Template:ar-conj

From Wiktionary, the free dictionary
Jump to navigation Jump to search

Usage

This template creates a conjugation table for all Arabic verbs.

This template should be placed in a conjugation section, headed by a ====Conjugation==== section header. The template automatically categorizes the entry into the appropriate Arabic verb subcategory (e.g. Category:Arabic form-IV verbs). Form-I verbs are also categorized according to the last vowel in the past and non-past, since these vowels vary from verb to verb (e.g. Category:Arabic form-I verbs with past vowel a and non-past vowel u). Verbs that are conjugate geminate, such as form-X اِسْتَرَدَّ (istaradda), are categorized accordingly, in this case in Category:Arabic geminate form-X verbs. If the verb is irregular, it will be categorized into Category:Arabic irregular verbs.

Basic usage

This template has a single required parameter, which is used to specify the verb form (I, II, III, ..., XV, as well as Iq ... IVq). For non-form-I verbs, often that is all that needs to be specified, although typically you should also specify whether there is a passive voice and if so, what type of passive (e.g. full passive, impersonal passive, passive-only, etc.). An example of a form-II verb ِis رَدَّدَ (raddada, to repeat), which can be declared as {{ar-conj|<II.pass>}}, where <II.pass> is an angle-bracket spec and the parts of the angle-bracket spec between periods/full stops are called indicators. In this spec there are two indicators, II (the verb form) and pass (indicating that a full passive occurs). In practice, this is written just as {{ar-conj|II.pass}}, since angle brackets can be omitted when the spec contains angle brackets at the beginning and end, and nowhere else. (Beware that if you embed a template like {{m}} or {{q}} in a footnote — see below for footnotes — you will need to use the form with explicit angle brackets, because {{m}}, {{q}} and several other templates generate HTML under the hood, which contains angle brackets that confuse the parser if the outer angle brackets have been omitted.)

Form-I verbs need extra information specified, including the past and non-past vowels as well as the verbal nouns (which are unpredictable) and in some cases the active participle(s), when they are unpredictable and do not follow the normal فَاعِل (fāʕil) pattern. A typical example is كَتَبَ (kataba, to write), which might use the following spec: {{ar-conj|I/a~u.pass.vn:كِتَابَة,كَتْب,كِتَاب}}. This contains three indicators, the first of which (I/a~u) is a compound indicator that embeds several pieces of information, in this case the form (I), the past vowel a (i.e. kataba, not #katiba or katuba) and the non-past vowel u (i.e. yaktubu, not #yaktibu or yaktaba). The third indicator is also compound and is an override, explicitly specifying the forms in a slot (in this case the verbal noun slot, whose abbreviation is vn). This is needed for form-I verbs because there is no default verbal noun for these verbs.

You can also specify the verb explicitly, e.g. {{ar-conj|ردد<II.pass>}}. This is useful on pages where the pagename is not the verb lemma, e.g. test and documentation pages as well as multiword expressions, which are described in more detail below. It is also used in {{ar-verb form}}, which is used on pages describing non-lemma verb forms but takes the same syntax as {{ar-conj}}.

Non-form-I examples

For the form-II verb رَدَّدَ (raddada, to repeat), use {{ar-conj|II.pass}} (as described above), which produces:

For the form-VIII geminate verb اِضْطَرَّ (iḍṭarra, to compel, to force), use {{ar-conj|VIII.pass}}, which produces:

Note that the module correctly knows how to conjugate all verb forms and weaknesses. In the above form-VIII verb, there are multiple possible jussive and imperative forms, all of which are generated and shown in the table.

For the form-X hollow hamzated verb استضاء (to be lit; to seek enlightenment), use {{ar-conj|X.II:و.ipass}}, which produces:

Here, ipass indicates that the verb has only an impersonal passive (i.e. only third-person masculine singular forms exist) rather than a full passive, and II:و specifies that the second radical is a و (wāw), which cannot be inferred from the lemma. Note that for non-form-I verbs, it's not strictly necessary to specify the non-inferrable second radical of hollow verbs or third radical of final-weak verbs, because it doesn't affect the conjugation. Doing so, however, allows the verbs to be correctly categorized by weak radical. For example, the above verb will be categorized into Category:Arabic form-X verbs with و as second radical.

