The lead section of this article may need to be rewritten. (September 2009) (Learn how and when to remove this template message)
On the other hand, Modern Hebrew grammar is also fusional synthetic:inflection plays a role in the formation of verbs and nouns (using non-concatenative discontinuous morphemes realised by vowel infixation) and the declension of prepositions (i.e. with pronominal suffixes).
Examples of Hebrew are represented using the International Phonetic Alphabet (IPA) as well as native script. Although most speakers collapse the phonemes /?, ?/ into /?, ?/, the distinction is maintained by a limited number of speakers and will therefore be indicated here for maximum coverage. In the transcriptions, /r/ is used for the rhotic, which is more commonly a lax voiced uvular approximant.
Word order in Modern Hebrew is somewhat similar to that in English: as opposed to Biblical Hebrew, where the word order is Verb-Subject-Object, the usual word order in Modern Hebrew is Subject-Verb-Object. Thus, if there is no case marking, one can resort to the word order. Modern Hebrew is characterized by an asymmetry between definite Objects and indefinite Objects. There is an accusative marker, et, only before a definite Object (mostly a definite noun or personal name). Et-ha is currently undergoing fusion and reduction to become ta. Consider ten li et ha-séfer "give:2ndPerson.Masculine.Singular.Imperative to-me ACCUSATIVE the-book" (i.e. "Give me the book"), where et, albeit syntactically a case-marker, is a preposition and ha is a definite article. This sentence is realised phonetically as ten li ta-séfer.
In sentences where the predicate is a verb, the word order is usually subject-verb-object (SVO), as in English. However, word order can change in the following instances:
Generally, Hebrew marks every noun in a sentence with some sort of preposition, with the exception of subjects and semantically indefinite direct objects. Unlike English, indirect objects require prepositions (Hebrew "? ?" /hu na'tan li ?et ha-ka'dur/ (literally "he gave to-me direct-object-marker the ball) in contrast to English "He gave me the ball") and semantically definite direct objects are introduced by the preposition /et/ (Hebrew "? ?" /hu na'tan li ?et ha-ka'dur/ (literally "he gave to-me direct-object-marker the ball) in contrast to English "He gave me the ball").
Hebrew also produces sentences where the predicate is not a finite verb. A sentence of this type is called ? ? /mi?'pat ?ema'ni/, a nominal sentence. These sentences contain a subject, a non-verbal predicate, and an optional copula. Types of copulae include:
While /lo/ ("not") precedes the copula in the past and future tenses, it follows the copula (a subject pronoun) in the present tense.
Sentences are generally divided into three types:
A simple sentence is a sentence that contains one subject, one verb, and optional objects. As the name implies, it is the simplest type of sentence.
Two or more sentences that do not share common parts and can be separated by comma are called /mi?'pat me?u'bar/, a compound sentence. In many cases, the second sentence uses a pronoun that stands for the other's subject; they are generally interconnected. The two sentences are linked with a coordinating conjunction ( ? /mi'lat ?i'bur/). The conjunction is a stand-alone word that serves as a connection between both parts of the sentence, belonging to neither part.
Like English, Hebrew allows clauses, ? /psuki'jot/ (sing. ? /psu'kit/), to serve as parts of a sentence. A sentence containing a subordinate clause is called ? ? /mi?'pat mur'kav/, or a complex sentence. Subordinate clauses almost always begin with the subordinating conjunction -? /?e-/ (usually that), which attaches as a prefix to the word that follows it. For example, in the sentence ? /'josi ?o'mer ?e-'hu ?o'?el/ (Yossi says that he is eating), the subordinate clause ? /?e-'hu ?o'?el/ (that he is eating) serves as the direct object of the verb /?o'mer/ (says). Unlike English, Hebrew does not have a large number of subordinating conjunctions; rather, subordinate clauses almost always act as nouns and can be introduced by prepositions in order to serve as adverbs. For example, the English As I said, there's nothing we can do in Hebrew is ?, ? /kfi ?e-?a'marti, ?en ma la?a'sot/ (literally As that-I-said, there-isn't what to-do).
