Are these Spanish sentences grammatically acceptable?



Partial koreferenz in Spanish as a problem of grammar, teaching and translation *1


Partial co-reference in Spanish as a grammar, teaching and translation problem.



Carsten Sinner **

** Prof Dr CarstenSinner, Institute for Applied Linguistics and Translatology. University of Leipzig. E-mail: [email protected]



Aims: Investigation of the partial koreference in Spanish and German from the perspective of grammar, foreign language teaching and translatology.
Method: After an approximation of the description of the partial koreference as a phenomenon of languages ​​in general, the partial koreference in Spanish will be examined. Earlier attempts at description are taken into account and the contexts and structures in which this construction is considered grammatical or acceptable are considered.
Results: After an attempt to explain the restrictions of the construction in Spanish, the position of the Spanish norm is presented and the importance of the structure and the use of it in language acquisition of Spanish, in contrastive linguistics and especially in translation is considered.

Keywords: Partial koreference, Spanish, acceptability and grammaticality, translation, norm


Objetivo: analizar el papel de la correferencia parcial en español y en alemán desde la perspectiva de la gramática, la enseñanza de lenguas extranjeras y de la translatología.
Método: tras un acercamiento a la descripción de la correferencia parcial como fenómeno de las lenguas en general, ésta se analiza en español, teniendo en cuenta los intentos anteriores de descripción y determinando los contextos y las estructuras en los que sestruccia gramatical o acceptable esta.
Resultados: Después de explicar las restricciones que presenta la estructura española, se trata la posición que ha adoptado la norma española.
Conclusion: se indaga por la importancia de la estructura y en suertramiento en la adquisición del español, en la lingüística contrastiva y, particularmente, en la traducción.

Palabras clave: correferencia parcial, español, aceptabilidad y gramaticalidad, traducción, norma


Objective: To analyze partial co-reference in Spanish and German from the perspectives of grammar, language teaching and translation studies.
Method: after approaching the description of partial coreference as a phenomenon in languages ​​in general, partial co-reference in Spanish and German is analyzed, taking into account earlier attempts to describe or explain the structure and determining the contexts and structures in which the construction is judged grammatical or acceptable.
Results: After an attempt to explain the restrictions of partial co-reference in Spanish and German, the normative position is presented.

Key words: Partial co-reference, Spanish, acceptability and grammaticality, translation, norm


Objectif: analyzer le rôle de la coréférence partial en espagnol et en allemand depuis la perspective de la grammaire, de l'enseignement de langues étrangères et de la traductologie.
Method: après une approche de la description de la coréférence partielle en tant que phénomène de langues en général, l'on analyze la coréférence partielle en espagnol, en tenant compte des tentatives antérieures de description et en déterminant les contextes et les structures dans lesquels cette construction est considérée comme grammaticale ou acceptable.
Results: Après avoir expliqué les restrictions que présente la structure espagnole, l'on aborde la position adoptée par la norm espagnole.
Conclusion: on explore l'importance de la structure et son traitement dans l'acquisition de l'espagnol, en linguistique contrastive et, particulièrement, en traduction.

Mots-clés: coréférence partial, espagnol, acceptabilité et grammaticalité, traduction, norm




In linguistics, partial co-reference is considered to be complicated, and as a special case of inclusion it has also been recognized as a special problem in generative approaches (see already Lasnik 1989a, 1989b). In Hispanic linguistics it has been treated rarely, not exhaustively and hardly convincingly (García Calvo 1973, [1974] 1990, Sánchez López 1999, Ramírez Gelbes 2008). In linguistics, one speaks of partial koreference when a nominal syntagma in the plural contains a component to which another nominal syntagma in the singular refers. So co-reference means `` referring to the same thing ''. Two expressions, such as a noun phrase and a pronoun in a sentence, refer to the same entity; it is therefore a syntactic link. Some authors also speak of an inclusive co-reference, since a speaker is singular (incluido) in the speakers in the plural (incluyente) is included (see García Calvo 1973, Ramírez Gelbes 2008).

