User:A R King/private area/Basque verbs (DRAFT)

The verb is one of the most complex parts of Basque grammar. It is sometimes represented as a difficult challenge for learners of the language, and many Basque grammars devote most of their pages to lists or tables of verb paradigms. This article does not give an full list of verb forms; its purpose is to explain the nature and structure of the system.

Introduction: How complex is the Basque verb?
Even though Basque is best known to the international linguistic community as an oft-cited example of an ergative language, ever since the nineteenth century more has been written about the morphology of the Basque verb, whose conjugation has taken up most of the space in many traditional Basque grammars, than other aspects of the language. Furthermore, it has long been a commonplace that the Basque verb is very (even impossibly) difficult and complicated. However realistic, the impression that the Basque verb system is difficult to learn and complicated to describe can be explained as a cumulative effect, deriving from the following features:


 * 1) Many of the Basque verbal categories are expressed morphologically (i.e. through suffixes, prefixes or stem changes), giving rise to morphological complexity.
 * 2) Nevertheless, periphrastic (or compound) tense formations are so predominant in Basque verbal usage that a knowledge of these constructions and their part in the system is every bit as important as knowledge of the morphology of the simple tenses. (However, the morphology of auxiliary verbs is of course a necessary part of the periphrastic tenses.)
 * 3) The fact that up to three arguments (subject, direct object and indirect object) can be indexed morphologically on single verb forms is not only "exotic" for speakers of Indo-European languages, but makes for some very large paradigms and extensive paradigm tables.
 * 4) The existence in colloquial usage of further sets of synthetic allocutive forms to mark the social relationship between the speaker and the hearer leads to still more complex morphology.
 * 5) Not all elements and all combinations of elements in this system are used with equal frequency or are familiar to all speakers. Consequently, judgments may vary, depending on the theoretical or practical purpose, about which parts of the system to focus on and which to place first; the purposes of providing a pratical initiation or a comprehensive overview of the system as a whole may require conflicting emphases.
 * 6) Both the synthetic morphology and the periphrastic syntax are subject to some dialectal variation.

The rest of this article will outline these features of the Basque verb in greater detail.

Verb stems
One of the remarkable characteristics of the Basque verb is the fact that only a very few verbs can be conjugated synthetically (i.e. have morphological finite forms); the rest only have non-finite forms, which can enter into a wide variety of compound tense structures (consisting of a non-finite verb form combined with a finite auxiliary) and are conjugated in this way (periphrastically). Thus for example 'I come' is nator (a synthetic finite form), but 'I arrive' is iristen naiz (a periphrastic form, literally 'arriving I-am').

It should be noted that synthetically conjugated verbs like 'come' can also be conjugated periphrastically (etortzen naiz). In some such cases the synthetic/periphrastic contrast is semantic (e.g. nator and etortzen naiz are not generally interchangeable); in others the contrast is more a matter of style or register, or else of diachrony (some synthetic forms of conjugation are archaic or obsolete). A few synthetic forms occurring in twentieth-century Basque literature are even a posteriori extrapolations or back-formations of historically unattested forms, created for stylistic, poetic or puristic purposes.

Traditionally Basque verbs are cited using a non-finite form conventionally referred to as the participle (although not all its uses are really participial). Other non-finite forms can be derived from the participle, as will be seen in a later section. When the verb possesses synthetic finite forms, these are based on an ultimate stem (called the "basic stem" here) which is normally also present in the participle. For example, the verb etorri 'come' has the basic stem -tor- from which are derived both the participle etorri (with the non-finite prefix e- and the participle suffix -i) and the finite present stem -ator- and non-present stem -etor-.

The participle is generally obtained from the basic stem by prefixing e- or i- (there is no rule; if the stem begins with a vowel, j- is prefixed instead), and suffixing -i (to stems ending in a consonant) or -n (to stems ending in a vowel). Occasionally there is no suffix. The verbal noun stem, another non-finite form, is obtained by replacing the suffixes -i and -n (and also -tu or -du, see below) of the participle by either -tze or -te. A third non-finite form which we shall call the "short stem" is obtained from the participle by omitting any of these suffixes except -n, which is retained in the short stem in those verbs whose participle has it.