That said, relative clauses, which act as adjectives, are also formed using - /?e-/. For example, English Yosi sees the man who is eating apples is in Hebrew ? /'josi ro'?e ?et ha-'?i? ?e-?o'?el tapu'?im/ (literally Yosi sees [et] the-man that-eats apples). In this use ? /?e-/ sometimes acts as a relativizer rather than as a relative pronoun; that is, sometimes the pronoun remains behind in the clause ? /hi maki'ra ?et ha-'?i? ?e-di'barti ?a'lav/, which translates to She knows the man I talked about, literally means She knows [et] the-man that-I-talked about him. This is because in Hebrew, a preposition (in this case /?al/) cannot appear without its object, so the him - (/-av/) could not be dropped. However, some sentences, such as the above example, can be written both with relativizers and with relative pronouns. The sentence can also be rearranged into ? ? /hi maki'ra ?et ha-'?i? ?a'lav di'barti/, literally She knows [et] the-man about him I-talked., and translates into the same meaning. In that example, the preposition and its object /?a'lav/ also act as a relative pronoun, without use of - /?e-/.
A sentence may lack a determinate subject, then it is called ? /mi?'pat sta'mi/, an indefinite or impersonal sentence. These are used in order to put emphasis on the action, and not on the agent of the action. Usually the verb is of the 3rd person plural form.
When a sentence contains multiple parts of the same grammatical function and relate to the same part of the sentence, they are called collective parts. They are usually separated with the preposition - /ve-/ (and), and if there are more than two, they are separated with commas while the last pair with the preposition, as in English. Collective parts can have any grammatical function in the sentence, for instance:
When a collective part is preceded by a preposition, the preposition must be copied onto all parts of the collective.
The Hebrew verb (? /'po?al/) serves essentially the same functions as the English verb, but is constructed very differently. Hebrew verbs have much more internal structure. Every Hebrew verb is formed by casting a three- or four-consonant root (? /'?ore?/) into one of seven derived stems called /binja'nim/ (, meaning buildings or constructions; the singular is /bin'jan/, written henceforth as binyan). Most roots can be cast into more than one binyan, meaning more than one verb can be formed from a typical root. When this is the case, the different verbs are usually related in meaning, typically differing in voice, valency, semantic intensity, aspect, or a combination of these features. The "concept" of the Hebrew verb's meaning is defined by the identity of the triliteral root. The "concept" of the Hebrew verb assumes verbal meaning by taking on vowel-structure as dictated by the binyan's rules.
Each binyan has a certain pattern of conjugation and verbs in the same binyan are conjugated similarly. Conjugation patterns within a binyan alter somewhat depending on certain phonological qualities of the verb's root; the alterations (called [?izra], meaning "form") are defined by the presence of certain letters composing the root. For example, three-letter roots (triliterals) whose second letter is ? /vav/ or ? /jud/ are so-called hollow or weak roots, losing their second letter in binyan /hif'?il/, in /huf'?al/, and in much of /pa?al/. The feature of being conjugated differently because the second root-letter is ? or ? is an example of a gizra. These verbs are not strictly irregular verbs, because all Hebrew verbs that possess the same feature of the gizra are conjugated in accordance with the gizra's particular set of rules.
Every verb has a past tense, a present tense, and a future tense, with the present tense doubling as a present participle. Other forms also exist for certain verbs: verbs in five of the binyanim have an imperative mood and an infinitive, verbs in four of the binyanim have gerunds, and verbs in one of the binyanim have a past participle. Finally, a very small number of fixed expressions include verbs in the jussive mood, which is essentially an extension of the imperative into the third person. Except for the infinitive and gerund, these forms are conjugated to reflect the number (singular or plural), person (first, second, or third) and gender (masculine or feminine) of its subject, depending on the form. Modern Hebrew also has an analytic conditional~past-habitual mood expressed with the auxiliary haya.
