti ʃavir neg breakable.adj.msg ‘unbreakable’ and bilti savir neg reasonable.adj.msg ‘improbable’. Finally, the prefixes ʔi- and ʔa- may be used to negate certain nouns and adjectives, though it may be better to regard the prefix ʔa- entirely as a borrowing rather than as a productive morpheme in Modern Hebrew: ʔi-havana neg-understanding.fsg ‘misunder- standing’, ʔi-raʦjonali neg-rational.adj.msg ‘irrational’, ʔi-zugi neg-even.adj.msg ‘odd (lit: ‘not even/dual’)’ and ʔa-politi neg-political.adj.msg (or perhaps merely ʔapoliti apo- litical.adj.msg) ‘apolitical’. 6 LEXICON The majority of the Modern Hebrew lexicon is inherited from earlier stages of the language (i.e., Biblical Hebrew, Rabbinic Hebrew, Medieval Hebrew), though a significant portion of the lexicon is first attested in Modern Hebrew (i.e., loan words, neologisms). Since the revival of Hebrew as a spoken language, the Academy of the Hebrew Language has played a significant role in innovating new words and disseminating them by means of the media. 602  Philip Zhakevich and Benjamin Kantor Nevertheless, in a typical Modern Hebrew text, most words can be traced back to Biblical Hebrew, though often with a further-developed semantic range. The high presence of Bib- lical Hebrew terms in Modern Hebrew texts should not be surprising, for these lexemes are often function words such as copulas, common verbs, pronouns, numbers, prepositions and adverbs. Moreover, the utilization of Hebrew as a literary language from the 3rd century ce onward preserved vocabulary from all strata of the language up until its revival as a spoken language in the late 19th century (Schwarzwald 2013c: 535, Sivan 1980: 27). Loan words, which are quite common, are a result of language contact and can be divided into two groups: (a) those that entered pre-modern layers of Hebrew and are pre- served in Modern Hebrew and (b) those that have entered Modern Hebrew since the Jewish Enlightenment of the 18th century. Examples of the former include ktav ‘writing’ (Biblical Hebrew < Aramaic), talmid ‘pupil’ (Biblical Hebrew < Aramaic < Akkadian), sfog ‘sponge’ (Rabbinic Hebrew < Greek), safsal ‘bench’ (Rabbinic Hebrew < Latin) and merkaz ‘center’ (Medieval Hebrew < Arabic) (Sáenz-Badillos 1993: 256, Bar-Asher 2013: 534–5, Kogan 2013: 528–32). Examples of the latter include kugel ‘noodle casserole’ (German), ʤuk ‘cockroach’ (Russian), tembel ‘fool’ (Turkish) and hepi-end ‘happy end’ (English). Arabic accounts for many borrowed food terms (e.g., χumus ‘hummus’, falafel ‘falafel’), slang expressions (wala ‘wow’, uχti ‘sis’) and obscenities. While loan words are generally not register-specific, those derived from Jewish languages (e.g., Yiddish, Ladino) and Pales- tinian Arabic tend to be relegated to a lower register (Schwarzwald 2013c: 538–9). On the other hand, loan words from Slavic languages or from English that fill a lexical void in Hebrew (e.g., norma ‘norm’, banana ‘banana’, bakterja ‘bacteria’) are used in all registers. 7 SAMPLE TEXT The text represents an online article found on a website providing medical advice. This particular article was found in a section dealing with children and youth. The article discusses the positive and negative effects of the internet on children and teenagers. The original article can be found here: www.drtal.co.il/%D7%90%D7%99%D7%A0%D7%9 8%D7%A8%D7%A0%D7%98. Line 1: barur ʃe=ʔi-ʔefʃar limnoaʕ mi=jelad-im legamre(j) ʔet clear.adj.msg comp=neg-possible.pred.adj prevent.inf from=child-mpl totally.adv obj Line 2: ha-ʃimuʃ b=a-ʔinternet, ʔaχ ribuj ha-katav-ot ʔodot def-usage in=def-internet.msg but.conj multiplicity.msg def-article.f-pl concerning.prep Line 3: pgiʕa b=a-jelad-im