A Noun phrase is a phrase constructed around a Noun. In this sense, we will say that the Noun 'heads' its phrase. Noun phrases in Euskara have a very fixed word order, in contrast to the sentences, where phrases can be arranged in many different ways. Let us consider a few examples:
a. gure haur txiki-a
we-gen baby small-the
'our small baby'
b. neska gazte hau
girl young this
'this young girl'
c. Bilboko zazpi gizon
Bilbo-from seven man
'seven men from Bilbo'
d. azkarra den emakume-a
smart-the is-that woman-the
'the woman that is smart'
As we can see in (1a) and (1b), adjectives follow the Noun, and articles and demonstratives follow the [Noun + Adj] group (1a,b). Other modifiers, such as possesive phrases, postpositional phrases, relative clauses and most quantifiers, always precede the noun. Thus, for instance, in (1a), the possesive phrase gure 'our' appears before the Noun; in (1c), the postpositional phrase Bilboko 'from Bilbo', and the quantifier zazpi, 'seven', both precede the Noun gizon 'man'. Similarly, in (1d), the relative clause azkarra den 'who is smart' precedes its head Noun, emakume 'woman'. We can now consider a longer example, where more elements are combined:
atzo etorri ziren Ondarruko mutil alai haiek
yesterday arrived were Ondarru-from boy happy those
'those happy guys from Ondarru that came yesterday'
As you can easily see in this example, the word order in the Basque Noun phrase happens to be exactly the reverse of its English translation. Word-order within the Noun Phrase follows, rather strictly, the general tendency in Euskara to place heads of phrases in final position, while complements and other modifiers precede them. In other words, Euskara is a 'head-final' language. We can say that the Noun follows its complements and heads the Noun Phrase as illustrated in (3):
In the same fashion, since articles and demonstratives follow the Noun, as shown in examples in (1) and (2), we can say that articles and demonstratives, grouped under the common name of 'determiners', follow the Noun Phrase and head the Determiner Phrase, as in (4):
[[[emakumearen] argazkiNP] aDP]
woman-the-gen photo the
'the photograph of the woman'
However, a few items appear to break this head-final pattern: The Adjective follows the Noun for instance, as shown in previous examples. There is also a small subset of modifiers that can either precede or follow the Noun:
(I) modifiers with the morpheme dun, which denotes a possesed entity,
(II) modifiers ending in the morpheme tar, which denote geographic origin, and
(III) modifiers without the morpheme tar which also denote geographic origin.
They are all illustrated in the examples in (5):
a. dirudun emakumea
'a rich woman/ a woman that has money'
b. emakume diruduna
c. Gasteiztar laguna
'the friend from Gasteiz'
d. lagun Gasteiztarra
e. frantses liburua
'a french book'
f. liburu frantsesa
These modifiers can either precede or follow the Noun, without any difference in meaning. It must be noted that example (5e) can refer to a french book, but also more specifically to a book to learn French, while (5f) can only refer to a french book. That is, the string in (5e) can correspond also to a compund noun.
2.1. Noun phrases headed by common Nouns.
2.2. Noun Phrases headed by proper Nouns
Let us now start from the bottom of a Noun Phrase: the Noun. Regarding the types of Noun Phrases they build, we can distinguish two main kinds of Nouns, proper Nouns and common Nouns. They differ in their relation to Determiners:
(I) Noun phrases headed by common Nouns require Determiners, but
(II) Noun Phrases headed by proper Nouns don't occur with determiners.
We will discuss each of these generalizations now.
2.1. Noun phrases headed by common Nouns require Determiners.
a. *emakume gazte etorri da
woman young arrived is
b. emakume gaztea etorri da
woman young the arrived is
'the young woman has arrived'
The few exceptions to generalization (I) have to do with syntactic conditions external to the structure of the Noun Phrase, and they are overviewed in the section corresponding to Determiners, more specifically in the discussion on the determiner a (section 6.1.). One exception to be mentioned here, since it is not syntactically conditioned, is the case of common Nouns that can be used as proper Names: names of family relations like iseko 'aunt', can be used as proper Names and thus display no Determiner:
a. iseko ikusi dut
aunt seen have-I
'I have seen auntie'
It is important to note that Noun Phrases like (7) are grammatical only if they refer to the speaker's or the hearer's aunt or relative. The family relations that are subject to this treatment are the following: ama 'mother', aita 'father', osaba 'uncle', amama 'grandmother', aitite 'grandfather'. We use iseko 'aunt' for the illustration in the text because its ending is phonologically dintinct from the determiner -a. Nouns like errege 'king', or faraoi 'pharaoh', were also used as proper names in older stages of the language, when referring to one's king or pharaoh.
2.2. Noun Phrases headed by proper Nouns cannot appear with a determiner:
a. Irune etorri da
Irune arrived is
'Irune has arrived'
b. *Irunea etorri da
Irune-det arrived is
The only exception to generalization (II) has to do with the nature of the Noun Phrase itself. If a Noun Phrase is headed by a proper name which is used to to refer to a single individual, but to a group of individuals (i.e. a group of individuals with the name "Alex"), or it is used to refer to different stages of the existence of an individual as if the stages were actually different individuals, then that Noun Phrase can take Determiners and modifiers that are otherwise not possible for standard proper Nouns. From a descriptive point of view, we can say that, in these cases, the proper Noun is used almost as a commoun Noun. A few different types of examples are provided below: (a) Demonstratives and bat 'one', can cooccur with proper Nouns used as common Nouns, as shown in (9):
a. Nor da Alex hori?
Who is Alex that
'Who is that Alex?
b. Aspaldi ezagutu nuen Patxi bat
Long-ago known had-I Patxi one
'A Patxi that I knew long ago'
Even in these cases, however, proper Nouns never cooccur with the determiner a, as shown below in (10). In this respect, the behavior of this determiner is distinct from all other elements we have included in the category of Determiners, and it is also distinct from definite articles in indo-european languages, where it is common to have the definite determiner in these cases, as well as the demonstratives:
a. gure Alex
c. Guk miresten genuen Irune hura
we-E admiring had-we Irune that
'That Irune that we admired'
d.*Guk miresten genuen Irune-a
we-E adimiring had-we Irune-the
As the examples in (10) illustrate, even when they are used to refer to a group of individuals with the same name, proper Nouns cannot take the determiner a on them. Note that if instead of the proper Nouns Alex and Irune we put the common Nouns gizon 'man' and emakume 'woman' in the examples, then (10a) would be ungrammatical and (10b, c, d) would be grammatical. For a type of construction where the determiner a and proper Nouns appear to coocur, see discussion on examples (13), (14) and (15) below. (b) Proper Nouns can occasionally take the partitive marker, although this is not a common usage in the language:
a. ez dut Irunerik ikusi
not have-I Irune-part seen
'I have not seen (any) Irune'
'I have not see Irune (at all)'
b. Josu Anderrik ikusi duzu?