A larger number of Basque verbs have no finite forms, but their non-finite forms follow the same pattern described above (they show an e-/i-/j- prefix, and the participle ends in -i, -n or occasionally zero.

There is also another large group of verbs which again have only non-finite forms, in which the non-finite stem is unanalysable (as a verb, at least), thus there is no e-/i-/j- prefix. In most cases the participle of such verbs has the suffix -tu (-du if the stem ends in n or l). Occasionally we find zero or -i instead. This is replaced by -tze or -te in the verbal noun, and by nothing in the short stem. The stems of these secondary verbs may be (1) a nominal or other non-verbal stem (e.g. poz-tu, garbi-tu...), (2) a phrase (e.g. ohera-tu), (3) a Latin or Romance verbal stem (e.g. barka-tu, kanta-tu...) or (4) an unanalysable (primary) verb stem (e.g. har-tu).

Izan ('be')
The verb 'to be', the most common verb in the language, is irregular and shows some stem allomorphy in its finite forms. Its participle is izan.

Egon
Another verb, egon, is used in western dialects (and in writing) as a second verb 'to be' in a way similar to estar in Spanish.

Izan ('have')
The verb 'to have', also extremely common, also shows irregularities in its finite conjugation. In western and central dialects and in standard Basque, 'izan' is used as its participle, i.e. the same participle as for 'to be'; the two meanings are disambiguated by the context. Given that Basque verbs are conventionally cited in their participle form, this presents a problem for metalinguistic terminology, since "the verb izan" is ambiguous.

Ukan/*Edun
Eastern dialects avoid this ambiguity by using ukan as the participle of 'to have', reserving izan for 'to be', and some grammarians employ izan and ukan in this way for convenience, but this could create confusion since most Basque speakers do not actually employ ukan (or even know it as a metalinguistic term). Other grammarians refer to 'to have' as *edun, which is a hypothetical, unattested form derived from the finite stem -du-; again, the problem is that *edun does not exist in real Basque usage.

To avoid such problems, this article simply refers to "the verb 'to be'" and "the verb 'to have'".

*Edin, *Ezan
The two standard aorist auxiliaries (see below) lack any non-finite forms, and so also have no obvious citation forms. As with *edun, some grammars construct hypothetical participles based on the finite stems, referring to *edin (the intransitive aorist auxiliary) and *ezan (the transitive aorist auxiliary).

Eduki
There is another verb which also means 'have', at least in western dialects, namely eduki. As a lexical verb (rather than an auxiliary), many speakers and writers frequently use this verb. (This is somewhat reminiscent of, though not entirely parallel to, the Spanish distribution of haber and tener.)

Esan
The verb 'to say', esan possesses finite forms which have a different stem, -io- (e.g. diot 'I say'). Some grammarians treat these as different defective verbs, while others consider them a single word with stem allomorphy.

Tense structure and stem forms
Synthetic (single-word) conjugation involves the following finite "tenses":

Finite verbs have a basic finite stem which is either an unanalysable lexical root (e.g. -bil- 'go about, move (intr.)') or such a root preceded by the causative/intensive prefix -ra- (e.g. -rabil- 'cause to move, use'). From regular basic stems two tense stems are derived as follows: the present stem with prefix -a- and the non-present stem with prefix -e-, e.g. -abil- and -ebil- are the regular present and non-present stems of -bil-, -arabil- and -erabil- are the corresponding tense stems of -rabil-, and so on. The present stem is used in the present tense, the present potential tense and the non-third-person imperative, e.g. present d-abil 'he/she/it goes about', present potential d-abil-ke 'he/she/it may go about', second person imperative h-abil! 'go about!'. The non-present stem is used in the past and hypothetic tenses (non-potential and potential), and in third-person imperative forms, e.g. z-ebil-en 'he/she/it went about', ba-l-ebil 'if he/she/it went about', z-ebil-ke-en 'he/she/it might or would have gone about', l-ebil-ke 'he/she/it might or would go about', b-ebil! 'let hime/her/it go about!' (not in common use).

Non-present stems are further characterised by prefixes containing an n whenever the primary index (defined below) is non-third-person, e.g. z-ebil-en 'he went about' but n-enbil-en 'I went about', h-enbil-en 'you went about'; l-erabil-ke 'he would use it' but n-inderabil-ke 'he would use me'.