dereχ ha-medja ha-zo meʕorer damage.fsg in=def-child-mpl through.prep def-media.fsg def-dem.fsg arouse.prs.msg Line 4: χaʃaʃ gadol bekerev hor-im rab-im. fear.msg big.adj.msg among.prep parent-mpl many.adj-mpl  Modern Hebrew 603 Line 5: ʔaz ʔe(j)χ be=ʔemet ha-ʔinternet maʃpiaʕ ʕal jelad=e(j)nu, so.adv how.int in=truth.fsg def-internet.msg influence.prs.msg on.prep child.mpl=poss.1cpl Line 6: ve=beʔikar ma ʔanaχnu ke=hor-im jeχol-im laʕasot conj=in.essence.adv what.int sbj.1cpl as=parent-mpl be.able.prs-mpl do.inf Line 7: ʕal menat limnoaʕ ʔet haʃpaʕ-ot=av ha-ʃlili-jot? in.order.to.prep prevent.inf obj influence.f-pl=poss.3msg def-negative.adj-fpl Line 8: ne-nase laʕanot ʕal ʃeʔel-ot ʔelu. 1cpl-try.fut answer.inf on.prep question.f-pl dem.cpl Line 9: ʔim be=χol zot niʃʔar-u la=χem ʃeʔel-ot if.conj in=all dem.fsg remain.pst-pl to=poss.2mpl question.f-pl Line 10: nosaf-ot, hitlabtu-jot ʔo sfek-ot mumlaʦ lifnot additional.adj-fpl indecision.f-pl or.conj doubt.m-pl recommended.adj.msg turn.inf Line 11: le=ʔiʃ mikʦoaʕ ha-mumχe le=tipul be=jelad-im ve=bn-e(j) to=man.msg profession.msg def-expert.msg to=treatment.msg in=child-mpl conj=son-mpl.bnd Line 12: noʕar le=kabala-t ʕeʦa ʔo hadraχa-t hor-im b=a-nose. youth.msg to=reception-fsg.bnd counsel.fsg or guidance-fsg.bnd parent-mpl in=def-subject.msg Line 13: Lo tamid neχon-a ha-hanaχa ha-rovaχ-at neg always.adv correct.adj-fsg def-assumption.fsg def-widespread.adj-fsg Line 14: ʃe=jelad-im ha-meval-im b=a-ʔinternet hem davka jelad-im comp=child-mpl def-spend.prs-mpl in=def-internet.msg sbj.3mpl particularly.adv child-mpl Line 15: boded-im ve=lo mekubal-im. lonely.adj-mpl conj=neg accepted.adj-mpl Line 16: lehefeχ, peʕam-im rab-ot jelad-im ve=bn-e(j) noʕar on.the.contrary.adv time.f-pl many.adj-fpl child-mpl conj=son-mpl.bnd youth.msg 604  Philip Zhakevich and Benjamin Kantor Line 17: ha-peʕil-im b=a-ʔinternet hem gam peʕil-im meʔod def-active.adj-mpl in=def-internet.msg sbj.3mpl also.adv active.adj-mpl very.adv b=a-χevra, in=def-society.fsg Line 18: ve=ze paʃut ʕaruʦ nosaf b=o hem miʃtamʃ-im conj=dem.msg simply.adv channel.msg additional.adj.msg in=poss.3msg sbj.3mpl use.prs-mpl Line 19: l=a-peʕilut ha-χevrati-t ʃel=ahem ha-ʕanef-a mimela. to=def-activity.fsg def-activity.adj-fsg gen=poss.3mpl def-diverse.adj-fsg anyway.adv ‘1 It is clear that it is impossible to completely prevent children from 2 using the internet. However, the multiplicity of articles concerning 3 the damage caused to children through this medium does raise considerable 4 fear among many parents. 5 So how does the inter- net actually influence our children, 6 and what in essence can we as parents do 7 in order to prevent its negative influences? 8 We will attempt to answer these questions. 9 Neverthe- less, if you still have additional 10 questions, concerns, or doubts, we recommend that you turn to 11 a professional with expertise in treating children and 12 youth to receive counsel or parental guidance on the subject. 13 It is not always the case, as the common assumption would have it, that 14 children who spend time on the internet are particularly lonely 15 children or ‘not accepted’. 16 On the contrary, in many instances children and youth 17 who are active on the internet are also very active in society 18 and it (the internet) just so happens to be another channel that they use 19 for their already diverse social activity.’ NOTES 1 Note that in Ashkenazi Hebrew, /t/ > [t] and /θ/ > [s]. 2 A horizontal line above the letter (e.g., ‫) ֿב‬, known as rafe, was a mark developed by the Tiberian Masoretes to indicate the fricative pronunciation of the consonants ‫ב ג ד כ פ ת‬ as opposed to the plosive pronunciation of the consonants ‫ּב ּג ּד ּכ ּפ ּת‬, which was signi- fied with dagesh. It is used conventionally here for the diachronic discussion in light of its historical use, even though it is not used in Modern Hebrew orthography. 