Josu Ander-I seen have-you
'Have you seen (any) Josu Ander?'
'Have you seen Josu Ander (at all)?'
c. Bada Garikoitzik hemen
yes-is Garikoitz-part here
'There are Garikoitz here'
'There is (someone named) Garikoitz here'
As the English translations try to convey, these sentences can be used to refer to groups of individuals with the same name, or to talk about a single individual, in which case the second translation is more accurate. (c) Quantifiers can coocur with proper nouns used as common Nouns:
a. Alex asko
b. Zenbait Irune
Quantifiers that require the determiner a can also coocur with proper Nouns. In these cases, the determiner a seems to appear in the same constituent as the proper Noun (13):
a. Alex guzti-ak
'All (of) the Alexes'
b. Irune gehien-ak
It is likely that the Nouns in (13a, b) are complements of the quantifiers, which would account for the fact that the quantifiers appear following the Noun rather than preceeding it (remember the discussion of examples in (1) and (2)). It is significant that also in English, these constructions appear to take the Noun as complement of the Quantifier: 'All [(of) the guests]', 'most [of the guests]'. The example (13b) is a superlative construction, where arguably the quantifier takes the preceding Noun Phrase as a complement. Interestingly, there is a further difference between common Nouns and proper Nouns that surfaces in this construction:
a. emakume(rik) gehi-en-ak
'most (of the) women'
b. Irune(*rik) gehi-en-ak
whereas comon nouns can optionally take the partitive marker, this is not possible for a proper Noun in this construction.
(d) There are two other constructions where a proper Noun used as a common Noun appears to take the determiner a. They are illustrated in (15):
a. Peru gure-a
b. Alex gaixo-a
'the poor Alex'
The example in (15a) has a posessor following the Noun (a non-canonical order, since possesors typically precede the Noun, as shown in example (1a)). The posessor has the determiner a attached. It is likely that the determiner in (15a) is not attached to the Noun Phrase headed by the proper Noun Peru. Rather, this construction is probably best analized as containing an appositional Noun Phrase, which is atributive to the preceding Noun Phrase that contains the proper Noun. The structure would then be like (16):
Assuming this to be the constituent structure of the construction, a number of its distinctive properties could be accounted for: (a) It is impossible to have the determiner a with a proper Noun if the possesor appears in its canonical place, preceding the Noun, as showh in (10b); (b) Only the determiner a can appear following the atributive. Demonstratives and indefinite articles are barred from this construction, as shown in (17):
a. *Peru gure hau
Peru our this
'This Peru of ours'
b. *Peru gure bat
Peru our one
'A Peru of ours'
c. gure Peru hau our Peru this
'This Peru of ours'
d. gure Peru bat
our Peru one
'A Peru of ours'
The equivalents of (15) in English and Spanish have the structure in (17): 'Mary [the great]', *'Mary [this great]', 'Pedro [el grande]', *'Pedro [este grande]', where the definite article does not correspond to the proper Noun, but rather to the atributive that follows, and where any determiner other than the definite article yields ungrammaticality. The example in (15b) constitutes the best case for a Noun Phrase headed by a proper Noun bearing a determiner a. Compare (18a) and (18b), with (10b) and (10d):
a. Gure Alex txiki-a
our Alex little-the
'Our little Alex'
b. guk miresten genuen Irune gazte-a
we admiring 4E-havepst-Comp Irune young-the
'The young Irune we admired'
It is not clear how examples like (15b), and (18a,b) could be argued to have a structure similar to (15a), where the determiner and the possesor make a constituent leaving out the proper Noun. These examples do seems to constitute genuine cases of proper Nouns heading Noun Phrases that have the determiner a attached to them. As shown by the data, the presence of the Adjective is required to make the construction grammatical. Summary: proper Nouns almost never coocur with the definite article a in Euskara. Only when used as denominators of a set of elements can proper Nouns coocur with demonstratives or the indefinite article. Even in this later case, proper Nouns resist cooccurrence with the determiner a.
2.3. Gender. There is no grammatical gender in the nominal system. The only area of Euskara grammar where gender morphology can be found is the familiar treatment in the verbal morphology. Nouns and adjectives have no distinct endings depending on gender.
In modern spoken language, and only in Western dialects, there can be found some instances were the gender endings of Spanish have been kept in borrowings and a distinction is made between masculine and femenine, but this is a rather modern and marginal phenomenon.
3.1. Adjectives and their complements.
As we have already seen, adjectives always follow the Noun in Euskara(1) (19):
a. zaldi zuri hau
horse white this
'this white horse'
b. *zuri zaldi hau
There is no word-order difference between apositive and atributive adjectives of the sort found in certain Romance languages; both atributive and apositive adjectives follow the Noun. More than one adjective can appear following the noun:
zaldi zuri txiki argal hau
horse white small thin this
'this thin small white horse'
3.1. Adjectives and their complements. Adjectives appear to have some difficulty taking syntactic complements in Euskara, that is, they do not easily take complements with which they make a separate constituent. For instance, constructions like 'a woman [proud of her work]' or 'a man [fond of his paintings]' are not possible:
*emakume [bere lanaz harro]a
woman her work-ins proud-the
(A woman proud of her work)
Rather, Euskara resorts to relative clauses or participial infinitivals in those cases:
bere lanaz harro dagoen emakumea
his work-ins proud is-3A-Comp woman
'A woman who is proud of his work'
Although we are far from being able to suggest a reason why bare adjectives in Euskara are unable to have complements, it is intriguing to relate this fact with a well known restriction on the distribution of Adjectival phrases in English. Although the canonical position of the Adjective in English is prenominal, Adjectives with complements are banned from that position:
a. a proud (*of his son) father
b. a father proud *(of his son)
With respect to the placement of phrasal heads, Euskara and English are mirror images: English is head-initial and Euskara is head-final. Note that adjective placement is also the mirror image: they are prenominal in English, and postnominal in Euskara. Since the equivalents of English Adjectival Phrases appear as prenominal complements of Euskara (although they never display the bare Adjective as head), it is possible that at a deeper level of analysis both phenomena will turn out to be the same constraint.
