Arabic verbs (فِعْل fiʿl; pl. أَفْعَال afʿāl), like the verbs in other Semitic languages, and the entire vocabulary in those languages, are based on a set of two to five (but usually three) consonants called a root (triliteral or quadriliteral according to the number of consonants). The root communicates the basic meaning of the verb, e.g. ك-ت-ب k-t-b 'write', ق-ر-ء q-r-ʾ 'read', ء-ك-ل ʾ-k-l 'eat'. Changes to the vowels in between the consonants, along with prefixes or suffixes, specify grammatical functions such as person, gender, number, tense, mood, and voice.

Various categories are marked on verbs:

• Three tenses (present, past; future tense is indicated by the prefix sa- or the particle sawfa and the present tense).
• Two voices (active, passive)
• Two genders (masculine, feminine)
• Three persons (first, second, third)
• Three numbers (singular, dual, plural)
• Six moods in the non-past only (indicative, subjunctive, jussive, imperative, and short and long energetics)
• Nineteen forms, the derivational systems indicating derivative concepts such as intensive, causative, reciprocal, reflexive, frequentative etc. For each form, there is also an active and a passive participle (both adjectives, declined through the full paradigm of gender, number, case and state) and a verbal noun (declined for case; also, when lexicalized, may be declined for number).

Weakness is an inherent property of a given verb determined by the particular consonants of the verb root (corresponding to a verb conjugation in Classical Latin and other European languages), with five main types of weakness and two or three subtypes of each type.

Arabic grammarians typically use the root ف-ع-ل f-ʿ-l to indicate the particular shape of any given element of a verbal paradigm. As an example, the form يتكاتب (root: ك-ت-ب) yutakātabu 'he is corresponded (with)' would be listed generically as يتفاعل yutafāʿalu (yuta1ā2a3u), specifying the generic shape of a strong Form VI passive verb, third-person masculine singular present indicative.

The maximum possible total number of verb forms derivable from a root — not counting participles and verbal nouns — is approximately 13 person/number/gender forms; times 9 tense/mood combinations, counting the س- sa- future (since the moods are active only in the present tense, and the imperative has only 5 of the 13 paradigmatic forms); times 17 form/voice combinations (since forms IX, XI–XV exist only for a small number of stative roots, and form VII cannot normally form a passive), for a total of 1,989. Each of these has its own stem form, and each of these stem forms itself comes in numerous varieties, according to the weakness (or lack thereof) of the underlying root.

Inflectional categories

Each particular lexical verb is specified by four stems, two each for the active and passive voices. In a particular voice, one stem (the past stem) is used for the past tense, and the other (the non-past stem) is used for the present and future tenses, along with non-indicative moods, e.g. subjunctive and imperative. The past and non-past stems are sometimes also called the perfective stem and imperfective stem, respectively, based on a traditional misinterpretation of Arabic stems as representing grammatical aspect rather than grammatical tense. (Although there is still some disagreement about the interpretation of the stems as tense or aspect, the dominant current view is that the stems simply represent tense, sometimes of a relative rather than absolute nature. There are some unusual usages of the stems in certain contexts that were once interpreted as indicating aspectual distinctions, but are now thought to simply be idiosyncratic constructions that do not neatly fit into any aspectual paradigm.)^{[citation needed]}

To the past stem, suffixes are added to mark the verb for person, number and gender, while to the non-past stem, a combination of prefixes and suffixes are added. (Very approximately, the prefixes specify the person and the suffixes indicate number and gender.) A total of 13 forms exist for each of the two stems, specifying person (first, second or third); number (singular, dual or plural); and gender (masculine or feminine).

There are six separate moods in the non-past: indicative, imperative, subjunctive, jussive, short energetic and long energetic. The moods are generally marked by suffixes. When no number suffix is present, the endings are -u for indicative, -a for subjunctive, no ending for imperative and jussive, ـَنْ -an for shorter energetic, ـَنَّ -anna for longer energetic. When number suffixes are present, the moods are either distinguished by different forms of the suffixes (e.g. ـُونَ -ūna for masculine plural indicative vs. ـُو -ū for masculine plural subjunctive/imperative/jussive), or not distinguished at all. The imperative exists only in the second person and is distinguished from the jussive by the lack of the normal second-person prefix ـت ta-/tu-.

The third person masculine singular past tense form serves as the "dictionary form" used to identify a verb, similar to the infinitive in English. (Arabic has no infinitive.) For example, the verb meaning 'write' is often specified as كَتَبَ kataba, which actually means 'he wrote'. This indicates that the past-tense stem is كَتَبْـ katab-; the corresponding non-past stem is ـكْتُبْـ -ktub-, as in يَكْتُبُ yaktubu 'he writes'. Using the third person masculine singular as the dictionary citation form is more useful in that the vowels that appear in the remaining present tense forms are evident. Especially in form I verbs, without prior knowledge, these vowels are often not evident based purely on the past-tense forms.