In listings such as dictionaries, Hebrew verbs are sorted by their third-person masculine singular past tense form. This differs from English verbs, which are identified by their infinitives. (Nonetheless, the Hebrew term for infinitive is shem po?al, which means verb name.) Further, each of the seven binyanim is identified by the third-person masculine singular past tense form of the root ?-?-? (P-?-L, meaning doing, action, etc.) cast into that binyan: /'pa?al/, /nif'?al/, /pi'?el/, /pu'?al/, /hif'?il/, /huf'?al/, and /hitpa'?el/.
Binyan pa?al, also called binyan or qal /qal/ (light), is the most common binyan. Pa?al verbs are in the active voice, and can be either transitive or intransitive. This means that they may or may not take direct objects. Pa?al verbs are never formed from four-letter roots.
Binyan pa?al is the only binyan in which a given root can have both an active and a passive participle. For example, /ra't?suj/ (desirable) is the passive participle of /ra't?sa/ (want).
Binyan pa?al has the most diverse number of gzarot (pl. of gizra), and the small number of Hebrew verbs that are strictly irregular (about six to ten) are generally considered to be part of the pa'al binyan, as they have some conjugation features similar to pa?al.
Verbs in binyan nifal are always intransitive, but beyond that there is little restriction on their range of meanings.
The nifal is the passive-voice counterpart of paal. In principle, any transitive paal verb can be rendered passive by taking its root and casting it into nifal. Nonetheless, this is not nif?al's main use, as the passive voice is fairly rare in ordinary Modern Hebrew.
More commonly, it is paal's middle- or reflexive-voice counterpart. Ergative verbs in English often translate into Hebrew as a paal-nifal pair. For example, English he broke the plate corresponds to Hebrew ? /hu ?a'var et ha-t?sa'la?at/, using paa'al; but English the plate broke corresponds to Hebrew /ha-t?sa'la?at ni?'bera/, using nifal. The difference is that in the first case, there is an agent doing the breaking (active), while in the second case, the agent is ignored (although the object is acted upon; passive). (Nonetheless, as in English, it can still be made clear that there was an ultimate agent ? /hu hi'pil ?et ha-t?sa'la?at ve-hi ni?'bera/, he dropped the plate and it broke, uses nif'al.) Other examples of this kind include /pa'ta?// /nif'ta?/ (to open, transitive/intransitive) and /?a'mar//? /ni?'mar/ (to end, transitive/intransitive).
Other relationships between a paa'al verb and its nifa'al counterpart can exist as well. One example is /za'?ar/ and /niz'kar/: both mean to remember, but the latter implies that one had previously forgotten, rather like English to suddenly remember. Another is ? /pa'?a?/ and /nif'?a?/: both mean to meet, but the latter implies an intentional meeting, while the former often means an accidental meeting.
Finally, sometimes a nifal verb has no pa'al counterpart, or at least is much more common than its pa?al counterpart; /nid'baq/ (to stick, intransitive) is a fairly common verb, but /da'vak/ (to cling) is all but non-existent by comparison. (Indeed, /nid'bak/'s transitive counterpart is /hid'bik/, of binyan hif?il; see below.)
Like pa'al verbs, nifal verbs are never formed from four-letter roots.
Nifal verbs, unlike verbs in the other passive binyanim (pua'al and hufa'al, described below), do have gerunds, infinitives and imperatives.
Binyan pi'el, like binyan pa'al, consists of transitive and intransitive verbs in the active voice, though there is perhaps a greater tendency for pi?el verbs to be transitive.
Most roots with a pa'al verb do not have a pi?el verb, and vice versa, but even so, there are many roots that do have both. Sometimes the pi'el verb is a more intense version of the pa?al verb; for example, /ki'pet?s/ (to spring) is a more intense version of /ka'fat?s/ (to jump), and ? /?i'ber/ (to smash, to shatter, transitive) is a more intense version of /?a'var/ (to break, transitive). In other cases, a pi?el verb acts as a causative counterpart to the pa'al verb with the same root; for example, /li'med/ (to teach) is essentially the causative of /la'mad/ (to learn). And in yet other cases, the nature of the relationship is less obvious; for example, /si'per/ means to tell / to narrate or to cut hair, while /sa'far/ means to count, and /pi'te.a?/ means to develop (transitive verb), while /pa'ta?/ means to open (transitive verb).