3 While it was held for some time that the Hebrew “emphatic” consonants were origi- nally pharyngealized, it is likely that they were actually glottalic ejectives in the earli- est stages of Hebrew. In the Tiberian Biblical Hebrew reading tradition of the Middle Ages, however, these consonants were pronounced with pharyngealization as in Ara- bic, probably due to Arabic influence. 4 Because ʕ is preserved in the speech of some speakers, we have decided to transcribe etymological *ʕ as ʕ throughout this chapter. Nevertheless, this is merely a transcrip- tion convention and one should keep in mind that what we have transcribed as ʕ is actually realized as ʔ or even Ø by a majority of speakers. Relevant phonological and morphological phenomena regarding etymological *ʕ are detailed explicitly in this chapter.  Modern Hebrew 605 5 The symbol ə is not necessarily meant to represent the phonetic realization of shewa, since various traditions pronounced shewa differently. For example, while in Tiberian Hebrew shewa was typically pronounced as [a], in the Palestinian tradition it was pro- nounced as either [e] or [a]. It is used here merely as a convention. 6 Elsewhere in this chapter, the optional offglided pronunciation of historical tsere is represented with a [j] in parentheses: i.e., [e(j)]. 7 In highly formal speech and among some elderly speakers, this rule does not apply to the 2p suffixes (-tem, -ten) on the past tense: e.g., [ktav-ˈtem]/[ktav-ˈten] write.pst-2mp/ write.pst-2fpl ‘you wrote’ (see §4.4.1). 8 While -ot is the most regular plural form, there is some variation in particular patterns. Feminine nouns that end in -ut have a plural of -ujot and feminine nouns that end in -it have a plural of -ijot. REFERENCES Agmon, Galit. “Negation of Adjectives.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 814–15. Leiden: Brill, 2013. Bar-Adon, Aaron. The Rise and Decline of a Dialect: A Study in the Revival of Modern Hebrew. The Hague: Mouton, 1975. Bar-Asher, Moshe. “On the Multiple Facets of Contemporary Hebrew.” Ha-ʕIvrit 58 (2010): 5–26. Bar-Asher, Moshe. “Lexicon: Rabbinic Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 532–5. Leiden: Brill, 2013. Barak, Dan. “Orthography: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 956–4. Leiden: Brill, 2013. Bat-El, Outi. “Denominal Verbs: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 704–9. Leiden: Brill, 2013. Berman, Ruth A. “Modern Hebrew.” In The Semitic Languages, edited by Robert Hetzron, 312–33. London: Routledge, 1997. Berman, Ruth A. “Noun Phrase.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 873–9. Leiden: Brill, 2013. Bolozky, Shmuel. “Israeli Hebrew Phonology.” In Phonologies of Asia and Africa, Vol. 1, edited by Alan S. Kaye, 287–311. Winona Lake, Indiana: Eisenbrauns, 1997. Bolozky, Shmuel. “Bgdkpt Consonants: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 262–8. Leiden: Brill, 2013a. Bolozky, Shmuel. “Imperative and Prohibitive: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 245–8. Leiden: Brill, 2013b. Bolozky, Shmuel. “Phonology: Israeli Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 113–22. Leiden: Brill, 2013c. Bolozky, Shmuel. “Pretonic Lengthening: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 229–30. Leiden: Brill, 2013d. Bolozky, Shmuel. “Resh: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 389–90. Leiden: Brill, 2013e. Bolozky, Shmuel. “Vowel Length: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 