Tense

There are three tenses in Arabic: the past tense (اَلْمَاضِي al-māḍī), the present tense (اَلْمُضَارِع al-muḍāriʿ) and the future tense. The future tense in Classical Arabic is formed by adding either the prefix ‏سَـ‎ sa- or the separate word ‏سَوْفَ‎ sawfa onto the beginning of the present tense verb, e.g. سَيَكْتُبُ sa-yaktubu or سَوْفَ يَكْتُبُ sawfa yaktubu 'he will write'.

In some contexts, the tenses represent aspectual distinctions rather than tense distinctions. The usage of Arabic tenses is as follows:

• The past tense often (but not always) specifically has the meaning of a past perfective, i.e. it expresses the concept of 'he did' as opposed to 'he was doing'. The latter can be expressed using the combination of the past tense of the verb كَانَ kāna 'to be' with the present tense or active participle, e.g. كَانَ يَكْتُبُ kāna yaktubu or كَانَ كَاتِبٌ kāna kātibun 'he was writing'. There are some special verbs known as "compound verbs" that can express many grammatical aspects such as Inchoative, Durative etc., for example بَدَأ يُلْفِتُ النَظرَ badaʾa yulfitu al-naẓara means "It started to attract attention" which badaʾa conveys the meaning of "to start doing something (in the past)"
• The two tenses can be used to express relative tense (or in an alternative view, grammatical aspect) when following other verbs in a serial verb construction. In such a construction, the present tense indicates time simultaneous with the main verb, while the past tense indicates time prior to the main verb. (Or alternatively, the present tense indicates the imperfective aspect while the past tense indicates the perfective aspect.)

In all but Form I, there is only one possible shape for each of the past and non-past stems for a given root. In Form I, however, different verbs have different shapes. Examples:

• كَتَبَ يَكْتُبُ kataba yaktubu 'write'
• كَسِبَ يَكْسِبُ kasaba yaksibu 'earn'
• قَرَأَ يَقْرَأُ qaraʾa yaqraʾu 'read'
• قَدِمَ يَقْدَمُ qadima yaqdamu 'turn'
• كَبُرَ يَكْبُرُ kabura yakburu 'become big, grow up'

Notice that the second vowel can be any of a i u in both past and non-past stems. The vowel a occurs in most past stems, while i occurs in some (especially intransitive) and u occurs only in a few stative verbs (i.e. whose meaning is 'be X' or 'become X' where X is an adjective). The most common patterns are:

• past: a; non-past: u or i
• past: a, non-past: a (when the second or third root consonant is a "guttural," i.e. one of ʾ ʿ h ḥ)
• past: i; non-past: a
• past: u; non-past: u

Mood

There are three moods (حَالَات ḥālāt, a word that also means "cases"; sg. ‏حَالَة‎ ḥālah), whose forms are derived from the imperfective stem: the indicative mood (‏مَرْفُوع‎ marfūʿ), usually ending in u; the subjunctive (‏مَنْصُوب‎ manṣūb), usually ending in a; and the jussive (‏مَجْزُوم‎ majzūm), with no ending. In less formal Arabic and in spoken dialects, the subjunctive mood is used as the only imperfective tense (subjunctivism) and the final ḥarakah vowel is not pronounced.

The imperative (صِيغَة اَلْأَمْر ṣīghat al-amr) (positive, only 2nd person) is formed by dropping the verbal prefix (ت-) from the imperfective jussive stem, e.g. قَدِّم qaddim 'present!'. If the result starts with two consonants followed by a vowel (a or i), an elidible alif (ا) is added to the beginning of the word, usually pronounced as "i", e.g. اِغْسِلْ ighsil 'wash!' or اِفْعَل ifʿal 'do!' if the present form vowel is u, then the alif is also pronounced as u, e.g. أُكْتُب uktub 'write!'. Negative imperatives are formed from the jussive.

The exception to the above rule is the form (or stem) IV verbs. In these verbs a non-elidible alif ا pronounced as a- is always prefixed to the imperfect jussive form, e.g. أرسل arsil "send!", أضف ^[1]aḍif 'add!'.

The subjunctive is used in subordinate clauses after certain conjunctions. The jussive is used in negation, in negative imperatives, and in the hortative la+jussive. For example: 2. sg. m.:

• imperfect indicative تفعلُ tafʿalu 'you are doing'
• subjunctive أن تفعلَ an tafʿala 'that you do'
• jussive لا تفعلْ lā tafʿal its meaning is dependent upon the prefix which attaches to it; in this case, it means 'may you do not do!'
• short energetic تفعلنْ tafʿalan its meaning is dependent upon the prefix which attaches to it; if the prefix is "la" it means 'you should do'
• long energetic تفعلنَّ tafʿalanna it has more emphasis than the short energetic, its meaning is dependent upon the prefix which attaches to it; if the prefix is "la" it means 'you must do'
• imperative افعل ifʿal 'do!'.