The suffix -(e)n is a marker of the past tenses, and -ke of the potential tenses (the past potential has both: -ke-en). The hypothetic non-potential tense usually occurs with the subordinator prefix ba- 'if', which will therefore be shown in examples; use of ba- is not restricted to the hypothetic, however (e.g. ba-dabil 'if he goes about', etc.). Apart from the tense markers mentioned, third person prefixes distinguish between present, past, hypothetic and imperative tenses, as will be seen below.

Synopses of two verbs are given in the following table as illustrations. The verb 'to be' (izan) is irregular but in extremely frequent use, since it also serves as an important auxiliary. The verb ibili 'go about, move, etc.' (root -bil-) is regularly conjugated, although not all its synthetic forms are in widespread use. This synoptic table shows third person forms.

Primary person indices
All conjugating verb stems (unless defective) can take the following set of person-indexing prefixes: n- (first person singular), h- (second person singular informal), g- (first person plural), z- (second person singular formal and second person plural). With intransitive verbs, these prefixes index the subject; with transitives, they index the direct object. For convenience, we shall refer to this as the set of 'primary person indices'.

The following table shows some examples of how these prefixes combine with verb stems to produce a wide range of finite verb forms.

Third-person forms
Third person verbs (here the 'person' again refers to the subject in intransitive verbs but the object in transitives) also take a prefix, which is invariable for number (singular or plural) but varies for tense, as follows: d- is used in the present tense, z- in the past, l- in the hypothetic and b- in third-person imperative forms (generally archaic or literary).

Some illustrative examples follow.

Plural marking
Plural number is marked in finite verbs in various ways, depending on the arguments whose plurality is being indexed. One set of plural forms are 'primary', that is, once again they refer to either the 'intransitive subject' or the 'transitive object'. The form of primary plural marking varies irregularly according to the verb stem, and may involve miscellaneous stem changes or the placement immediately after the singular stem of a plural marker (most commonly -tza or -z). Singular and plural forms of some finite verb stems are shown in the following table.

Primary plural marking occurs whenever the indexed argument (subject or direct object) is plural. The second person singular polite (pronoun zu) is also treated as plural for this purpose, although syntactically and semantically singular. To index the second person plural (pronoun zuek), in addition to the markers corresponding to zu a further ('secondary') plural marker -te is suffixed.

Ergative person and number suffixes
The ergative case is the case of subjects of transitive verbs. Such arguments are indexed in a different way from 'primary' arguments. Person of the ergative marker may be indexed in one of two ways: using suffixes or prefixes. The ergative-index plural marker is always a suffix (-te). The ergative person suffixes are as follows; those for the first and second person singular end in -a whenever another suffix morpheme follows them. The absence of an ergative suffix in transitive verbs (except those discussed in the next section) implies a third-person subject.

A few sample paradigms follow.

Ergative person prefixes
Instead of the ergative suffixes, ergative prefixes are used to index first or second person ergative arguments if the tense is non-present and the direct object is third person (see the gaps in the previous table). The ergative prefixes are identical to the primary prefixes in the singular, but in the plural -en- is added to the primary prefix forms:

The ergative plural suffix -te only occurs when required (a) to indicate the third person plural, or (b) to indicate the (real) second person plural.

Dative argument indices
Finite verbs that have an argument in the dative case also index the dative argument using the following set of dative suffixes (which are identical in form to the ergative suffixes except in the third person):

Both intransitive and transitive verbs may take dative indices, and the mechanism for incorporating these is the same in either case. Dative suffixes immediately follow the verb stem, preceding other suffixes such as the ergative suffixes (thus in d-i-da-zu 'you have it to me', -da- is the dative suffix and -zu is the ergative suffix) or the potential suffix -ke (as well as the past suffix -(e)n, which is always word-final).

Only the primary plural marker, if present, and the dative-argument marker precede the dative suffix. The dative-argument marker, whose regular form is -ki-, is added to basic verb stems to indicate that these are taking a dative argument. With -ki-, the primary plural marker always takes the form of -z- immediately preceding -ki-. A few verb stems have an irregular dative-argument form.