3.1.1. Participials. Participials can function as adjectives in Euskara, as the examples in (24) illustrate:
a. hosto eroriak
b. itsaso harrotua
'the swollen sea'
In these cases, the participial behaves exactly like and adjective, that is, it follows the Noun and precedes the Determiner. Participials in adjectival position such as the ones in (24) can hardly constitute clauses with complements. However, especially in eastern dialects, construction of the sort in (21) are possible to a certain extent with adjectival participials. Thus, for instance, examples like the ones in(25) are acceptable in these dialects, according to EGLU (1991) and Euskaltzaindia (1993):
a. haur ongi ikasi batzu
child well learned some
'some well learned children'
b. bere seme gudan hilei
her son war-in dead-det-D
'to her sons dead in the war'
Participial clauses, that is, clauses headed by a participial form of the verb, precede the Noun like all other complements, as shown in (26):
a. [lurrean eroritako] hostoak
ground-in fallen-ta-ko leave-detpl
'the leaves fallen on the ground'
b. [harrak jotako] sagarra
worm-det-E bitten-ta-ko apple-det
'the apple eaten by the worm'
as we can see in the examples, these participial clauses have the ending ta and to this ending the mofpheme ko is attached. Therefore, they belong tot he class of ko phrases, which are discussed immediately below.
4.1. ko and ren phrases.
4.2. beste 'other'.
Uner this heading, we willl group all other phrases that appear in the Noun Phrase. There are basically two big groups, the first of which we will name 'ko and ren phrases', the second one includes the modifier beste 'other'. The modifiers in both groups must precede the Noun.
4.1. 'ko' and 'ren' phrases.The morpheme ko can indicate location, and this is why it is sometimes referred to as a 'locative genitive', but as we will see location is not the only relation it can convey. However, one general guideline that is helpful in distinguishing the use of ko and ren phrases involves location: ko is attached to phrases that denote location, or phrases that denote a property. All other relations a phrase may bear with respect to a Noun are dealt with by means of the morpheme ren. Let us consider a few examples in detail. Consider first the examples in (27):
a. [etxe handiko] leihoak
house big-ko window-detpl
'the windows from/in the big house'
b. [margolari ezagun horren] erretratoa
painter known that-ren portrait-det
'that (well)known painter's portrait'
In (27a), the relation between the big house (etxe handi) and the windows (leihoak) is one of location: the big house is the place where the windows are located. That is why the big house is placed in a ko phrase. In (27b), the relation between that well-known painter (margolari ezagun hori) and the portrait (erretratoa) is not one of location. That is why the well known painter is placed in a ren phrase. The relationship expressed by ren in (27b) can be either: (a) posession, where the painter owns the portrait in question, regardless of who painted it, or (b) agency, where the painter is the author of the portrait, but not necessarily the owner, or (c) theme, where the painter is the entity portraid in the portrait, which may have been painted by someone else, and owned by someone else as well. To continue in a little more detail with ko phrases, let us add that they can also relate a property with the head noun. Thus for instance, in (28):
a. [jenio biziko] neska
temper lively-ko girl-det
'a quick-tempered girl'
b. [bihotz oneko] mutila
heart good-ko boy-det
'a good-hearted boy'
the two examples illustrate ko phrases that convey properties which are predicated of the head Nouns, not locative relations.
Included in the predicative type of ko phrases are the examples involving participial clauses, like the ones illustrated previously in example (26). Participials can carry one of the two endings ta or rik that form resultatives. These resultative phrases can then take the ending ko and become modifiers of the Noun, as we further illustrate in (29):
a. [atzo Bilbon erositako] eraztuna
yesterday Bilbo-in bought-ta-ko ring-det
'the ring bought yesterday in Bilbao '
b. [txoriak kabira ekarririko] abarra
bird-det-E nest-to brought-rik-ko branch-det
'the branch brought to the nest by the bird'
In (29a), we have ko phrase containing a resultative phrase headed by ta, which in turn contains a participle erosi 'bought', and a locative and a time adverbial belonging to it. In (29b), we see a ko phrase containing a resultative phrase headed by the morpheme rik, with contains a participle ekarri 'brought' and its subject argument txoriak 'the bird (ergative case)', plus a postpositional complement kabira 'to the nest'. Ordinal numerals can be considered a subtype of ren phrases. They are headed by the morpheme garren. See more about ordinals in the section devoted to numeral quantifiers. Ocassionally, ko and ren phrases can appear apposited to the right of the Noun phrase, particularly if it is headed by a proper name. This type of construction has been considered in example (15), within the discussion of proper Nouns.
4.2. Other: beste. This word, beste 'other', 'another', does not naturally fall under any of the categories we have considered so far. Distributional facts discriminate it from quantifiers, determiners and other complements of the Noun. Beste must always precede the Noun, never follow it:
a. beste neska alai bat
other girl happy one
'another happy girl'
b. *neska alai beste bat
girl happy other one
Beste attaches to a full Noun Phrase, which must conform to the description provided so far. When the Noun Phrase is complete, beste can always appear at the beggining of it:
a. beste Bilboko hiru neska alai hauek
other Bilbo-from three girl happy these
'These other happy three girls from Bilbo'
beste must always precede the numeral quantifier if there is one. But it seems that it can either precede or follow the ko and ren phrases.
a. amaren beste Bilboko hiru lagunak
mother-gen other Bilbo-from three friend-detpl
'mother's other three friends from Bilbo'
b. amaren Bilboko beste hiru lagunak
d. beste amaren Bilboko hiru lagunak
e. *amaren Bilboko hiru beste lagunak
The modifier beste must also be placed following relative clauses, which, in general, tend to be the most external modifier of the Noun:
a. [etorri den] beste lagun hori
arrived is-comp other friend that
'that other friend that has arrived'
b. *beste [etorri den] lagun hori
It appears therefore that beste can be placed in between ren complements and ko complements, but always preceding, that is, taking scope over numeral quantifiers. To see this, consider now the word order alternations between ko phrases and quantifiers:
a. hiru Bilboko lagun
three Bilbo-from frien
'three friends from Bilbo'
b. Bilboko hiru lagun
As shown in (34), a ko phrase can either follow or precede a numeral quantifier. It has been shown in (32), that beste can either follow or precede the ko phrase. However, if we include the modifier beste in a Noun phrase like the one in (34), as we have done in (35), we see that beste, unlike Bilboko, cannot follow the quantifier:
a. Bilboko beste hiru lagun
Bilbo-from other three friend
'Three other friends from Bilbo'
b. beste Bilboko hiru neska
c. *hiru beste Bilboko neska
d. beste hiru Bilboko neska
When the Noun is deleted, the resulting paradigm regarding complement distribution changes slightly. As shown in (36), if the Noun is deleted, the modifiers and determiner, if there is one, stay put, and no extra element needs to be added (unlike the case of English, as you can judge from the translations):
'the other one'
b. beste hiru
c. beste hiruak
'the other three'
But if we delete the Noun in a Noun phrase containing beste and either a ko phrase or a ren phrase, then beste must follow, not precede:
a. Bilboko besteak
b. *beste Bilbokoak
c. Mariren bestea
d. *beste Marirena
5.1.3. A partial list of numerals
5.2. Quantifiers that require a determiner.
5.3. Determinerless quantifiers.
We will divide quantifiers into three groups, depending on their ability to cooccur with a determiner:
(5.1) Numeral quantifiers;
(5.2) Quantifiers that require a determiner;
(5.3) Indefinite/determinerless quantifiers;
Regarding their distribution within the Noun Phrase, numerals in group (5.1), precede the Noun except for bat 'one', which must follow, and bi 'two', which must follow the Noun only in Western varieties of Euskara. In group (5.2) there are only two quantifiers, and both of them must follow the [Noun+Adjective] group. Finally, in group (5.3), most quantifiers precede the noun while a few of them must follow the [Noun+Adjective] group, as the examples below will illustrate.
5.1. Numerals. Numeral quantifiers must appear in determinerless Noun Phrases when they are indefinite (38a), and with determiners when they are definite (38b, c). In this latter case, Noun phrases containing numeral quantifiers can accept either the determiner a or a demonstrative.
a. Hiru txori
b. hiru txoriak
'the three bird'
c. hiru txori hauek
three bird thispl
'these three bird'
As mentioned above, numerals precede the Noun. The only exception is the numeral bat 'one', which must follow the [Noun+Adj] group in all dialects, and the numeral bi 'two', which also follows the Noun in western varieties of Euskara (39c), but it patterns like all other numerals in all other dialects (39d):
a. txori bat c. txori bi
bird one bird two
'one bird' 'two birds'
b.*bat txori d. bi txori
one bird two bird
5.1.1. Ordinal quantifiers are built by attaching the suffix garren to the cardinal (40a,b), except for the case of 'first', which is not *batgarren but it is independently formed as lehen, lehenengo or lehendabiziko, depending on the dialect. Ordinal quantifiers always precede the Noun, even in the case of ordinals for 'first' and 'second', in all varieties of the language, as shown in examples (40c,d):
a. zazpigarren etxea
'the seventh house'
b. hirugarren leihoa
'the third window'
c. lehenengo etxea
'the first house'
d. bigarren leihoa
'the second window'
As discussed previously in example (29), ordinal quantifiers can be thought of as a subclass of ko and ren phrases. It is therefore not surprising that Noun phrases containing ordinal quantifiers must have a determiner. There is no definite/indefinite contrast like the one in (38), depending upon the presence of the determiner. As the examples in (40) already illustrate, the presence of the determiner is required, and its absence yields ungrammaticality.
5.1.2. Distributive quantifiers are built by attaching the suffix na to cardinal quantifiers: thus for instance, from sei 'six', the distributive seina 'six each' can be constructed, or from the cardinal bi 'two', the distributive bina 'two each'. Distributive quantifiers are placed wherever the cardinal they are formed upon is placed. Thus, the distributive bana 'one each', formed upon bat 'one', always follows the Noun, while seina 'six each', mentioned above, behaves like its cardinal sei and precedes the Noun:
a. seina musu
'six kisses each'
b. musu bana
'one kiss each'
Noun phrases containing distributive quantifiers do not take determiners, as shown by the ungrammaticality of (42):
These distributive quantifiers are rather interesting when compared to Germanic or Romance languages. In English, for example, the translations of (41a, b) involve what is referred to as 'Binomial each' (Safir & Stowell (1988)), which is dependent not on the Noun Phrase being distributed but rather on the Noun Phrase that is the recipient of the distribution:
(i) [NP2The women] bought [NP1three books] [NP2each]
(ii) [NP2The women] [NP2each] bought [NP1three books]
In Euskara, the distributive suffix is attached to the quantifier of the Noun Phrase being distributed, as the examples in the text show. Moreover, the quantifier the suffix attaches to belongs in the distributed Noun Phrase and cannot be placed anywhere else; that is, it is not of a 'floating' kind. Noun Phrases containing the ditributive quantifier are constrained to appear in environments similar to reciprocal pronouns like elkar 'each other'. Thus, for instance, they cannot appear in subject positions:
*emakume binak gu ikusi gaituzte
woman two-each-erg we seem us-have-they
('two women each have seen us')
5.1.3. A partial list of numerals in Euskara. Here we provide a list of numerals in euskara, as well as instructions to construct numerals. First, let us consider the numbers up to twenty:
Twenty is an important number; it is the base to construct numerals up to a hundred. Thus, the numbers from twenty one up to forty are built repeating the list above after twenty, like this:
21 hogeita bat
22 hogeita bi...
and when we reach the munber thirty, we add twenty and ten:
30 hogeita hamar
31 hogeita hamaika...
up to number forty, which is something like 'again twenty':
Up to sixty, we keep on adding the numbers from one to twenty after the number for forty, like this:
41 berrogeita bat
42 berrogeita bi...
so you already know that fifty is 'forty and ten':
50 berrogeita hamar
51 berrogeita hamaika...
Now all you need to know is the numbers for sixty and eighty, which are translatable as 'three twenties' and 'four twenties', respectively:
So we are done up to ninety nine, which is, as you know now, laurogeita hemeretzi. Some numbers beyond this one are:
100 ehun 1.000 mila
200 berrehun 2.000 bimila
300 hirurehun 3.000 hirumila
400 laurehun 4.000 laumila...