Binyan pu?al is the passive-voice counterpart of binyan pi?el. Unlike binyan nif?al, it is used only for the passive voice. It is therefore not very commonly used in ordinary speech, except that the present participles of a number of pu?al verbs are used as ordinary adjectives /mevul'bal/ means mixed-up (from /bul'bal/, the passive of /bil'bel/, to confuse), /meun'jan/ means interested, /mefur'sam/ means famous (from /pur'sam/, the passive of /pir'sem/, to publicize), and so on. Indeed, the same is true of many pi?el verbs, including the pi?el counterparts of two of the above examples /meval'bel/, confusing, and /me?an'jen/, interesting. The difference is that pi?el verbs are also frequently used as verbs, whereas pu?al is much less common.
Pu?al verbs do not have gerunds, imperatives, or infinitives.
Binyan hif?il is another active binyan. Hif?il verbs are often causative counterparts of verbs in other binyanim; examples include /hi?'tiv/ (to dictate; the causative of /ka'tav/, to write), /hid'lik/ (to turn on (a light), transitive; the causative of ? /nid'lak/, (for a light) to turn on, intransitive), and /hir'?im/ (to impress; the causative of /hitra'?em/, to be impressed). Nonetheless, not all are causatives of other verbs; for example, /hiv'tia?/ (to promise).
Binyan huf'al is much like binyan pu'al, except that it corresponds to hif'il instead of to pi'el. Like pu'al, it is not commonly used in ordinary speech, except in present participles that have become adjectives, such as /mu'kar/ (familiar, from /hu'kar/, the passive of ? /hi'kir/, to know (a person)) and ? /mu?'zam/ (excessive, from /hu?'zam/, the passive of /hi?'zim/, to exaggerate). Like pu?al verbs, huf?al verbs do not have gerunds, imperatives, or infinitives.
Binyan hitpa'el is rather like binyan nif'al, in that all hitpa'el verbs are intransitive, and most have a reflexive sense. Indeed, many hitpa'el verbs are reflexive counterparts to other verbs with the same root; for example, /hitra'?et?s/ (to wash oneself) is the reflexive of /ra'?at?s/ (to wash, transitive), and /hit?a'lea?/ (to shave oneself, i.e. to shave, intransitive) is the reflexive of /?i'lea?/ (to shave, transitive). Some hitpa?el verbs are a combination of causative and reflexive; for example, /hista'per/ (to get one's hair cut) is the causative reflexive of /si'per/ (to cut (hair)), and ? /hit?sta'lem/ (to get one's picture taken) is the causative reflexive of /t?si'lem/ (to take a picture (of someone or something)).
Hitpa'el verbs can also be reciprocal; for example, /hitka'tev/ (to write to each other, i.e. to correspond) is the reciprocal of /ka'tav/ (to write).
In all of the above uses, the hitpa'el verb contrasts with a pu'al or huf'al verb in two ways: firstly, the subject of the hitpa'el verb is generally either performing the action, or at least complicit in it, whereas the subject of the pu'al or huf'al verb is generally not; and secondly, pu'al and huf'al verbs often convey a sense of completeness, which hitpa'el verbs generally do not. So whereas the sentence /a'ni met?su'lam/ (I am photographed, using pu'al) means something like there exists a photo of me, implying that the photo already exists, and not specifying whether the speaker caused the photo to be taken, the sentence ? /a'ni mit?sta'lem/ (I am photographed, using hitpa'el) means something like I'm having my picture taken, implying that the picture does not exist yet, and that the speaker is causing the picture to be taken.
In other cases, hitpa'el verbs are ordinary intransitive verbs; for example, /hitna'he?/ (to behave), structurally is the reciprocal of /na'ha?/ (to act), as in ? /ne'hag be-?o?'ma/ (act wisely). However, it is used sparsely, only in sayings as such, and the more common meaning of naha? is to drive; for that meaning, /hitna'he?/ is not a reciprocal form, but a separate verb in effect. For example: in talking about a car that drives itself, one would say /me?o'nit ?e-no'he?et ?et ?at?s'mah/ (a car that drives itself, using nahag), not /me?o'nit ?e-mitna'he?et/ (a car that behaves, using hitnaheg).