Voice

Arabic has two verbal voices (صِيغَات ṣīghāt "forms", sg. صِيغَة ṣīghah), active (صِيغَة اَلْمَعْلُوم ṣīghat al-maʿlūm), and passive (صِيغَة اَلْمَجْهُول ṣīghat al-majhūl). The passive voice is expressed by a change in vocalization. For example:

• active فَعَلَ faʿala 'he did', يَفْعَلُ yafʿalu 'he is doing'
• passive فُعِلَ fuʿila 'it was done', يُفْعَلُ yufʿalu 'it is being done'

Thus, the active and passive forms are spelled identically in Arabic; only their vowel markings differ. There are some exceptions to this in the case of weak roots.

Participle

Every verb has a corresponding active participle, and most have passive participles. E.g. معلم muʿallim 'teacher' is the active participle to stem II. of the root ع-ل-م ʿ-l-m ('know').

• The active participle to Stem I is فاعل fāʿil, and the passive participle is مفعول mafʿūl.
• Stems II–X take prefix مـ mu- and nominal endings for both the participles, active and passive. The difference between the two participles is only in the vowel between the last two root letters, which is -i- for active and -a- for passive (e.g. II. active مفعِّل mu-faʿʿil, and passive مفعَّل mu-faʿʿal).

Verbal noun (maṣdar)

In addition to a participle, there is a verbal noun (in Arabic, مَصْدَر maṣdar, pl. مَصَادِر maṣādir, literally meaning 'source'), sometimes called a gerund, which is similar to English gerunds and verb-derived nouns of various sorts (e.g. "running" and "a run" from "to run"; "objection" from "to object"). As shown by the English examples, its meaning refers both to the act of doing something and (by frequent semantic extension) to its result. One of its syntactic functions is as a verbal complement of another verb, and this usage it corresponds to the English gerund or infinitive (He prevented me from running or He began to run).

• verbal noun formation to stem I is irregular.
• the verbal noun to stem II is تفعيل tafʿīl. For example: تحضير taḥḍīr 'preparation' is the verbal noun to stem II. of ح-ض-ر ḥ-ḍ-r ('to be present').
• stem III often forms its verbal noun with the feminine form of the passive participle, so for ساعد sāʿada, 'he helped', produces the verbal noun مساعدة musāʿadah. There are also some verbal nouns of the form فعال fiʿāl: جاهد jāhada, 'he strove', yields jihād جهاد 'striving' (for a cause or purpose).

Some well-known examples of verbal nouns are فتح fatḥ (see Fatah) (Form I), تنظيم tanẓīm (Form II), جهاد jihād (Form III), إسلام islām (Form IV), انتفاضة intifāḍah (feminine of Form VIII verbal noun), and استقلال istiqlāl (Form X).

Derivational categories, conjugations

The system of verb conjugations in Arabic is quite complicated, and is formed along two axes. One axis, known as the form (described as "Form I", "Form II", etc.), is used to specify grammatical concepts such as causative, intensive, reciprocal, passive or reflexive, and involves varying the stem form. The other axis, known as the weakness, is determined by the particular consonants making up the root. For example, defective (or third-weak or final-weak) verbs have a و w or ي y as the last root consonant (e.g. ر-م-ي r-m-y 'throw', د-ع-و d-ʿ-w 'call'), and doubled (or geminated) verbs have the second and third consonants the same (e.g. م-د-د m-d-d 'extend'). These "weaknesses" have the effect of inducing various irregularities in the stems and endings of the associated verbs.

Examples of the different forms of a sound verb (i.e. with no root weaknesses), from the root ك-ت-ب k-t-b 'write' (using ح-م-ر ḥ-m-r 'red' for Form IX, which is limited to colors and physical defects):

The main types of weakness are as follows:

Conjugation

Regular verb conjugation for person-number, tense-aspect-mood, and participles

In Arabic the grammatical person and number as well as the mood are designated by a variety of prefixes and suffixes. The following table shows the paradigm of a regular sound Form I verb, kataba (كتب) 'to write'. Most of the final short vowels are often omitted in speech, except the vowel of the feminine plural ending -na, and normally the vowel of the past tense second person feminine singular ending -ti.

The initial vowel in the imperative (which is elidable) varies from verb to verb, as follows:

• The initial vowel is u if the stem begins with two consonants and the next vowel is u or ū.
• The initial vowel is i if the stem begins with two consonants and the next vowel is anything else.
• There is no initial vowel if the stem begins with one consonant.

In unvocalised Arabic, katabtu, katabta, katabti and katabat are all written the same: كتبت. Forms katabtu and katabta (and sometimes even katabti) can be abbreviated to katabt in spoken Arabic and in pausa, making them also sound the same.

ا (alif) in final ـُوا (-ū) is silent.

Weak roots

Roots containing one or two of the radicals و w (wāw), ي y (yāʾ ) or ء ʾ (hamzah) often lead to verbs with special phonological rules because these radicals can be influenced by their surroundings. Such verbs are called "weak" (verba infirma, 'weak verbs') and their paradigms must be given special attention. In the case of hamzah, these peculiarities are mainly orthographical, since hamzah is not subject to elision (the orthography of ء hamzah and ا alif is unsystematic due to confusion in early Islamic times).^{[citation needed]} According to the position of the weak radical in the root, the root can be classified into four classes: first weak, second weak, third weak (or final weak) and doubled, where both the second and third radicals are identical. Some roots fall into more than one category at once.