500 bostehun 1.000.000 milioi bat
600 seiehun 2.000.000 bi milioi/milioi bi
700 zazpiehun 3.000.000 hiru milioi
And finally, a few examples for your practice:
2001 bi mila eta bat
1984 mila bederatziehun eta laurogeita lau
666 seiehun eta hirurogeita sei
77 hirurogeita hamazazpi
5.2.Quantifiers that require a determiner. The quantifiers guzti 'all', bakoitz 'each', and gehien 'most' require the presence of a determiner, as shown in (44):
a. haur guzti-ak
'All the children'
b. lur guzti-a
'All the earth'
c. haur bakoitz-a
d. *haur bakoitz-ak
e. haur(rik) gehienak
As shown in (45e), the Noun preceding the quantifier gehien 'most' can carry the partitive marker, which indicates that gehien takes an entire determiner phrase as its complement, ([[DP] DP]) much in the way of the examples in (47) with guzti 'all. There are two main differences between guzti 'all' and bakoitz 'each': guzti 'all' can take plural or singular determiners (44a, b); it can also take demonstratives as a determiners (45a, b). On the other hand, bakoitz 'each' does not accept plural determiners (44d) or any kind of demonstratives, (45c,d):
a.haur guzti hauek
child all thispl
'All these children'
b.lur guzti hau
earth all this
'all this earth'
c.*haur bakoitz hau
child each this
d.*haur bakoitz hauek
child each thispl
The only way in which demonstratives can be made to coocur within the same Noun Phrase as bakoitz 'each' is by resorting to a partitive construction, as shown in (46):
a. hauetariko haur bakoitza
thispl-part-gen child each-the
'each of these children'
In this construction, the demonstrative hauek 'these' takes the partitive marker rik and the ending ko, becoming a ko phrase that acts as a complement of the Noun haur 'child'. Regarding the coocurrence of the universal quantifier guzti 'all' with demonstratives, it must be noted that (45a,b) are not the only choices. There is another usage, which is also current and in fact prevalent in written records, illustrated in (47):
a. haur hauek guztiak
child these all-detpl
'all these children'
b. lur hau guztia
earth this all-det
'all this earth'
In these examples, the quantifier appears apposed after the Noun phrase. This is why there are two determiners: the demonstrative heading the first phrase, which takes the Noun in it, and the determiner a, in its plural form in (47a) and in its singular form in (47b), constituting the second phrase, which take the quantifier in it.
5.3. Determinerless quantifiers. The group of quantifiers that never take determiners is the group of indefinite quantifiers. Most of them cannot cooccur with any determiner:
a. zenbait gizon
b. ume asko
The examples in (48) illustrate the indefinite quantifiers zenbait 'some' and asko 'many'. While the first one precedes the Noun, the other one follows it. None of them accept the presence of a determiner, as shown in (49):
a. *zenbait gizon-ak
b. *zenbait gizon-a
d. *ume asko-ak
f. *ume asko-a
It must be noted that, contrary to English, the quantifier zenbait can only quantify count Nouns, not mass Nouns, whereas the quantifier asko can quantify both over count and mass Nouns, that is, asko can be translated both as 'many' and as 'much', if, for example, it modified a mass noun such as gari 'wheat' (gari asko 'much wheat', 'a lot of wheat'). Other quantifiers that do not allow the presence of a determiner are: hainbat 'many, much', gutxi 'few, little', edozein, zeinnahi, whichever, oro 'all', and the interrogatives zenbat, 'how many', 'how much' and zein, 'which'. Examples containing all these quantifiers are given in (50):
a. hainbat aburu
b. irudimen gutxi
c. ikasle gutxi
d. edozein gona
e. herrialde oro
f. zenbat lagun
how many friend
'how many friends'
g. zein esku
In these cases, the presence of the determiner, regardless of number or syntactic environment, induces ungramaticality. The only possible exception to this statement is the quantifier gutxi 'few', which can take the plural indefinite determiner batzuk 'some, ones', as in the following example:
ikasle gutxi batzuk
student few ones
'a few students'
this quantifier can occasionally take the determiner a as well, in eastern varieties, but in this case it is not the equivalent of English 'a/the few...' but rather, a free variant of the down-entailing quantifier 'few' illustrated in (50c).
6.1. The determiner 'a'.
6.1.1. Definite Noun Phrases.
6.1.2. Indefinite environments.
220.127.116.11. Generic sentences.
18.104.22.168. Indefinite objects and subjects.
22.214.171.124. Existential or presentational sentences.
6.2.1. Emphatic demonstratives.
6.2.2. Demonstratives used as pronouns.
Within the class of determiners, we will group the demonstratives and the determiner a. This is the category that must appear last in the order of elements in the Noun Phrase. The Determiner takes the entire Noun Phrase as its complement, constituting the Determiner Phrase. That is to say, there is only one determiner corresponding to each Noun phrase, as the examples in (52) illustrate:
a. [irakurle gazte]a
'the young reader'
b. [asto zahar] hau
[donkey old] this
'this old donkey'
With the exception of proper Nouns discussed above, Noun Phrases in Euskara present a strong tendency to be headed by an overt Determiner; that is, there are no instances of 'bare plurals' and hardly any instances of 'bare nouns'. A second type of Noun phrase that displays no final determiner is constituted by those containing indefinite quantifiers, which have been discussed in the previous section. We start in 6.1. with an overview of the usage of the determiner a, which is perhaps the most intricate one in this group. Next, in 6.2., we will discuss demonstratives.
6.1. The determiner 'a'. This determiner is also called 'article' in many descriptions of Euskara. As we will see in some detail, it is used in all environments where a definite article is required, but its usage goes well beyond the definite article, since it also heads generic and indefinite Noun Phrases, as well as some predicative phrases even when they do not contain Nouns. The determiner a appears to be the unmarked determiner, in the sense that it often surfaces in environments where other languages display determinerless Noun Phrases.
When the determiner a is added to a word that ends in the vowel a, the two vowels fuse into one: gona 'skirt' gona+a > gona 'a/the skirt'. Let us consider the distribution of this determiner:
6.1.1. Definite Noun Phrases. The determiner a is used to convey definiteness, in Noun phrases containing common Nouns:
a. goizeko izarra
'the morning star'
b. Euskal Herriko lehendakari-a
Basque Country-of president-the
'the president of the Basque Country'
Definite Noun phrases headed by proper Nouns or pronouns do not allow the presence of the determiner a.