The most commonly used dative verb forms are those of the irregular verbs 'to be' and 'to have' which are in constant use as tense auxiliaries, when these verbs have no lexical meaning of their own. This is the reason why many of the glosses given below sound odd (e.g. dit 'he has it to me'); an example of a more natural-sounding use of this form as an auxiliary would be eman dit 'he has given it to me'. Nevertheless, the following table serves to clarify the morphological structure of dative-argument verb forms.

Familiar forms and allocutive indices (hika)
In colloquial Baque, an informal relationship and social solidarity between the speaker and a single interlocutor are expressed by employing a special mode of speech often referred to in Basque as either hika or hitano (both derived from hi, the informal second-person pronoun; in other places the same phenomenon is named toka and noka for male and female interlocutors respectively). The obligatory grammatical characteristics of this mode are:


 * The personal pronoun hi is used (rather than the polite second person singular pronoun zu).
 * All finite verb forms that index a second-person argument take (as one would expect) the corresponding hi forms, e.g. haiz 'you are' (rather than zara), duk or dun 'you have it' (rather than duzu), etc.:


 * Obligatorily in independent declarative clauses with finite verb forms not indexing a true second-person argument, an additional second-person index is incorporated. This is known as the allocutive construction, and we may refer to these second-person indices which do not refer to a syntactic argument of the verb as 'allocutive indices'.

The allocutive index may be incorporated in three different manners depending on the verb and the verb form in question. The first of these is only found with forms of the verb 'to be' (izan) without a dative argument, and consists of replacing 'be' with 'have': the subject of 'be' is indexed in the normal manner of a direct object (thus it remains 'primary'), while the ergative indices (normally referring to the subject) take on allocutive value, as in the following examples:

The second manner is limited to the verb 'have' itself, again without a dative argument. Such forms of 'to have' are replaced by the corresponding forms also employed when the verb has a dative argument, and the dative indices fulfill the allocutive function, as in the following examples:

In all other verb forms, the procedure is as follows: (1) an additional second-person suffix (-k / -a- for males, -n / -na- for females, the same as in the dative and ergative suffixes) is added (after a dative suffix but before an ergative one); (2) sometimes (there is considerable dialectal variation on this point), the third-person present-tense primary prefix d- changes to z- and/or the present-tense stem formant -a- changes to -ia- or -e- in the allocutive forms. e.g.

Eastern Basque dialects extend the allocutive system to the more polite form of address, zu (known as zuka or zutano), or the affectionate variant xu. The rules are similar. Such dialects have three levels of address: allocutive hi (with a male/female distinction) is the most intimate, allocutive zu or xu is polite but friendly, and the absence of allocutive constructions is the most neutral or formal. But most dialects lack the middle level.

Compound tense stem forms
Compound tense forms consist of a non-finite verb form (the compound tense stem) and a finite auxiliary form. We shall begin by looking at the non-finite stems. Each verb has four: the perfect, future, imperfect and short stems. The perfect stem is identical to the participle (see above). The future stem is obtained from the participle by adding -ko (-go after n). The imperfect stem is the verbal noun (see above) plus the suffix -n. The form of the short stem was discussed above. Some examples follow.

Compound tense auxiliaries
By combining the four compound tense stems with various auxiliaries, one obtains four groups of compound tense, sometimes referred to in Basque grammar as "aspects", which we shall call Imperfect, Perfect, Future and Aorist (= "aspect"-less) respectively.

The choice of auxiliary depends on the "aspect" and also on whether the verb is intransitive or transitive. Except in the Aorist tenses, the auxiliary for intransitives is the verb 'to be', while that for transitives is the verb 'to have'. In the Aorist a different pair of auxiliaries is used, one for intransitives and another for transitives. Since neither of the latter is used other than as an auxiliary, and neither has a participle (or other non-finite form) to provide a convenient citation form, we shall simply refer to them as the (intransitive and transitive) aorist auxiliaries.

The auxiliaries adopt all the argument indices (for subject, direct object and/or indirect object as the case may be, as well as the allocutive where applicable) that correspond to the verb within its clause.

The above diagram illustrates the patterns with auxiliaries in the present tense. However, the same auxiliaries may be used in a wide variety of tenses, not only in the present. The following two tables lay out synoptically the possible auxiliary/tense combinations for intransitive and transitive auxiliaries respectively.