Note also that the module knows the rules for determining the correct seat of hamzas, which follows complex rules and for which there are sometimes alternatives, as in the third-person masculine plural non-past indicative of the above verb, which is either يَسْتَضِيئُونَ (yastaḍīʔūna) (with the hamza over a ي) or يَسْتَضِيؤُونَ (yastaḍīʔūna) (with the hamza over a و).

Form-I examples

For form-I كَتَبَ (kataba, to write), use {{ar-conj|I/a~u.pass.vn:كِتَابَة,كَتْب,كِتَاب}} (as described above), which produces:

For form-I تَرَّ (tarra, to be cut off, to be severed), which has non-past vowel either u or i, use {{ar-conj|I/a~u,i.nopass.vn:تَرّ,تُرُور}}, which produces:

Here, commas separate the different possible non-past vowels, and nopass means there is no passive.

Form-I stative verbs (generally those whose past vowel is i or u) typically have unpredictable active participles as well as verbal nouns, and both must be specified. An example is تَلِفَ (talifa, to perish, to be spoiled), which uses {{ar-conj|I/i~a.ipass.vn:تَلَف.ap:+,+++,+an}}, producing:

Here, there is both a verbal noun override specifying the verbal noun تَلَف (talaf) and an active participle override specifying three possible active participles. Because certain active participle patterns are very common, shortcuts are provided for them. Here, + indicates the default فَاعِل (fāʕil) participle shape; +++ indicates the فَعِل (faʕil) participle shape; and +an indicates the فَعْلَان (faʕlān) participle shape. This is documented in more detail below.

If there is more than one possible combination of past/non-past vowels, but the combinations differ in both the past and non-past vowels, it is recommended to use multiple conjugation tables. An example is رَفَقَ (rafaqa, to be kind; to be a friend) with non-past vowel u, which also occurs as رَفِقَ (rafiqa) with non-past vowel a and رَفُقَ (rafuqa) with non-past vowel u. This would be written as follows:

{{ar-conj|I/a~u.ipass.vn:رِفْق,مَرْفِق,مِرْفَق,مَرْفَق.ap:+,++}}
{{ar-conj|I/i~a.ipass.vn:رَفَق.ap:+,++}}
{{ar-conj|I/u~u.ipass.vn:رَفَاقَة,رِفْق.ap:++}}

which produces

Note that in this case, the different vowel patterns are also associated with different verbal nouns and (to some extent) different active participles.

In general, you must specify the past and non-past vowels for a form-I verb, or an error is thrown. There are only two exceptions: (1) passive-only verbs, and (2) final-weak verbs. An example of the latter is بَكَى (bakā, to cry; to weep over), which would be written {{ar-conj|I.pass.vn:بُكَاء,بُكًى}}, which produces:

Here, the vowels are inferred as a~i, consistent with the last letter of the unvocalized form. (Vowel pattern a~u ends in ا and vowel pattern i~a ends in ي.)

For form-I weak verbs, it is important to know the identity of all radicals, as it determines the form of the passive participles. This applies in particular to hollow and final-weak verbs. Most of the time, radical can be safely inferred from the non-past vowel, but this is not the case when the non-past vowel is a. In this situation, the radical must be given or an error occurs. An example is نَامَ (nāma, to sleep; to forget about), which has vowel pattern i~a and an impersonal passive, hence a passive participle (مَنُوم (manūm) since the second radical is و). For this verb, the second radical must be given explicitly or an error occurs. Thus it should be written {{ar-conj|I/i~a.II:و.ipass.vn:نَوْم,مَنَام.ap:+}}, which produces:

A similar situation occurs with form-I final-weak verbs with vowel pattern i~a. An example is رَضِيَ (raḍiya, to please), which should be written {{ar-conj|I.III:و.pass.vn:رِضًا,رِضًى,رِضْوَان,رُضْوَان,مَرْضَاة.pp:+,مَرْضِيّ}}, which produces:

Here, the last radical is و, leading to an expected passive participle مَرْضُوّ (marḍuww). This does occur, but so does an analogical passive participle مَرْضِيّ (marḍiyy), so we list both using an override.

Indicators

This section gives full information on the indicators that can occur in |1=. Indicators are separated by a period (.). The first indicator is required and specifies the verb form (I, II, etc.) and, for form-I verbs, the past and non-past vowels. Additional indicators are optional and can come in any order.