6.1.2. Indefinite environments. There are many other syntactic environments where this determiner is used despite the fact that the phrase it heads is not definite. Among those cases we find the following:
126.96.36.199. Atributes. Many predicative atributes in Euskara require the determiner a. Predicates in copular sentences like the ones in (54), for instance, require the determiner a:
a.Josu on-a da
Josu good-det is
'Josu is good'
b.Miren irakasle-a da
Miren teacher-det is
'Miren is a teacher'
c.liburu hau interesgarri-a iruditzen zait
book this interesting-det seems is-to me
'This book seems interesting to me'
188.8.131.52. Generic sentences always require the determiner a, whether their subjects are singular or plural:
a.Edurr-a zuri-a da
snow-det white-det is
'Snow is white'
b.Txakurr-ak ugaztun-ak dira
dog-detpl mammal-detpl are
'Dogs are mammals'
184.108.40.206. Indefinite objects and subjects, which can often appear determinerless in many languages, also require the determiner a:
a.Guk arto-a erein dugu
we-E corn-det planted have-we
'We have planted corn'
b.Zuek sagarr-ak jan dituzue
You-E apple-detpl eaten have-you
'You have eaten apples'
c.Kamioi-ak etorri dira
truck-detpl arrived are
'Trucks have arrived'
There are no cases in Euskara were objects can appear as bare Noun Phrases, regardless of number. In the realm of subjects, there is formally no difference between unaccusative and intransitive predicates in the sense of Perlmutter (1978), in that both require the determiner a. However, a clear difference can be found when looking at the interpretation of the Noun Phrases involved: whereas objects of transitive verbs and subjects of ergtive predicates ( that is, all arguments bearing an initial 2 relation, in Relational Grammar terms, or being D-Structure complements of the Verb as in Burzio's (1986) work within the Government and Binding framework), are subject to an existential interpretation, the subject of the unergative predicate cannot receive such existential interpretation; rather, it must be interpreted as either definite or universal. Considering the examples in the text, this means that (56a, b, c) can be naturally interpreted as 'some corn', 'some apples' and 'some trucks' repectively, but (57) cannot be interpreted as 'some men':
gizonek negar egin dute
men-det-E cry made have
'the men have cried'
It is only interpretable as 'all men' or 'the men', which are truth-functionally equivalent.
220.127.116.11. Existential or presentational sentences, which involve an indefinite subject, bear the determiner a:
a.Bada ogia mahai gainean
yes-is bread-det table top-in
'There is bread on the table'
b.Badira sagu-ak etxe honetan
yes-are mouse-detpl house this-in
'There are mice in this house'
These existential sentences can also display partitive case, but the presence of the determiner a is also possible, maintaining the indefiniteness of its Noun phrase.
6.2. Demonstratives. There are three demonstratives, hau 'this', hori 'that' and hura 'that (further)'. The demonstratives indicate varying degrees of proximity in either real or figurative space or time. Thus, hau is closer to the speaker, hori is closer to the entity addressed, and hura is not close to any of them.
zuhaitz hau 'this tree'
zuhaitz hori 'that tree'
zuhaitz hura 'that tree (farther)'
The plural forms of the demonstratives are:
zuhaitz hauek 'these trees'
zuhaitz horiek 'those trees'
zuhaitz haiek 'those trees (farther)'
6.2.1. Emphatic demonstratives. Demonstratives have an emphatic form, which is constructed adding the morpheme xe, as shown in (59):
a. hauxe da irakurri dudan liburua
this-xe is read have-I-that book-det
'this is the book I've read'
b. horixe esan dut nik
that-xe said have-I I-E
'that (is what) I have said'
c. zuhaitz huraxe da aititek landatu zuena
tree that-xe is grandfather-E planted had-that
'It is that tree that grandfather planted'
there is a certain degree of variation among dialects regarding whether the morpheme xe is added before or after the case ending or the postposition if there is one, with the exception of ergative and genitive, where the morpheme is always inserted after the demonstrative but before the case ending:
a. honexek/horrexek/harexek ekartzen du egunkaria
this-xe-E/that-xe-E/that-xe-E brings has newspaper-det
'this one/that one/that one brings the newspaper'
b. honexen/horrexen/harexen arraina da freskuena
this-xe-gen/that-xe-gen/that-xe-gen fish is freshest-det
'this one's/that one's/ fish is the freshest
In the case of the dative case and other postpositions, some varieties of the language insert the xe morpheme before the case or postposition, whereas others insert it after, as the contrast between (61a) and (61b) illustrates for the dative:
a. hone-xe-ri erosi diot arraina
this-xe-D bought have fish-det
'(it is) to this one (that) I have bought the fish'
b. hon-i-xe erosi diot arraina
this-D-xe bought have fish-det
'(it is) to this one (that) I have bought the fish'
Emphatic demonstratives are most naturally used when the phrase headed by the demonstrative is the galdegaia of the sentence.
6.2.2. Demonstratives used as pronouns. There are no distinct forms for third person pronouns in Euskara, and demonstratives are used as third person pronominals.
When considering the use of demonstratives as third person pronouns, there is a third series of demonstratives that has special relevance. This series is formed adding the prefix ber 'same, again' to the demonstrative: berau, berori, bera (this third one is formed not by the combination of ber and hura, but rather by the combination of ber and a, which is a variant of hura, still in use in western varieties of Euskara). They are used anaphorically, that is, when the entity they referred to is already known in the discourse. The second one in the series, berori, is still nowadays used as a very polite form of second person singular, to address authority figures, such as a priest, or a doctor, among others (62a). The third one in the series, bera, is very frequently used as third person pronoun (62b), alternating with the third demonstrative hura (62c):
a. berori joango da?
that go-irr is
'will that one go?'
'will your honour go?'
b. bera etorri da
that arrived is
c. hark ikusi nau
that-E seen me-has
'she/he has seen me'
The criteria that determine when to use bera and when to use hura are rather complex and vary from dialect to dialect. We can mention a few of them, (listed also in EGLU):
(I) if the antecendent and the demonstrative are in the same sentence, bera must be used, and not hura
(II) if the antecedent and the demonstrative are not in the same sentence, eastern varieties prefer hura whereas western varieties prefer bera if the antecedent has been mentioned. This western usage has two exceptions:
(a) if the sentence containing bera contains another Noun phrase that could count as its antecedent, hura is preferred even if its antecedent has beenmentioned previously in another sentence;
(b) if the antecedent of the demonstrative belongs to a group that has been mentioned.
7.1. Number and Determiners.
7.2. Proximity plural determiner ok.
Singular is the unmarked case, and only plural is marked overtly. Hence, the best way to think of the category number is to consider it in terms of a binary category [+plural] versus [- plural].
7.1. Number and Determiners. The specification for number in the Noun Phrase belongs in the Determiner category and it is morfologically inseparable from it. Therefore, determinerless Noun Phrases cannot be marked for number even if they are semantically plural (63c). Only Noun Phrases that are headed by an overt determiner can have plural marking on them (63 b,d).
a. txakurr-a c. lau txakur
dog-det four dog
'the dog' 'four dogs'
b. txakurr-ak d. lau txakurr-ak
dog-detpl four dog-detpl
'the dogs' 'the four dogs'
There is no way of marking the Noun phrase in (63c) with a plural morpheme without involving the determiner as in (63d). Therefore, there are no morphologically plural determinerless Noun phrases in Euskara. Plurality can also be encoded in the demonstratives, since they belong in the class of determiners. Their plural forms have been illustrated in 6.2., where demonstratives were overviewed.