Assimilated (first-weak) roots

Most first-weak verbs have a و w as their first radical. These verbs are entirely regular in the past tense. In the non-past, the w drops out, leading to a shorter stem (e.g., (يَجِدُ) وَجَدَ wajada (yajidu) 'to find'), where the stem is ـجِدـ -jid- in place of a longer stem like ـجْلِدـ -jlid- from the verb جَلَدَ (يَجْلِدُ) jalada (yajlidu) 'to whip, flog'. This same stem is used throughout, and there are no other irregularities except for the imperative, which has no initial vowel, consistent with the fact that the stem for the imperative begins with only one consonant.

There are various types of assimilated (first-weak) Form I verbs:

Hollow (second-weak) roots

The following shows a paradigm of a typical Form I hollow (second-weak) verb قَالَ (قُلْتُ، يَقُولُ) qāla (qultu, yaqūlu) (root: ق-و-ل q-w-l) 'to say', parallel to verbs of the فَعَلَ (يَفْعُلُ) faʿala (yafʿulu) type. See notes following the table for explanation.

All hollow (second-weak) verbs are conjugated in a parallel fashion. The endings are identical to those of strong verbs, but there are two stems (a longer and a shorter) in each of the past and non-past. The longer stem is consistently used whenever the ending begins with a vowel, and the shorter stem is used in all other circumstances. The longer stems end in a long vowel plus consonant, while the shorter stems end in a short vowel plus consonant. The shorter stem is formed simply by shortening the vowel of the long stem in all paradigms other than the active past of Form I verbs. In the active past paradigms of Form I, however, the longer stem always has an ā vowel, while the shorter stem has a vowel u or i corresponding to the actual second root consonant of the verb.

No initial vowel is needed in the imperative forms because the non-past stem does not begin with two consonants.

There are various types of Form I hollow verbs:

• قَالَ، قُلْنَ (يَقُولُ، يَقُلْنَ) (root: ق-و-ل) qāla, qulna (yaqūlu, yaqulna) 'to say', formed from verbs with و w as their second root consonant and parallel to verbs of the فَعَلَ (يَفْعُلُ) faʿala (yafʿulu) type
• سَارَ، سِرْنَ (يَسِيرُ، يَسِرْنَ) (root: س-ي-ر) sāra, sirna (yasīru, yasirna) 'to get going, to travel', formed from verbs with ي y as their second root consonant and parallel to verbs of the فَعَلَ (يَفْعِلُ) faʿala (yafʿilu) type
• خَافَ، خُفْنَ (يَخَافُ، يَخَفْنَ) (root: خ-و-ف) khāfa khufna (yakhāfu yakhafna) 'to fear', formed from verbs with و w as their second root consonant and parallel to verbs of the فَعِلَ (يَفْعَلُ) faʿila (yafʿalu) type
• نَامَ، نِمْنَ (يَنَامُ، يَنَمْنَ) (root: ن-ي-م) nāma, nimna (yanāmu, yanamna) 'to sleep', formed from verbs with ي y as their second root consonant and parallel to verbs of the فَعِلَ (يَفْعَلُ) faʿila (yafʿalu) type

The passive paradigm of all Form I hollow verbs is as follows:

• قِيلَ، قِلْنَ (يُقَالُ، يُقَلْنَ) qīla qilna (yuqālu yuqalna) 'to be said'

Defective (third-weak) roots

فَعَى (يَفْعِي) faʿā (yafʿī)

The following shows a paradigm of a typical Form I defective (third-weak) verb رَمَى (يَرْمِي) ramā (yarmī) (root: ر-م-ي r-m-y) 'to throw', parallel to verbs of the فَعَلَ (يَفْعِلُ) faʿala (yafʿilu) type. See notes following the table for explanation.

Two stems each

Each of the two main stems (past and non-past) comes in two variants, a full and a shortened. For the past stem, the full is رَمَيـ ramay-, shortened to رَمـ ram- in much of the third person (i.e. before vowels, in most cases). For the non-past stem, the full is rmiy-, shortened to rm- before -ū -ī. The full non-past stem رْمِيْـ rmiy- appears as رْمِيـ rmī- when not before a vowel; this is an automatic alternation in Classical Arabic. The places where the shortened stems occur are indicated by silver (past), gold (non-past).

Irregular endings

The endings are actually mostly regular. But some endings are irregular, in boldface:

• Some of the third-person past endings are irregular, in particular those in رَمَى ram-ā 'he threw', رَمَوْا⁩ ram-aw 'they (masc.) threw'. These simply have to be memorized.
• Two kinds of non-past endings are irregular, both in the "suffixless" parts of the paradigm (largely referring to singular masculine or singular combined-gender). In the indicative, the full stem ـرْمِي -rmī actually appears normally; what is irregular is the lack of the -u normally marking the indicative. In the jussive, on the other hand, the stem actually assumes a unique shortened form ـرْمِـ -rmi, with a short vowel that is not represented by a letter in the Arabic.