985. Leiden: Brill, 2013f. 606  Philip Zhakevich and Benjamin Kantor Boneh, Nora. “Aspect: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 205–18. Leiden: Brill, 2013a. Boneh, Nora. “Mood and Modality: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 693–703. Leiden: Brill, 2013b. Boneh, Nora. “Tense: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 742–51. Leiden: Brill, 2013c. Borochovsky, Esther. “The Hierarchy in Post-Nominal Modifiers.” Lešonenu 50 (1986): 103–18. Botwinik, Irena. “Phrasal Verbs: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 142–3. Leiden: Brill, 2013a. Botwinik, Irena. “Prepositions: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 212–19. Leiden: Brill, 2013b. Burstein, Ruth. “Interrogative Particles and Adverbs in Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 316–20. Leiden: Brill, 2013. Coffin, Edna Amir, and Shmuel Bolozky. A Reference Grammar of Modern Hebrew. Cambridge: Cambridge University Press, 2005. Cohen-Gross, Dalia. “Syllable Structure in Modern Hebrew (in Hebrew).” In A Tribute to Raphael Nir: Studies in Communication, Linguistics and Language Teaching (in Hebrew), edited by Ora (Rodrigue) Schwarzwald, Shoshana Blum-Kulka, and Elite Olshtein, 359–69. Jerusalem: Carmel, 2003. Cohen-Gross, Dalia. “Syllable Structure: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 676–9. Leiden: Brill, 2013. Cohen, Eran. “The Modern Hebrew Prepositional Relative Clause Strategy.” Studies in Language 40.4 (2016): 733–64. Cohen, Evan-Gary, and Adam Ussishkin. “Stress: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 625–8. Leiden: Brill, 2013. Daniels, Peter. “Alphabet, Origin of.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 87–95. Leiden: Brill, 2013. Danon, Gabi. “Syntactic Definiteness in the Grammar of Modern Hebrew.” Linguistics 39 (2001): 1071–116. Danon, Gabi. “Case: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 395–400. Leiden: Brill, 2013a. Danon, Gabi. “Copula.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 627–31. Leiden: Brill, 2013b. Danon, Gabi. “Definiteness: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 684–90. Leiden: Brill, 2013c. Doron, Edit. “Binyanim: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 364–71. Leiden: Brill, 2013. Doron, Edit and Irit Meir. “Construct State: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 581–9. Leiden: Brill, 2013. Edzard, Lutz. “Compound Nominals.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 516–21. Leiden: Brill, 2013a. Edzard, Lutz. “Genitive.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 44–7. Leiden: Brill, 2013b.  Modern Hebrew 607 Engel, Edna. “Script, History of Development.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 485–502. Leiden: Brill, 2013. Even-Shoshan, Abraham. Milon Even Shoshan: Renewed and Updated for the 2000’s. Edited by Moshe Azar et  al. Tel Aviv: Am Oved, Kineret, Zmora-Bitan, Dvir, and Yediot Aharonot, 2003. Faust, Noam. “Noun.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 864–9. Leiden: Brill, 2013 Fruchtman, Maya. Ha-Yaduaʕ Ve-Ha-Satum: Qategoryot Shel Yiduaʕ Ve-Tixum Ba-ʕivrit Ha-Yisreʔelit. Tel-Aviv: Papyrus, 1982. Fruchtman, Maya. “Indefinite Pronouns.