7.2. Proximity plural determiner ok. The plural determiner ak has the variant ok, which indicates proximity in real or imaginary space or time. Thus, for instance, it is often used as the determiner of a vocative Noun phrase that refers to the audience addressed, for its proximity to the speaker:
let's go, friend-detpl
'let us go, my friends'
It is also often used to refer to a group that includes the speaker:
gizakiok ez dugu lurra ondo zaintzen
human-detpl not have-we earht-det well take-care-of-imp
'we humans do not take good care of the earth'
But it can be used generally to refer to any plural entity that is near the speaker:
a. kendu platerok mahai gainetik
take plate-detpl table top-from
'take those plates from the table'
b. egun-o-tan ez dugu berririk jaso
day-detpl-in not have-we new-prt received
'we have not received news these days'
(66a) illustrates and example where the proximity is spatial, since the plates are near the speaker. The example in (66b) illustrates the proximity determiner in a postpositional phrase. The postpositional phrase is headed by the locative postposition, and the determiner looses its final k. In this second example, the proximity of the days is of course temporal.
8.1. Person pronouns
8.1.1. Emphatic person pronouns.
8.2. Interrogative pronouns.
8.3. Indefinite pronouns derived from interrogatives.
8.3.1. Existential indefinites.
8.3.2. Universal, free-choice indefinites.
8.3.3. Negative Polarity Items.
8.3.4. Plural interrogatives.
8.4. Anaphors and reciprocals.
Pronouns are perhaps best thought of as determiners that do not take Noun phrases, but we will consider them in a separate class of their own. It must be noted, however, that often times the border that separates indefinite quantifiers, determiners and pronouns is not very clear, indicating possibly that they all belong in the same category. In this respect, the curious reader is encouraged to compare some indefinite quantifiers considered in 5.3., with what we will call indefinite pronouns in this section, to asses their similarities.
8.1. Person pronouns. The basic paradigm of personal pronouns is the following:
ni: first person singular pronoun. It always refers to the speaker:
ni joango naiz
I go-irr am
'I will go'
hi: Second person singular pronoun. It always refers to the hearer. This pronoun is only used in family or friendship settings, and it is not used in all varieties of Euskara. It has the peculiarity of obligatorily triggering addressee agreement, which is discussed in chapter 4.
hi joango haiz?
you go-irr are
'will you go?'
zu: Second person singular pronoun. It always refers to the hearer. It is used in all varieties of the language, and in varieties where hi is used, zu is used in all environments where the former is not appropriate.
zu joango zara
you go-irr are
'you will go'
gu: First person plural pronoun. It always refers to a group that includes the speaker.
gu joango gara
we go-irr are
'we will go'
zuek: Second person plural pronoun. It always refers to a group that includes the hearer.
zuek joango zarete
you go-irr are
'you will go'
There are no special forms for third person pronouns. Euskara makes use of the demonstrative system to refer to third person entities. A given form of a demonstrative is also used as a very polite second person pronoun. Pronominal uses of demonstratives have been discussed in 6.2.2.
8.1.1. Emphatic person pronouns. There is a second series of emphatic personal pronouns that can be used alone or following a basic person pronoun. The emphatic pronouns have different forms depending on the variety of Euskara:
emphatic of ni: neu, nerau, nihaur.
emphatic of hi: heu, herori, hihaur.
emphatic of zu: zeu, zerori, zuhaur.
emphatic of gu: geu, gerok, guhaur.
emphatic of zuek: zeuek, zerok, zuihauk.
In general, western varieties of Euskara have a stronger tendency to use emphatic pronouns than eastern varieties. Emphatic pronouns can be used in the following circumstances:
(I) they are used alone when the pronoun is the galdegaia of the sentence, especially if it is used contrastively (72)
a. neuk ekarri dut hori
I-emph-E brought have that
'I have brought that/ It's me who brought that'
b. zeuk agindu duzu
you ordered have
'you have ordered it'
Emphatic pronouns in galdegaia function cannot be used in negative sentences, regardless of what galdegaia position is chosen:
a.*zeuek ez duzue ekarri
you-emph-E not have brought
b.*ez dun heuk ekarri
not have you-emph-E brought
(II) Emphatic pronouns are also used in other contrastive environments, typically in topic functions, even if they are not the galdegaia of the sentence, as in (74). When immediately following a normal pronoun they also constitute topics, equivalents of English 'as for me', as shown in (74b).
a. geuk behintzat, ez dugu hori ekarri
we-emph-E at least, not have that brought
'as for us, we have not brought that'
b. nik neuk, ez dakit zer egin
I-E I-emph-E not know what do
'as for me, I don't know what to do'
Emphatic pronouns cannot be used as vocatives:
a. *heu mutil, erdu hona!
you, boy, come here
b. hi mutil, erdu hona!
'you boy, come here!'
8.2. Interrogative pronouns. Interrogative pronouns are used to construct partial questions. Here, we present the basic list of interrogative pronouns:
These interrogative pronouns can inflect for case:
nor nori nork
who who-dative who-ergative
zer zeri zerk
what what-dative what-ergative
zein zeini zeinek
which which-dative which-ergative
They can also take postpositions, and constitute various interrogative postpositional phrases:
nor-en zer-tan zein-etatik
who-gen what-in which-from
'whose' 'in what' 'from which'
Adverbial interrogatives in general cannot be directly derived by combining one of these basic interrogative pronouns with a postposition. Consider, for instance, non 'where' and noiz 'when'. Other adverbial interrogatives are derived by combining the base non 'where' with the relevant postposition: non-dik 'where-from', no-ra 'where-to', etc...
8.3. Indefinite pronouns derived from interrogatives. There are several paradigms of indefinite pronouns that are formed taking the interrogative pronoun as a base.
8.3.1. Existential indefinites. They are formed by adding the morpheme bait: norbait 'someone', zerbait 'something', nonbait 'somewhere'... in western varieties, these are formed by repeating the interrogative and inserting the conjunctive edo 'or': nor edo nor 'someone', zer edo zer 'something', non edo non 'somewhere'.
8.3.2. Universal, free-choice indefinites. There are two ways to construct them:
(a) in western varieties, they are derived by prefixing edo to the interrogative: edonor 'whoever', edozer, 'whatever', edozein, 'whichever', edonon 'wherever' etc...
(b) in eastern varieties, they are derived by suffixing nahi to the interrogative pronoun: nornahi 'whoever', zernahi 'whatever', zeinhai 'whichever', nonahi 'wherever' etc...