Verb form and vowel indicator

The first indicator is required and specifies the verb form: I, II, ... XV or Iq ... IVq. See Appendix:Arabic verbs for more information. Non-form-I verbs just specify the indicator directly, but form-I verbs normally need to specify a compound indicator with additional information given:

  1. Rarely, the weakness needs to be explicitly given following a hyphen, i.e. either I-assimilated or I-sound. This occurs only with form-I verbs whose first radical is و, and even then fairly rarely. Specifically, form-I verbs whose first radical is و are assumed to be assimilated (i.e. they lose the و in the non-past) unless the vowel pattern is i~a or u~u, in which case the verb is assumed to be sound (regular); in other words, the verb is assumed to be assimilated if the vowel pattern is a~u, a~i, a~a or i~i (or some weird pattern like i~u that is almost never seen and probably an error). This works correctly with almost all verbs, hence وَهَرَ (wahara, to frighten), non-past يَهِرُ (yahiru, he frightens) vs. وَهِلَ (wahila, to be frightened), non-past يَوْهَلُ (yawhalu, he is frightened). But a small number of verbs don't follow this pattern, and need the weakness explicitly given, such as وَسِعَ (wasiʕa, to be wide; to house, to accommodate), non-past يَسَعُ (yasaʕu, he is wide; he houses, he accommodates) rather than expected but non-existent #يَوْسَعُ (yawsaʕu). This verb would be specified as {{ar-conj|I-assimilated/i~a.pass.vn:سَعَة.ap:+}}.
  2. In most cases, the past and non-past vowel need to be given following a slash, in the format e.g. a~u for past vowel a, non-past vowel u; for example I/a~u. Multiple comma-separated vowels can be given for either type of vowel, e.g. a~u,i meaning past vowel a and non-past vowel either u or i; or a,u~u meaning past vowel either a or u and non-past vowel u. As mentioned above, more complex patterns (e.g. either a~u or i~a) should be specified using multiple conjugation templates. (It would be possible to extend the syntax to accomodate this but it would have to be displayed using multiple tables in any case, since the result of mixing all forms into a single table would be the equivalent of a,i~u,a, which would be extremely misleading.) It is possible to attach a footnote to a given vowel, e.g. a,u[rare]~u indicating that past vowel u specifically is rare; all forms involving this past vowel will bear the specified footnote. This is described below in more detail.

Passive indicator

Whether the verb has a passive voice and what sort of passive forms exist can be specified using one of the following passive-voice indicators:

  • pass: Verb has a full passive. Generally, all transitive verbs have a full passive. Verbs in Almaany that have a full passive are generally shown with the passive participle after the word مَفْعُول (mafʕūl).
  • ipass: Verb has an impersonal passive. Generally, intransitive verbs with prepositional objects have an impersonal passive. Verbs in Almaany that have a full passive are generally shown with the passive participle followed by a preposition + object (e.g. بِهِ (bihi), عَلَيْهِ (ʕalayhi)) after the word مَفْعُول (mafʕūl).
  • nopass: Verb has no passive. Verbs in Almaany that have a full passive are generally shown without any passive participle. (However, some verbs with full or impersonal passives are also missing passive participles, especially obsolete ones. Those with full passives contain example sentences with an object in the accusative, and those with impersonal passives either contain example sentences with prepositional objects or have the preposition listed after the lemma, i.e. past-tense, form of the verb.)
  • onlypass: Verb is passive-only, with a full passive.
  • onlypass-impers: Verb is passive-only, with an impersonal passive.

Root consonant indicator

For verbs with weak radicals, sometimes the radical cannot be inferred from the lemma form. In these cases, the radical can be explicitly given using one of the following indicators:

  • I:... to indicate the first radical; e.g. I:و to indicate that the first radical is و (w).
  • II:... to indicate the second radical; e.g. II:و to indicate that the second radical is و (w).
  • III:... to indicate the second radical; e.g. III:ي to indicate that the third radical is ي (y).
  • IV:... to indicate the fourth radical (only for quadriliteral verbs); e.g. IV:ي to indicate that the fourth radical is ي (y).

Radicals can be followed by transliteration after //. For example, the verb أَنْجَلَ (ʔangala) can be conjugated using III:ج//g, indicating that the third radical is ج, which should be transliterated as g.