The Hebrew noun (? /?em ?et?sem/) is inflected for number and state, but not for case and therefore Hebrew nominal structure is normally not considered to be strictly declensional. Nouns are generally related to verbs (by shared roots), but their formation is not as systematic, often due to loanwords from foreign languages. Hebrew nouns are also inflected for definiteness by application of the prefix (ha) before the given noun. Semantically, the prefix "ha" corresponds roughly to the English word "the".
Every noun in Hebrew has a gender, either masculine or feminine (or both); for example, /'sefer/ (book) is masculine, while /'delet/ (door) is feminine. There is no strict system of formal gender, but there is a tendency for nouns ending in ? (/-t/) or ? (usually /-a/) to be feminine and for nouns ending in other letters to be masculine. There is a very strong tendency toward natural gender for nouns referring to people and some animals. Such nouns generally come in pairs, one masculine and one feminine; for example, /i?/ means man and ? /i'?a/ means woman. (When discussing mixed-sex groups, the plural of the masculine noun is used.)
Hebrew nouns are inflected for grammatical number; as in English, count nouns have a singular form for referring to one object and a plural form for referring to more than one. Unlike in English, some count nouns also have separate dual forms, for referring to two objects; see below.
Masculine nouns generally form their plural by adding the suffix /-im/:
The addition of the extra syllable usually causes the vowel in the first syllable to shorten if it is Kamatz:
Many common two-syllable masculine nouns accented on the penultimate syllable (often called segolates, because many (but not all) of them have the vowel /se'?ol/ (/-e-/) in the last syllable), undergo more drastic characteristic vowel changes in the plural:
Feminine nouns ending in /-a/ or /-at/ generally drop this ending and add /-ot/, usually without any vowel changes:
Nouns ending in /-e-et/ also replace this ending with /-ot/, with an /-e-/ in the preceding syllable usually changing to /-a-/:
Nouns ending in /-ut/ and /-it/ replace these endings with /-ujot/ and /-ijot/, respectively:
A large number of masculine nouns take the usually feminine ending /-ot/ in the plural:
A small number of feminine nouns take the usually masculine ending /-im/:
Many plurals are completely irregular:
Hebrew also has a dual number, expressed in the ending /-ajim/, but even in ancient times its use was very restricted. In modern times, it is usually used in expressions of time and number, or items that are inherently dual. These nouns have plurals as well, which are used for numbers higher than two, for example:
|/'pa?am a'?at/ (once)||? /pa?a'majim/ (twice)||/?a'losh pe?a'mim/ (thrice)|
|/?a'vua? e'?ad/ (one week)||/?vu'?ajim/ (two weeks)||? ? /?lo'?a ?avu'?ot/ (three weeks)|
|/'me?a/ (one hundred)||/ma?'tajim/ (two hundred)||/'?losh me'?ot/ (three hundred)|
The dual is also used for some body parts, for instance:
In this case, even if there are more than two, the dual is still used, for instance /le'?elev je? '?arba? ra?'lajim/ ("a dog has four legs").
The dual is also used for certain objects that are "inherently" dual. These words have no singular, for instance /mi?ka'fajim/ (eyeglasses) and /mispa'rajim/ (scissors). As in the English "two pairs of pants", the plural of these words uses the word ? /zu?/ (pair), e.g. /?ne zu'?ot mispa'rajim/ ("two pairs-of scissors-DUAL"). Similarly, the dual can be found in some place names, such as the city /giv?a'tajim/ (Twin Peaks, referring to the two hills of the landscape on which the city is built).
In Hebrew, as in English, a noun can modify another noun. This is achieved by placing the modifier immediately after what it modifies, in a construction called /smi'?ut/. The noun being modified appears in its construct form, or status constructus. For most nouns, the construct form is derived fairly easily from the normal (indefinite) form:
There are many words (usually ancient ones) that have changes in vocalization in the construct form. For example, the construct form of /'bajit/ (house) is /bet/.