These quantifiers most often take the semantic value that 'free choice any' has in English. Consider a few examples:
a. edonork egin dezake hori
anyone do it-can that
'anyone can do that'
b. edozer eros daiteke diru horrekin
anything buy can-be money that-with
'anything can be bought with that money'
c. edonon aurkitzen dira bedar hauek
anywhere find-hab are grass these
'these grass can be found anywhere'
8.3.3. Negative Polarity Items. They are formed by prefixing e or i to the interrogative pronoun:
and they can be declined for case, or take postpositions in the same fashion that simple interrogatives do. Negative Polarity Items can only appear under the scope of downward entailing operators such as negation (77a) (Ladusaw (1979)), quantifiers such as gutxi 'few' (77b), conditionals (77c), and yes/no questions (77d), for instance:
a. ez da inor etorri
not is anybody arrived
b. ikasle gutxik ikasi dute ezer
student few-E learned have anything
'Few students have learned anything'
c. inon aurkitzen baduzu, harrituko naiz
anywhere find-hab if-have-you, surprise-irr am
'if you find it anywhere, I will be surprised'
d. inork ekarriko al du?
anyone-E bring-irr int has
'Will anyone bring it?'
Environments that are not downward entailing do not permit the presence of these pronouns, as the ungrammaticality of (78a, b) shows:
a. *ikasle guztiek ikasi dute ezer
student all-det learned have anything
b. *zoriona inon dago
happiness-det anywhere is
These Negative Polarity Items can appear in environments that are not downward entailing, and receive a 'free-choice' interpretation, similar to the pronouns overviewed in 8.3.2. Examples are given in (79):
a. inoren eritzia da hori
anyone-gen opinion-det is that
'that is anyone's opinion'
b. inork esango luke erregearen alaba zarela!
anyone say-irr would king-det-gen daughter-det are-that
'Anyone would say that you are the king's daughter'
The example in (79a) refers to an opinion that anyone can hold, and the example in (79b) is a sardonic exclamation, only applicable to someone who is absolutely not the king's daughter. Occasionally, and particularly in ready-made sentences and aphorisms, these polar pronouns can also take on meanings such as 'someone else'. We provide an example in (80):
inork beti errua
anyone-E always blame-det
'someone else always (bears) the blame'
8.3.4. Plural Interrogatives. In western varieties mostly, interrogative pronouns can take the plural morpheme tzu to indicate plurality. The reader is invited to read again the considerations made regarding the relationship between number and the determiner class. Given what was said there, the presence of a plural morpheme that can be attached to interrogatives strengthens the hypothesis that determiners and pronouns in general constitute a natural class in Euskara.
The forms created by the addition of the plural marker are:
nor 'who' nortzu 'who (plural)'
zer 'what' zertzu 'what (plural)'
zein 'which' zeintzu 'which (plural)'
8.4. Anaphors and reciprocals. Strictly speaking, there are no anaphoric pronouns in Euskara. Anaphors pronouns in Euskara, like in many other languages of the world, make reference to a body part. In the case of Euskara, the body part is the head. Hence, 'my own head' is the translation of the Noun phrase corresponding to English 'myself'. The anaphor is thus a determiner phrase headed by the determiner a. The determiner phrase contains a Noun phrase, headed by the Noun buru 'head'. This Noun phrase contains a genitive phrase which contains the relevant personal pronoun. This 'russian doll' structure is illustrated in (81):
[ [ [ [ neu ] re ] buru] a ]
[DP[NP[PP[DPpronoun]genitive] noun] determiner]
The paradigm of anaphors is:
neure burua myself
heure burua yourself
bere burua her/himself
geure burua(k) ourselves
zeuen burua(k) yourselves
bere burua(k) themselves
In the plural persons, the determiner can either be singular or plural; the parenthesis indicates this option. Third person anaphors are made by using the anaphoric pronominal bere, overviewed in 6.2.2.
The reciprocal pronoun in Euskara is elkar 'each other'. There is a variant of this reciprocal: bata bestea, literally 'the one the other', which is also used as a reciprocal. Although ther antecedent must be plural, elkar and bata bestea themselves are not plural, and therefore they do not trigger plural agreement on the verb:
a. Anek eta Jonek elkar maite dute
Ane-E and Jon-E each-other love have-they
'Ane and Jon love each other'
b. Anek eta Jonek batak bestea maite dute
Ane-E and Jon-E one-det-E other-det love have
'Ane and Jon love each other'
Anaphor phrases and reciprocals can be inflected for case, and they can also take postpositions to form postpositional phrases:
a. gaur ez naiz neure buruarekin ondo konpontzen
today not am my own head-with well get-along
'today I am not getting along well with myself'
b. Ane eta Jon elkarrekin bizi dira
Ane and Jon each-other-with live are
'Ane and Jon live with each other'
Anaphors and reciprocals must have their antecedent in the same sentence:
a. Mirenek bere burua zaintzen badaki
Miren-E herself take-care-hab yes-knows
'Miren knows (how) to take care of herself'
b. Jonek bere buruari hitz egiten dio sarri
Jon-E himself-D word make-hab has-it-him often
'Jon often talks to himself'
The antecedent of the anaphor need not precede it linearly, as the comparison between the examples in (82) and (83) illustrate:
a. bere burua zaintzen badaki Mirenek
herself take-care-hab yes-knows Miren-E
'Miren knows (how) to take care of herself'
b. bere buruari hitz egiten dio Jonek sarri
himself-D word make-hab has-it-him Jon-E often
'Jon often talks to himself'
It is generally accepted that the antecedent must be higher in the basic syntactic structure than the anaphor or the reciprocal. Thus, for instance, anaphors and reciprocals do not usually appear as subjects, presumably because there is no antecedent high enough to command them:
a. *elkar gurekin etorri gara
each-other we-with arrived are
b. gu elkarrekin etorri gara
we each-other-with arrived are
'we have arrived together (with each other)'
However, this issue might turn out to be a little more complex. The examples in (85) appear rather acceptable, despite the fact that the anaphoric expression is marked with ergative case, the subject case, which there are good reasons to believe is the highest in the basic syntactic structure:
a. egunotan, neure buruak kezkatzen nau
day-det-in my-own head-det-E worry-hab me-has-it
'these days, my(own)self worries me'
b. neure buruak agintzen dit zer egin eta zer ez
my-own head-E order-hab has-me-it what do and what not
'My own self orders me what to do and what not (to do)'
Whatever turns out to be the explanation for examples like the ones in (85), it is generally true however, that most predicates do not allow subject anaphors:
*neure buruak ikusi nau ni
my-own head-E seen me-has-it I
(*myself has seen me)
1. There is only one exception to this rule, which involves the noun jente 'people', in combination with the adjective gazte 'young' :
(i) gazte jentea
'the young people'
It is used in some varieties of the language to refer to 'the youth'.