فَعَا (يَفْعُو) faʿā (yafʿū)

The following shows a paradigm of a typical Form I defective (third-weak) verb دَعَا (يَدْعُو) (root: د-ع-و) daʿā (yadʿū) 'to call', parallel to verbs of the فَعَلَ (يَفْعُلُ) faʿala (yafʿulu) type. Verbs of this sort are entirely parallel to verbs of the فَعَا (يَفْعِي) faʿā (yafʿī) type, although the exact forms can still be tricky. See notes following the table for explanation.

Verbs of this sort are work nearly identically to verbs of the فَعَى (يَفْعِي) faʿā (yafʿī) type. There are the same irregular endings in the same places, and again two stems in each of the past and non-past tenses, with the same stems used in the same places:

• In the past, the full stem is دَعَوـ daʿaw-, shortened to دَعـ daʿ-.
• In the non-past, the full stem is دْعُوْـ dʿuw-, rendered as دْعُوـ dʿū- when not before a vowel and shortened to دْعـ dʿ- before ـُو، ـِي -ū, -ī.

The Arabic spelling has the following rules:

• In the third person masculine singular past, regular ا ʾalif appears instead of ى ʾalif maqṣūrah: hence دَعَا, not *دَعَى.
• The otiose final alif appears only after the final wāw of the plural, not elsewhere: hence تَدْعُو 'you (masc sg) call (ind)' but تَدْعُوا 'you (masc pl) call (sjv)', even though they are both pronounced تَدْعُو tadʿū.

فَعِيَ (يَفْعَى) faʿiya (yafʿā)

The following shows a paradigm of a typical Form I defective (third-weak) verb نَسِيَ (يَنْسَ) nasiya (yansā) (root: ن-س-ي) 'to forget', parallel to verbs of the فَعِلَ (يَفْعَلُ) faʿila (yafʿalu) type. These verbs differ in a number of significant respects from either of the above types.

Multiple stems

This variant is somewhat different from the variants with ـِي -ī or ـُو -ū in the non-past. As with other third-weak verbs, there are multiple stems in each of the past and non-past, a full stem composed following the normal rules and one or more shortened stems.

• In this case, only one form in the past uses a shortened stem: نَسُو nas-ū 'they (masc) forgot'. All other forms are constructed regularly, using the full stem نَسِيْـ nasiy- or its automatic pre-consonant variant نَسِيـ nasī-.
• In the non-past, however, there are at least three different stems:

1. The full stem ـنْسَيـ -nsay- occurs before ـَ/ـَى -a/ā- or ـنـ -n-, that is before dual endings, feminine plural endings and energetic endings corresponding to forms that are endingless in the jussive.
2. The modified stem ـنْسَى -nsā occurs in "endingless" forms (i.e. masculine or common-gender singular, plus 1st plural). As usual with third-weak verbs, it is shortened to ـنْسَ -nsa in the jussive. These forms are marked with red.
3. Before endings normally beginning with ـِ/ـِي -i/ī- or ـُ/ـُو -u/ū-, the stem and endings combine into a shortened form: e.g. expected *تَنْسَيِينَ *ta-nsay-īna 'you (fem sg) forget', *تَنْسَيُونَ *ta-nsay-ūna 'you (masc pl) forget' instead become تَنْسَيْنَ ta-nsayna, تَنْسَوْنَ ta-nsawna respectively. The table above chooses to segment them as تَنْسَيْنَ ta-nsa-yna, تَنْسَوْنَ ta-nsa-wna, suggesting that a shortened stem ـنْسَـ -nsa- combines with irregular (compressed) endings ـيْنَ -yna < *ـِينَ *-īna, ـوْنَ -wna < ـُونَ *-ūna. Similarly subjunctive/jussive تَسنَوْا ta-nsaw < تَسنَيُوْا *ta-nsay-ū; but note energetic تَنْسَوُنَّ ta-nsawunna < تَنْسَيُنَّ *ta-nsay-unna, where the original ـيُـ *-yu- has assimilated to ـوُـ -wu-. Consistent with the above analysis, we analyze this form as تَنْسَوُنَّ ta-nsa-wunna, with an irregular energetic ending ـوُنَّ -wunna where a glide consonant has developed after the previous vowel. However, since all moods in this case have a form containing ـنْسَوـ -nsaw-, an alternative analysis would consider ـنْسَوـ -nsaw and ـنْسَيـ -nsay as stems. These forms are marked with gold.