In addition, the definite article is never placed on the first noun (the one in the construct form).
However, this rule is not always adhered to in informal or colloquial speech; one finds, for example, /ha-'o?e? din/ (literally the law organiser, i.e. lawyer).
Possession is generally indicated using the preposition /?el/, of or belonging to:
In literary style, nouns are inflected to show possession through noun declension; a personal suffix is added to the construct form of the noun (discussed above). So, ? /sif're/ (books of) can be inflected to form ? /sfa'raj/ (my books), /sfa're)?a/ (your books, singular masculine form), ? /sfa'renu/ (our books), and so forth, while ? /di'rat/ (apartment of) gives /dira'ti/ (my apartment), ? /dirat'?a/ (your apartment; singular masculine form), /dira'tenu/ (our apartment), etc.
While the use of these forms is mostly restricted to formal and literary speech, they are in regular use in some colloquial phrases, such as ? ? /ma ?lom'?a?/ (literally "what peace-of-you?", i.e. "what is your peace?", i.e. "how are you?", singular masculine form) or /leda?a'ti/ (in my opinion/according to my knowledge).
In addition, the inflected possessive is commonly used for terms of kinship; for instance, /bni/ (my son), ? /bi'tam/ (their daughter), and /i?'to/ (his wife) are preferred to ? /ha-'ben ?e'li/, /ha-'bat ?elahem/, and ? /ha-?i'?a ?e'lo/. However, usage differs for different registers and sociolects: In general, the colloquial will use more analytic constructs in place of noun declensions.
In the same way that Hebrew verbs are conjugated by applying various prefixes, suffixes and internal vowel combinations, Hebrew nouns can be formed by applying various "meters" (Hebrew /mi?ka'lim/) and suffixes to the same roots. Gerunds are one example (see above).
Many abstract nouns are derived from another noun, or from a verb (usually one in binyan hitpa?el) using the suffix /-ut/:
The ? /kat'lan/ meter applied to a root, and the /-an/ suffix applied to a noun, indicate an agent or job:
The suffix /-on/ usually denotes a diminutive:
Though occasionally this same suffix can denote an augmentative:
Repeating the last two letters of a noun or adjective can also denote a diminutive:
The /ka'telet/ meter commonly used to name diseases:
However, it can have various different meanings as well:
New nouns are also often formed by the combination of two existing stems:
/ram'zor/ uses more strictly recent compound conventions, as the ? aleph (today usually silent but historically very specifically a glottal stop) is dropped entirely from spelling and pronunciation of the compound.
Some nouns use a combination of methods of derivation:
In Hebrew, an adjective (? /?em toar/) comes after the noun and agrees in gender, number, and definiteness with the noun which it modifies:
Adjectives ending in -i have slightly different forms:
Masculine nouns that take the feminine plural ending /-ot/ still take masculine plural adjectives, e.g. /meko'mot ja'fim/ (beautiful places). The reverse goes for feminine plural nouns ending in /-im/, e.g. ? /mi'lim ?aru'kot/ (long words).
Many adjectives, like segolate nouns, change their vowel structure in the feminine and plural.
In Hebrew, unlike in English, each attributive adjective follows the noun and takes the definite article if it modifies a definite noun (either a proper noun, or a definite common noun):
The case of a proper noun highlights the fact that all Hebrew adjectives can be interpreted as appositive nouns. For example, contrast the following:
Many adjectives in Hebrew are derived from the present tense of verbs. These adjectives are inflected the same way as the verbs they are derived from:
The Hebrew term for adverb is /'to?ar ha-'po?al/.
Hebrew forms adverbs in several different ways.