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 257–60. Leiden: Brill, 2013. Giora, Rachel. “The Function of Topicalization at the Sentence and Discourse Level.” In Studies in Discourse Analysis, edited by Shoshana Blum-Kulka, Yishai Tobin, and Raphael Nir, 263–301. Jerusalem: The Hebrew University of Jerusalem, 1982. Glinert, Lewis. “Negative and Non-Assertive in Contemporary Hebrew.” Bulletin of the School of Oriental and African Studies 45 (1982): 434–70. Glinert, Lewis. The Grammar of Modern Hebrew. Cambridge: Cambridge University Press, 1989. Glinert, Lewis. “Conjunctions: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 566–9. Leiden: Brill, 2013a. Glinert, Lewis. “Negation: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 811–14. Leiden: Brill, 2013b. Goldenberg, Gideon. “Hebrew as a Living Semitic Language.” In Evolution and Renewal: Trends in the Development of the Hebrew Language, edited by Yehoshua Blau, 148–90. Jerusalem: Israeli Academy of Sciences and Humanities, 1996. Halevy, Rivka. “Deixis.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 693–7. Leiden: Brill, 2013a. Halevy, Rivka. “Syntax: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 707–22. Leiden: Brill, 2013b. Hatav, Galia. “Relative and Absolute Tense Interpretation in Modern Hebrew.” Hebrew Studies 51 (2010): 261–85. Huehnergard, John. “On the Etymology of the Hebrew Relative šε-.” In Biblical Hebrew in Its Northwest Semitic Setting: Typological and Historical Perspectives, edited by Steven E. Fassberg, and Avi Hurvitz, 103–25. Jerusalem/Winona Lake, Indiana: Magnes/Eisenbrauns, 2006. Ilani, Noga, Sigal Shlomo, and Dina Goldberg. “Word Order: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 1002–7. Leiden: Brill, 2013. Khan, Geoffrey. “Tiberian Reading Tradition.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 769–78. Leiden: Brill, 2013. Kogan, Leonid. “Lexicon: Biblical Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 528–32. Leiden: Brill, 2013. Kotek, Hadas. “Syntactic-Semantic Interface in Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 683–8. Leiden: Brill, 2013. Laks, Lior. “Actionality (Aktionsart): Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 28–31. Leiden: Brill, 2013. 608  Philip Zhakevich and Benjamin Kantor Meir, Irit. “Numerals: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 903–8. Leiden: Brill, 2013. Melčuk, Igor and Baruch Podolsky. “Stress in Modern Hebrew Nominal Inflection.” Theoretical Linguistics 22 (1996): 154–94. Mor, Galila. “Adverb.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 44–50. Leiden: Brill, 2013. Morag, Shelomo. “ʕal Lashon Ve-ʔestetiqa Ve-Ha-ʕivrit Bat-Zmanenu.” Molad (n.s.) 8 (1980): 81–90. Oren, Mikhal. “Adverbial.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 50–3. Leiden: Brill, 2013. Ornan, Uzzi. The Simple Sentence. Jerusalem: Akademon, 1979. Ravid, Dorit and Yitzhaq Shlesinger. “Modern Hebrew Adverbials: Between Syntactic Class and Lexical Category.” In Between Grammar and Lexicon, edited by Ellen Contini-Morava and Yishai Tobin, 333–51. Amsterdam/Philadelphia: John Benjamins, 2000. Reshef, Yael. “Revival of Hebrew: Grammatical Structure and Lexicon.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 397–405. Leiden: Brill, 2013a. Reshef, Yael. “Revival of Hebrew: Sociolinguistic Dimension.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 408–15. Leiden: Brill, 2013b. Rubin, Aaron D. “Indefinite Article.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 256. Leiden: Brill, 2013. Sadan, Tsvi. “Verbal System: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 925–9. Leiden: Brill, 2013. Sáenz-Badillos, Angel. A History of the Hebrew Language. Cambridge: Cambridge University Press, 1993. Schwarzwald, Ora. “Modern Hebrew.” In The Semitic Languages, edited by Stefan Weninger, 523–36. Berlin: De Gruyter Mouton, 2011. Schwarzwald, Ora. “Defective Verbs.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 673–8. Leiden: Brill, 2013a. Schwarzwald, Ora. “Diphthongs: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 739. Leiden: Brill, 2013b. Schwarzwald, Ora. “Lexicon: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 535–42. Leiden: Brill, 2013c. Schwarzwald, Ora. “Mishqal.