Simple and compound tenses
The following are the most usual Basque tenses. By considering both simple and compound tenses as part of a single list, one can better see how the whole system fits together and compare the tenses with each other.

More periphrastic constructions
Some other constructions that commonly express a range of aspectual or modal notions show a greater degree of periphrasis than those considered so far. A brief selection of some of the most important of these are shown in the following table:

Non-finite verb forms
Basque verbs have a fairly wide range of non-finite forms. Morphologically these can all be derived via suffixation from the three non-finite forms presented at the beginning of this article: the participle, the verbal noun and the short stem. Apart from the short stem (which has a rather limited set of functions), all other forms are built on either the participle or the verbal noun.

The participle and derived forms
The participle and some other non-finite forms derived therefrom are as follows. To avoid repetition, mention will not be made of the use of the participle as a perfect stem in the formation of periphrastic tenses (see above).

The verbal noun and derived forms
The verbal noun and some other non-finite forms derived therefrom are as follows. Again, to avoid repetition, mention will not be made of the use of the -t(z)en form as an imperfect stem in the formation of periphrastic tenses (see above).

Compound verbs
Basque has a fairly large number of compound verbs of a type also known as light verb constructions, consisting of two parts. The first component is a lexical element which is often (but not always) an undeclined noun. The second is a common verb which contributes less semantic content to the construction but is the part that is conjugated, thus lending to the whole its verbal character. Details of conjugation depend on the light verb used, which may be one that has synthetic finite forms (e.g. izan), or a verb without synthetic finite forms (e.g. egin or hartu).

In synthetically conjugated light-verb constructions such as bizi naiz 'I live' or maite dut 'I love', care must be taken not to confuse the light verb (naiz, dut...) with tense auxiliaries; bizi naiz and maite dut are simple present forms, for example. The modal verbs nahi izan and behar izan are also of this kind. In the periphrastic tenses of compound verbs with izan, some contractions occur, e.g. in the future of bizi izan 'live', where we would expect bizi izango naiz for 'I will live', biziko naiz is more common, with -ko attached directly onto the lexical component bizi as if this were a verb.

Compound verbs, especially those with the light verb egin, offer an alternative way (besides direct derivation with -tu, as seen above) for incorporating new verbs into the language, either through the incorporation of onomatopeic words (kosk 'bite', oka 'vomit', hurrup 'sip', klik 'click'...) or of loanwords (dantza 'dance', salto 'jump' etc.) as lexical components.

Verbal particles
A small set of modal particles, including al, ote and omen only occur immediately preceding finite forms (i.e. in front of a synthetic finite form or the synthetic part of an auxiliary verb).

The only exception is that ote and omen are sometimes used in isolation where the ellipsis of a verb is understood. E.g. Egia ote? 'I wonder if it's true' is easily recognised by speakers to be an ellipsis of Egia ote da? Or if someone says Badator 'She's coming.' and someone else responds Omen! 'Supposedly!', this is as much as to say that the first utterance should incorporate omen, i.e. Ba omen dator 'Supposedly she is coming.'

Another set of preverbal particles consists of the affirmative particle ba- (by modern convention joined to a following finite verb form) and the negator ez. These are compatible with the modal particles, which they precede (e.g. ba omen dator in the preceding paragraph; ez al dakizu? 'don't you know?', etc.); apart from this, they too immediately precede the finite verb form.

Subordinator affixes
The forms of verbs cited throughout the general presentation of the finite verb system are normally those that occur in main clauses. (However, certain forms, such as the non-potential hypothetic, e.g. -litz, or the subjunctive, e.g. etor dadi-, never occur in such main-clause forms and these are therefore cited in subordinate forms such as balitz, etor dadin etc.)

In subordinate clauses, the finite verb takes a subordinator affix, i.e. a suffix or prefix which establishes (to some extent) the kind of subordination. Basically there are four such affixes, two suffixes and two prefixes, and one (and only one) of these is found in every subordinate form.

Both of the suffixes, however, may take further suffixes (mostly nominal declension suffixes) which serve to further specify the type of subordination. The following table provides a brief overview of some of the main uses and forms.