Irregular endings

The endings are actually mostly regular. But some endings are irregular in the non-past, in boldface:

• The non-past endings in the "suffixless" parts of the paradigm (largely referring to singular masculine or singular combined-gender). In the indicative and subjunctive, the modified stem ـنسَاـ -nsā appears, and is shortened to ـنسَـ -nsa in the jussive. In the forms actually appears normally; what is irregular is the lack of the ـُ -u normally marking the indicative. In the jussive, on the other hand, the stem actually assumes a unique shortened form ـنْسَـ -nsa, with a short vowel that is not represented by a letter in the Arabic script.
• In the forms that would normally have suffixes ـِ/ـِي -i/ī- or ـُ/ـُو -u/ū-, the stem and suffix combine to produce ـنْسَيـ -nsay-, ـنْسَوـ -nsaw-. These are analyzed here as consisting of a shortened stem form ـنْسَـ -nsa- plus irregular (shortened or assimilated) endings.

Doubled roots

The following shows a paradigm of a typical Form I doubled verb مَدَّ (يَمُدُّ) (root: م-د-د) madda (yamuddu) 'to extend', parallel to verbs of the فَعَلَ (يَفْعُلُ) faʿala (yafʿulu) type. See notes following the table for explanation.

All doubled verbs are conjugated in a parallel fashion. The endings are for the most part identical to those of strong verbs, but there are two stems (a regular and a modified) in each of the past and non-past. The regular stems are identical to the stem forms of sound verbs, while the modified stems have the two identical consonants pulled together into a geminate consonant and the vowel between moved before the geminate. In the above verb مَدَّ (يَمُدُّ) madda (yamuddu) 'to extend (s.th.)', the past stems are مَدَدـ madad- (regular), مَدّـ madd- (modified), and the non-past stems are مْدُدـ mdud- (regular), مُدّـ mudd- (modified). In the table, places where the regular past stem occurs are in silver, and places where the regular non-past stem occurs are in gold; everywhere else, the modified stem occurs.

No initial vowel is needed in most of the imperative forms because the modified non-past stem does not begin with two consonants.

The concept of having two stems for each tense, one for endings beginning with vowels and one for other endings, occurs throughout the different kinds of weaknesses.

Following the above rules, endingless jussives would have a form like تَمْدُد tamdud, while the corresponding indicatives and subjunctives would have forms like تَمُدُّ tamuddu, تَمُدَّ tamudda. As a result, for the doubled verbs in particular, there is a tendency to harmonize these forms by adding a vowel to the jussives, usually ـَ a, sometimes ـِ i. These are the only irregular endings in these paradigms, and have been indicated in boldface. The masculine singular imperative likewise has multiple forms, based on the multiple forms of the jussive.

There are various types of doubled Form I verbs:

Formation of derived stems ("forms")

Arabic verb morphology includes augmentations of the root, also known as forms, an example of the derived stems found among the Semitic languages. For a typical verb based on a triliteral root (i.e. a root formed using three root consonants), the basic form is termed Form I, while the augmented forms are known as Form II, Form III, etc. The forms in normal use are Form I through Form X; Forms XI through XV exist but are rare and obsolescent. Forms IX and XI are used only with adjectival roots referring to colors and physical defects (e.g. "red", "blue", "blind", "deaf", etc.), and are stative verbs having the meaning of "be X" or "become X" (e.g. Form IX iḥmarra 'be red, become red, blush', Form XI iḥmārra with the same meaning). Although the structure that a given root assumes in a particular augmentation is predictable, its meaning is not (although many augmentations have one or more "usual" or prototypical meanings associated with them), and not all augmentations exist for any given root. As a result, these augmentations are part of the system of derivational morphology, not part of the inflectional system.

The construction of a given augmentation is normally indicated using the dummy root f–ʿ–l (ف–ع–ل), based on the verb faʿala 'to do'. Because Arabic has no direct equivalent to the infinitive form of Western languages, the third-person masculine singular past tense is normally used as the dictionary form of a given verb, i.e. the form by which a verb is identified in a dictionary or grammatical discussion. Hence, the word faʿala above actually has the meaning of 'he did', but is translated as 'to do' when used as a dictionary form.

Verbs based on quadriliteral roots (roots with four consonants) also exist. There are four augmentations for such verbs, known as Forms Iq, IIq, IIIq and IVq. These have forms similar to Forms II, V, VII and IX respectively of triliteral verbs. Forms IIIq and IVq are fairly rare. The construction of such verbs is typically given using the dummy verb faʿlala (root: ف-ع-ل-ل). However, the choice of this particular verb is somewhat non-ideal in that the third and fourth consonants of an actual verb are typically not the same, despite the same consonant used for both; this is a particular problem e.g. for Form IVq. The verb tables below use the dummy verb faʿlaqa (root: ف-ع-ل-ق) instead.

Some grammars, especially of colloquial spoken varieties rather than of Classical Arabic, use other dummy roots. For example, A Short Reference Grammar of Iraqi Arabic (Wallace M. Erwin) uses فمل FaMaLa (root: ف-م-ل) and فستل FaSTaLa (root: ف-س-ت-ل) for three and four-character roots, respectively (standing for "First Middle Last" and "First Second Third Last"). Commonly the dummy consonants are given in capital letters.

The system of identifying verb augmentations by Roman numerals is an invention by Western scholars. Traditionally, Arabic grammarians did not number the augmentations at all, instead identifying them by the corresponding dictionary form. For example, Form V would be called "the tafaʿʿala form".