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 646–50. Leiden: Brill, 2013d. Schwarzwald, Ora. “Modern Hebrew: Language Varieties.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 668–82. Leiden: Brill, 2013e. Shatil, Nimrod. “Guttural Consonants: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 2, edited by Geoffrey Khan, 169–72. Leiden: Brill, 2013. Shlesinger, Yitzhak. “Borrowing in Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 1, edited by Geoffrey Khan, 375–81. Leiden: Brill, 2013. Sivan, Reuben. The Revival of the Hebrew Language. Jerusalem: Rubinstein, 1980.  Modern Hebrew 609 Steiner, Richard C. Affricated Ṣade in the Semitic Languages. New York: The American Academy for Jewish Research, 1982. Uziel-Karl, Sigal. “Predicate.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 205–8. Leiden: Brill, 2013. Wexler, Paul. The Schizoid Nature of Modern Hebrew: A Slavic Language in Search of a Semitic Past. Wiesbaden: Otto Harrassowitz, 1990. Wintner, Shuly. “Definiteness in the Hebrew Noun Phrase.” Journal of Linguistics 36 (2000): 319–63. Zewi, Tamar. “Relative Clause: Modern Hebrew.” In Encyclopedia of Hebrew Language and Linguistics, Vol. 3, edited by Geoffrey Khan, 359–63. Leiden: Brill, 2013. Zuckerman, Ghil’ad. Israeli, a Beautiful Language. Tel Avid: Am Oved, 2008. General Blau, Joshua. The Renaissance of Modern Hebrew and Modern Standard Arabic. Berkeley: University of California Press, 1981. Izre’el, Shlomo. “The Emergence of Spoken Israeli Hebrew.” In Corpus Linguistics and Modern Hebrew: Towards the Compilation of The Corpus of Spoken Israeli Hebrew, edited by Benjamin H. Hary, 85–104. Tel Aviv: Tel Aviv University, The Chaim Rosenberg School of Jewish Studies, 2003. Khan, Geoffrey, (ed.). Encyclopedia of Hebrew Language and Linguistics. Leiden: Brill, 2013. Netzer, Nisan. ʕivrit be-Jins (in Hebrew). Beer-Sheva: Ben Gurion University Press, 2007. Schwarzwald, Ora. Modern Hebrew. Munich: Lincom Europa, 2001. Schwarzwald, Ora. “Modern Hebrew.” In The Semitic Languages, edited by Stefan Weninger, 523–36. Berlin: De Gruyter Mouton, 2011. Yardeni, Ada. The Book of Hebrew Script. Jerusalem: Carta, 1997. Grammars Arad, Maya. Roots and Patterns: Hebrew Morphosyntax. Dordrecht: Springer, 2005. Bar-Asher, Moshe. “On the Multiple Facets of Contemporary Hebrew (in Hebrew).” Ha-ʕivrit 58 (2010): 5–26. Bolozky, Shmuel. “Israeli Hebrew Phonology.” In Phonologies of Asia and Africa, Vol. 1, edited by Alan S. Kaye, 287–311. Winona Lake, Indiana: Eisenbrauns, 1997. Coffin, Edna Amir and Shmuel Bolozky. A Reference Grammar of Modern Hebrew. Cambridge: Cambridge University Press, 2005. Glinert, Lewis. Modern Hebrew: An Essential Grammar. New York/London: Routledge,  2016. Sharvit, Yael. “Aspects of the Semantics of Tense in Modern Hebrew (in Hebrew).” In Theoretical Hebrew Linguistics, 139–62. Jerusalem: Magnes, 2008. Textbooks Etzion, Giore. The Routledge Introductory Course in Modern Hebrew. New York: Routledge, 2009. Ringwald, Vardit, Bonit Porath, Yaron Peleg, Esther Shorr, and Sara Hascal. Brandeis Modern Hebrew. Waltham, MA: Brandeis University Press, 2015. 610  Philip Zhakevich and Benjamin Kantor Dictionaries Doniach, Nakdimon S., and Ahuvia Kahane (eds.). The Oxford English – Hebrew Dictionary. Oxford: Oxford University Press, 1998. Even-Shoshan, Abraham. Milon Even Shoshan: Renewed and Updated for the 2000’s (in Hebrew), edited by Moshe Azar et al. Tel Aviv: Am Oved, Kineret, Zmora-Bitan, Dvir, and Yediot Aharonot, 2003. Levy, Ya’acov (ed.). Oxford English-Hebrew, Hebrew – English Dictionary. Jerusalem: Kernermann/Lonnie Kahn, 1995.