Slot overrides

Individual slots can be overridden using an indicator slot:override,override,.... Here slot is the name of a slot, e.g. vn for the verbal noun, ap for the active participle, pp for the passive participle, past_1s for the first-person singular past, juss_pass_3fp for the third-person feminine plural non-past jussive passive, etc. This is followed by one or more comma-separated overrides (the Arabic comma ، can be used in place of a regular comma ,, but no space may follow the comma). An override can take the form arabic//translit to specify both the Arabic script and equivalent transliteration, but this is rarely necessary.

The most likely slots you will find occasion to use are vn for the verbal noun, ap for the active participle and pp for the passive participle. In particular, form I verbs have no default verbal noun and require such an override; form I stative verbs (those with past vowel i or u) have no default active participle and likewise require an override; and many form III verbs have a second verbal noun of the فِعَال (fiʕāl) form (appropriately adjusted for weak verbs), which requires an override to specify. To simplify specifying overrides, some shortcuts are available:

  • + works for all slots and explicitly requests the default value. (For this purpose, form-I stative verbs are considered to have a default participle of the فَاعِل (fāʕil) form, appropriately adjusted for weak verbs, like other form-I verbs.) This will throw an error for form-I verbal nouns, which have no default.
  • ++ works for some participles and verbal nouns and requests the "secondary" default value. Specifically, the secondary default for form-I active and passive participles is of the فَعِيل (faʕīl) form, and the secondary default for form-III verbal nouns is of the فِعَال (fiʕāl) form. Other slots and conjugation forms have no secondary default. Note that all default values are appropriately adjusted for weak verbs. This means, for example, that the فَاعِل (fāʕil) form automatically becomes فَائِل (fāʔil) for hollow verbs, فَاعٍ (fāʕin) for final-weak verbs, فَالّ (fāll) for geminate verbs, etc.
  • +++ for form-I active participles requests the secondary stative default value, of the فَعِل (faʕil) form.
  • +cd for form-I active participles requests the "color/defect" default value, of the أفْعَل form.
  • +an for form-I active participles requests the "-ān" (characteristic) default value, of the فَعْلَان (faʕlān) form.

You can freely mix shortcuts and fully specified overrides.

Each slot override can be followed by footnotes of the form [footnote] and/or inline modifiers of various sorts. The possible inline modifiers are:

  • <id:...>: Specify a sense ID to link to (see {{senseid}} and {{l}}).
  • <t:...> or <gloss:...>: Specify a gloss for the override.
  • <q:...>: Specify a left qualifier for the override. Rarely needed.
  • <qq:...>: Specify a right qualifier for the override. Rarely needed.
  • <l:...>: Specify one or more comma-separated left labels for the override. Rarely needed.
  • <ll:...>: Specify one or more comma-separated right labels for the override. Rarely needed.
  • <pos:...>: Specify a part of speech for the override. Rarely if ever needed.
  • <g:...>: Specify one or more comma-separated genders for the override. Rarely if ever needed.
  • <lit:...>: Specify a literal meaning for the override. Rarely if ever needed.

Additional parameters

Most of the information is encoded in |1=. The other parameters are:

  • |t=: Specify a gloss for this verb conjugation as a whole. Useful if there are multiple verb conjugations of a given verb (conjugation) form under a given lemma. Ignored except by {{ar-verb form}}, where it is displayed next to verb forms derived from that verb conjugation.
  • |id=: Specify an ID for this verb conjugation as a whole. Useful if there are multiple verb conjugations of a given verb (conjugation) form under a given lemma. Ignored except by {{ar-verb form}}, where it can be used to select a subset of conjugations for extracting forms, using |conjid= or |noconjid=.
  • |pagename=: Override the pagename of the verb lemma; normally taken from the actual pagename. Useful on testing and documentation pages.
  • |noautolinktext=1: Disable the autolinking algorithm for text outside of the verb itself, in multiword expressions.
  • |noautolinkverb=1: Disable the autolinking algorithm for the verb. If you use both this and |noautolinktext=1, no autolinking will happen, with the result that non-lemma forms will be linked in their entirety in the table.
  • |json=1: Instead of outputting a table, output a JSON representation of all the inflected forms. Used by bots.