Each form can have either active or passive forms in the past and non-past tenses, so reflexives are different from passives.

Note that the present passive of forms I and IV are the same. Otherwise there is no confusion.

Sound verbs

Sound verbs are those verbs with no associated irregularities in their constructions. Verbs with irregularities are known as weak verbs; generally, this occurs either with (1) verbs based on roots where one or more of the consonants (or radicals) is w (wāw, و), y (yāʾ, ي) or the glottal stop ʾ (hamzah, ﺀ); or (2) verbs where the second and third root consonants are the same.

Some verbs that would be classified as "weak" according to the consonants of the verb root are nevertheless conjugated as a strong verb. This happens, for example:

• Largely, to all verbs whose only weakness is a hamzah radical; the irregularity is in the Arabic spelling but not the pronunciation, except in a few minor cases.
• Largely, to all verbs whose only weakness is a y in the first radical (the "assimilated" type).
• To all verbs conjugated in Forms II, III, V, VI whose only weakness is a و w or ي y in the first or second radicals (or both).

Form VIII assimilations

Form VIII has a ـتـ -t- that is infixed into the root, directly after the first root consonant. This ـتـ -t- assimilates to certain coronal consonants occurring as the first root consonant. In particular, with roots whose first consonant is د، ز، ث، ذ، ص، ط، ض، ظ d z th dh ṣ ṭ ḍ ẓ, the combination of root and infix ت t appears as دّ، زد، ثّ، ذّ، صط، طّ، ضط، ظّ dd zd thth dhdh ṣṭ ṭṭ ḍṭ ẓẓ. That is, the t assimilates the emphasis of the emphatic consonants ص، ط، ض، ظ ṣ ṭ ḍ ẓ and the voicing of د، ز d z, and assimilates entirely to the interdental consonants ث، ذ، ظ th dh ẓ. The consonant cluster ضط ḍṭ, as in اضطرّ iḍṭarra 'compel, force', is unexpected given modern pronunciation, having a voiced stop next to a voiceless one; this reflects the fact that ط ṭ was formerly pronounced voiced, and ض ḍ was pronounced as the emphatic equivalent not of د d but of an unusual lateral sound. (ض ḍ was possibly an emphatic voiced alveolar lateral fricative /ɮˤ/ or a similar affricated sound /dɮˤ/ or /dˡˤ/; see the article on the letter ض ḍād.)

Defective (third-weak) verbs

Other than for Form I active, there is only one possible form for each verb, regardless of whether the third root consonant is و w or ي y. All of the derived third-weak verbs have the same active-voice endings as (فعى (يفعي faʿā (yafʿī) verbs except for Forms V and VI, which have past-tense endings like (فعى (يفعي faʿā (yafʿī) verbs but non-past endings like (فعي (يفعى faʿiya (yafʿā) verbs. The passive-voice endings of all third-weak verbs (whether Form I or derived) are the same as for the (فعي (يفعى faʿiya (yafʿā) verbs. The verbal nouns have various irregularities: feminine in Form II, -in declension in Form V and VI, glottal stop in place of root w/y in Forms VII–X.

The active and passive participles of derived defective verbs consistently are of the -in and -an declensions, respectively.

Defective Form IX verbs are extremely rare. Heywood and Nahmad list one such verb, iʿmāya 'be/become blind', which does not follow the expected form اعميّ *iʿmayya.^[2] They also list a similarly rare Form XI verb اعمايّ iʿmāyya 'be/become blind' — this time with the expected form.

Hollow (second-weak) verbs

Only the forms with irregularities are shown. The missing forms are entirely regular, with w or y appearing as the second radical, depending on the root. There are unexpected feminine forms of the verbal nouns of Form IV, X.

Assimilated (first-weak) verbs

When the first radical is w, it drops out in the Form I non-past. Most of the derived forms are regular, except that the sequences uw iw are assimilated to ū ī, and the sequence wt in Form VIII is assimilated to tt throughout the paradigm. The following table only shows forms with irregularities in them.

The initial w also drops out in the common Form I verbal noun علة ʿilah (e.g. صلة ṣilah 'arrival, link' from وصلة waṣalah 'arrive'). Root: و-ع-ل

When the first radical is y, the forms are largely regular. The following table only shows forms that have some irregularities in them, indicated in boldface. Root: ي-ع-ل

Doubled verbs

Root: ف-ل-ل

Hamzated verbs

The largest problem with so-called "hamzated" verbs (those with a glottal stop ʾ or "hamzah" as any of the root consonants) is the complicated way of writing such verbs in the Arabic script (see the article on hamzah for the rules regarding this). In pronunciation, these verbs are in fact almost entirely regular.

The only irregularity occurs in verbs with a hamzah ء as the first radical. A phonological rule in Classical Arabic disallows the occurrence of two hamzahs in a row separated by a short vowel, assimilating the second to the preceding vowel (hence ʾaʾ ʾiʾ ʾuʾ become ʾā ʾī ʾū). This affects the following forms:

• The first-person singular of the non-past of Forms I, IV and VIII.
• The entire past and imperative of Form IV.