Some adjectives have corresponding one-word adverbs. In many cases, the adverb is simply the adjective's masculine singular form:
In other cases, the adverb has a distinct form:
In some cases, an adverb is derived from an adjective using its singular feminine form or (mostly in poetic or archaic usage) its plural feminine form:
Most adjectives, however, do not have corresponding one-word adverbs; rather, they have corresponding adverb phrases, formed using one of the following approaches:
The use of one of these methods does not necessarily preclude the use of the others; for example, slowly may be either /le'?at/ (a one-word adverb), ? /be-?iti'jut/ (literally "in slowness", a somewhat more elegant way of expressing the same thing) or /be'?ofen ?i'ti/ ("in slow fashion"), as mentioned above.
Finally, as in English, there are various adverbs that do not have corresponding adjectives at all:
Like English, Hebrew is primarily a prepositional language, with a large number of prepositions. Several of Hebrew's most common prepositions, however, unlike any in English, are prefixes rather than separate words; for example, English in the room is Hebrew /ba-'?eder/.
The preposition /?et/ plays an important role in Hebrew grammar. Its most common use is to introduce a direct object; for example, English I see the book is in Hebrew /?a'ni ro'?e ?et ha-'sefer/ (literally I see /?et/ the-book). However, /?et/ is used only with semantically definite direct objects, such as nouns with the, proper nouns, and personal pronouns; with semantically indefinite direct objects, it is simply omitted? ?ani ro?e sefer (I see a book) does not use /?et/. This has no direct translation into English, and is best described as an object particle -- that is, it denotes that the word it precedes is the direct object of the verb.
Finally, /?et/ has a number of special uses; for example, when the adjective ? /t?sa'ri?/ (in need (of)) takes a definite noun complement, it uses the preposition /?et/? ? /ha'jiti t?sa'ri? ?et ze/ (literally I-was in-need-of /?et/ this, i.e. I needed this). Here as elsewhere, the /?et/ is dropped with an indefinite complement? /ha'ju t?sri'?im jo'ter/ (literally they-were in-need-of more, i.e. they needed more). This is perhaps related to the verb-like fashion in which the adjective is used.
In Biblical Hebrew, there is possibly another use of /?et/. Waltke and O'Connor (pp. 177-178) make the point: "...(1) ...sign of the accusative ... (2) More recent grammarians regard it as a marker of emphasis used most often with definite nouns in the accusative role. The apparent occurrences with the nominative are most problematic ... AM Wilson late in the nineteenth century concluded from his exhaustive study of all the occurrences of the debated particle that it had an intensive or reflexive force in some of its occurrences. Many grammarians have followed his lead. (reference lists studies of 1955, 1964, 1964, 1973, 1965, 1909, 1976.) On such a view, /?et/ is a weakened emphatic particle corresponding to the English pronoun 'self' ... It resembles Greek 'autos' and Latin 'ipse' both sometimes used for emphasis, and like them it can be omitted from the text, without obscuring the grammar. This explanation of the particle's meaning harmonizes well with the facts that the particle is used in Mishnaic Hebrew as a demonstrative and is found almost exclusively with determinate nouns."
There is a form called the verbal pronominal suffix, in which a direct object can be rendered as an additional suffix onto the verb. This form allows for a high degree of word economy, as the single fully conjugated verb expresses the verb, its voice, its subject, its object, and its tense.
In modern usage, the verbal pronominal suffixes are rarely used, in favor of expression of direct objects as the inflected form of the separate word ?et. It is used more commonly in biblical and poetic Hebrew (for instance, in prayers).
Indirect objects are objects requiring a preposition other than /?et/. The preposition used depends on the verb, and these can be very different from the one used in English. A good dictionary is required to look these up. In the case of definite indirect objects, the preposition will replace /?et/.
The Hebrew grammar distinguishes between various kinds of indirect objects, according to what they specify. Thus, there is a division between objects for time ? (/te'?ur zman/), objects for place ? (/te?ur ma'kom/), objects for reason ? (/te'?ur si'ba/) and many others.
In Hebrew, there are no distinct prepositional pronouns; if the object of a preposition is a pronoun, but the preposition contracts with the object, and the inflected preposition thus formed can be considered the indirect object of the sentence.
As mentioned above, the direct object is often rendered with the word /?et/. /?et/ is excluded only when the direct object is a non-definite noun.