In addition, any place where a hamzat al-waṣl (elidable hamzah) occurs will optionally undergo this transformation. This affects the following forms:

• The entire imperative of Form I.
• The entire past and imperative of Form VIII, as well as the verbal noun of Form VIII.

There are the following irregularities:

• The common verbs ʾakala (أكل; root: ء-ك-ل) 'eat', ʾakhadha (أخذ; root: ء-خ-ذ) 'take', ʾamara (أمر; root: ء-م-ر) 'command' have irregular, short imperatives kul, khudh, mur.
• Form VIII of the common verb ʾakhadha 'take' is ittakhadha 'take on, assume', with irregular assimilation of the hamzah.
• The common verb saʾala yasʾalu 'ask' has an alternative non-past yasalu with missing hamzah.

Doubly weak verbs

Doubly weak verbs have two "weak" radicals; a few verbs are also triply weak. Generally, the above rules for weak verbs apply in combination, as long as they do not conflict. The following are cases where two types of weaknesses apply in combination:

• Verbs with a w in the first radical and a w or y in the third radical. These decline as defective (third-weak) verbs, and also undergo the loss of w in the non-past of Form I, e.g. waqā yaqī 'guard', wafā yafī 'complete, fulfill (a promise)', waliya yalī 'be near, follow'. These verbs have extremely short imperatives qi fi li (feminine qī fī lī, masculine plural qū fū lū, feminine plural iqna ifna ilna), although these are not normally used in Modern Standard Arabic. Similarly, verbs of this sort in Form IV and Form VIII are declined as defective but also have the normal assimilations of w-initial verbs, e.g. Form IV awfā yūfī 'fulfill a vow', Form VIII ittaqā yattaqī 'fear (God)', augmentations of wafā yafī and waqā yaqī, respectively (see above).
• Verbs with a hamzah in the first radical and a w or y in the third radical. These decline as defective (third-weak) verbs, and also undergo the assimilations associated with the initial hamzah, e.g. the common verb ʾatā yaʾtī 'come' (first singular non-past ʾātī 'I come') and the related Form IV verb ʾātā yuʾtī 'bring' (first singular non-past ʾūtī 'I bring').

The following are examples where weaknesses would conflict, and hence one of the "weak" radicals is treated as strong:

• Verbs with a w or y in both the second and third radicals. These are fairly common, e.g. rawā yarwī 'recount, transmit'. These decline as regular defective (third-weak) verbs; the second radical is treated as non-weak.
• Verbs with a w in the first radical and the second and third radicals the same. These verbs do not undergo any assimilations associated with the first radical, e.g. wadda (wadidtu) yawaddu 'to love'.
• Verbs with a hamza in the first radical and the second and third radicals the same. These verbs do not undergo any assimilations associated with the first radical, e.g. ʾajja yaʾujju 'burn', first singular non-past ʾaʾujju 'I burn', despite the two hamzahs in a row.

The following are cases with special irregularities:

• Verbs with a w or y in the second radical and a hamzah in the third radical. These are fairly common, e.g. the extremely common verb jāʾa yajīʾu 'come'. The only irregularity is the Form I active participle, e.g. jāʾin 'coming', which is irregularly declined as a defective (third-weak) participle (presumably to avoid a sequence of two hamzahs in a row, as the expected form would be *jāʾiʾ).
• The extremely common verb raʾā yarā 'see'. The hamzah drops out entirely in the non-past. Similarly in the passive, ruʾiya yurā 'be seen'. The active participle is regular rāʾin and the passive participle is regular marʾīy-. The related Form IV verb arā yūrī 'show' is missing the hamzah throughout. Other augmentations are regular: Form III rāʾā yurāʾī 'dissemble', Form VI tarāʾā yatarāʾā 'look at one another', Form VIII irtaʾā yartaʾī 'think'.
• The common verb ḥayiya yaḥyā 'live', with an alternative past tense ḥayya. Form IV aḥyā yuḥyī 'resuscitate, revive' is regular. Form X istaḥyā yastaḥyī 'spare alive, feel ashamed' also appears as istaḥayya and istaḥā.

Summary of vowels

The vowels for the various forms are summarized in this table:

Verbs in colloquial Arabic

The Classical Arabic system of verbs is largely unchanged in the colloquial spoken varieties of Arabic. The same derivational system of augmentations exists, including triliteral Forms I through X and quadriliteral Forms I and II, constructed largely in the same fashion (the rare triliteral Forms XI through XV and quadriliteral Forms III and IV have vanished). The same system of weaknesses (strong, defective/third-weak, hollow/second-weak, assimilated/first-weak, doubled) also exists, again constructed largely in the same fashion. Within a given verb, two stems (past and non-past) still exist along with the same two systems of affixes (suffixing past-tense forms and prefixing/suffixing non-past forms).

The largest changes are within a given paradigm, with a significant reduction in the number of forms. The following is an example of a regular verb paradigm in Egyptian Arabic.