In the case of partial koreference, according to principle C of the attachment theory, a pronoun must always be involved, since the R-expressions (non-pronouns) cannot have an expression with the same referential index (antecendents) or are neither locally nor non-locally bound (cf. Philippi 2008: 141– 143).2

In connection with the discussion of partial co-reference, the question of grammaticality and acceptability is important. Chomsky introduced the term of grammaticality for the level of competencethat of acceptability for that of performance (see Chomsky 1970: 12).3 Everything that is normal, natural and in no way noticeable for the speaker would then be acceptable; the 'possible' sentences, i.e. those that are most likely to occur, would be grammatical (1957: 16-17). Undoubtedly, however, there are utterances that, while entirely grammatical, are practically unacceptable to the speaker, since the scales of grammaticality and acceptability do not match. The acceptability of a linguistic utterance relates to its acceptability in linguistic usage as assessed or assessable by a - normally - knowledgeable speaker. It is essential that the acceptance criteria of linguistic utterances do not necessarily have to match the criteria for the grammaticality of linguistic units formulated in a grammar model. A unit that is regarded as grammatical in a grammar model can be unacceptable in terms of language usage, whereas units classified as ungrammatical by a grammar model can be entirely acceptable in linguistic usage. In contrast to acceptability that reflects speaker's judgments, grammaticality is bound by theory. With regard to the question of partial co-referentiality, it should be noted that since the mid-1990s the acceptability and grammaticality of linguistic utterances have increasingly been understood as graded quantities.

It is therefore a matter of a distinction that focuses on two different perspectives: on the one hand, on the grammar, either in the sense of a prescriptive grammar such as Duden, Dictionary or grammar of the Real Academia Española or in the sense of a certain grammar model, on the other hand on the descriptive level, the determination or representation of what the speakers actually do or would not do or would not do.

This distinction is essential, since utterances like `` you can't say that '', `` that is wrong '' or `` that is impossible '' always refer to a specific instance, to a level that is regarded as normal or norm you can see. Several linguists with Spanish mother tongue, who were asked by us about the pronouns and their actual or supposed incompatibility, i.e. the (im) possibility of partial co-reference, answered that certain constructions not possible even though we had shown them examples from spoken Spanish corpora. So it wasn't about them in their statements Possibility in custom, but obviously about what is not possible in the grammar model used or regarded as valid - in this case it was the theory of rule and attachment. Speakers saying or writing something that does not conform to the grammar model is no exception, but grammarians reject something because they cannot explain it is remarkable.

The partial co-reference has, syntactically speaking, two possible forms. This should be illustrated by examples (1) and (2), which are already dealt with in a similar form in Sinner (2005) for Spanish.




Partial koreference is obviously a phenomenon that, if they are made aware of it, fascinates native speakers of Spanish, but it is sometimes difficult to imagine. García Calvo (1973) tried to make Koreference understandable for his Spanish-speaking readers with drawings. However, the illustrations only help to a limited extent. In Figure 1 the author tries to illustrate how a scene in a novel could be understood in which the author reports how she looks at herself and a companion in the mirror.



In Figure 2 he tries to explain the relation between what is said and the speaker by comparing it with mirrors - what is said reflects what was thought, so to speak - in order to explain reflexivity.



Partial co-reference is in principle possible in Spanish if the nominal syntagm that depends on the other with regard to the reference is a stressed pronoun:

  1. a. yoi So puedo contar con nosotrosi + j (+ ...)
    'I can only count on us / only count on us'
    b. nosotrosi + j (+ ...) sólo podemos contar conmigoi
    'we can only count on me / only count on me'
  2. a. Peri he comprado la tarta para nosotrosi + j (+ ...)
    'I bought the cake for us'
    b. Peri + k (+ ...)hemos comprado la tarta para míi
    'we bought the cake for me'

It is irrelevant whether the pronominal phrase with the stressed pronoun is mandatory (4a and 4b) or optional (5a and 5b). In the case of unstressed pronouns, however, the situation is somewhat different; if the second nominal syntagma, which depends on the first with regard to the reference, is an unstressed pronoun, then the partial co-reference is obviously not possible in some cases or is rejected by the majority of speakers.4 According to the meager relevant literature, structures such as the following are considered ungrammatical:

  1. a. (*) Peri nosi + j (+ ...) hice un café
    'I made us a coffee'
    b. (*) Peri + j (+ ...) mei hicimos un café
    'we made myself a coffee'
  2. a. (*) Peri nosi + j (+ ...) compré las entradas
    'I bought the tickets for us'
    b. (*) Peri + j (+ ...) mei compramos las entradas
    'we bought the tickets'

The verbs votar and comprar illustrate the different functioning of unstressed (clitic) pronouns and stressed pronouns very impressively: Both verbs allow the realization of an argument or two arguments with a similar thematic role as a clitic or as a pronominal syntagm with stressed pronouns:

  1. a. Peri voté por nosotrosi + j (+ ...)
    'I voted for / chose us
    b. Peri + j (+ ...)votamos por míi
    'we voted for / chose me'
  2. a. (*) Peri nosi + j (+ ...) voté
    'I voted for / chose us'
    b. (*) Peri + j (+ ...) mei votamos
    'we voted for / chose me'
  3. a. Peri he comprado esta tarta para nosotrosi + j (+ ...)
    'I bought this cake for us'
    b. Peri + j (+ ...) hemos comprado esta tarta para míi
    'we bought this cake for me'
  4. a. (*) Peri nosi + j (+ ...) he comprado esta tarta
    'I bought this cake for us'
    b. (*) Peri + j (+ ...) mei hemos comprado esta tarta
    'we bought this cake'

In examples (8) and (9), the thematic argument appears as a prepositional addition or as an accusative object, which has a direct effect on the choice of stressed or unstressed pronouns and thus on the possibility of expressing partial co-reference. With the stressed pronouns nosotros and In (8) the co-reference is possible, whereas in (9) it is possible with the unstressed pronouns nos and me would not be, if one understands the marking * as the (only) correct view, i.e. assuming that it is ungrammatical structures. It should be noted here that the structure SINGULARi PLURALj + k (+ ...) in (8a) appears more acceptable than PLURALi + j (+ ...) SINGULARi in (8b). In view of the differences between examples (8) and (9) on the one hand and (10) and (11) on the other, it seems reasonable to assume that the apparent (but not all speakers see this) incompatibility of unstressed pronouns and structures of partial co-reference , i.e. the (apparently present) impossibility of the presence of unstressed pronouns in structures of partial co-reference can be explained by the special status of clitics in Spanish grammar. This would then mean that this restriction, if it really exists, has nothing to do with the binding restrictions or principles for pronouns set up in the rule and attachment theory.

But there are still a number of cases in which partial co-reference is obviously possible even with unstressed pronouns. (12) and (13) serve as examples. In both cases, an unstressed pronoun is found in a structure of partial co-reference, i.e. exactly the structure that was actually excluded for unstressed pronouns using the example of (8) and (9):

  1. Peri nosi + j (+ ...) veo demasiado cansados ​​como para ganar.
    'I think we're too tired to win'
  2. Todosi + ... mei vimos en la película.
    'we all saw me in the movie'

In fact, this partial co-reference with unstressed pronouns can be found in at least five different structures:

  1. with the verbs of perception (ver, sentir, notary);
  2. with the verbs of assessment or assessment (considerar, creer, tener por, ver in the sense of 'assess');
  3. with the verbs of conception or mental representation; sometimes you can find the name in Spanish verbos creadores de mundos 'Verbs of the creation of a world', with verbs like imaginary, suponer etc., which obviously also refers to the verbs of mental representation or where these categories at least overlap, as well as partial overlaps with the verbs of assessment - see (18) and (19) - can be ascertained.
  4. with the verbs of the creation of pictures (after Levin 1993, the subdivision of which includes 49 classes of verbs, including the Image creation verbsto which they the Verbs of Image Impression, Scribble verbs, Illustrate verbs and Transcribe verbs counts). In my opinion, verbs such as retratar 'portray, paint'.
  5. in causative structures.

Some examples are given below.

a. Verbs of perception

  1. No nos vi en la película que se grabó en la fiesta.
    'I didn't see us in the video that was made at the festival'
  2. En esta foto no nos veo.
    'I don't see us in the photo here'

b. Verbs of assessment or judgment

  1. nos considero incapaces de terminarlo.
    'I think we are unable to finish it'
  2. No nos creo capaces de realizarlo en el plazo previsto.
    'I don't think we're able to do that within the given deadline'
  3. Nos imaginaba mejor preparados.
    'I thought we were better prepared'
  4. Me imaginábamos ya ganador.
    'we have already introduced myself as a winner'

The structure in (16) is rejected by García Calvo (1990: 124) together with other verbs with the comment `` prohibido construir ''.

c. For imagining or creating an imaginary space5

  1. Ya nos veo con el cochazo que vamos a ganar.
    'I'll see you with the super sled we're going to win'
  2. Todavía nos veo con estos terribles pantalones de campana ... ¡Eso fue en 1973!
    'I'll still see you in those terrible bell-bottoms ... That was 1973!'

d. Verbs creating images

  1. Nos voy a retratar cenando en el jardín.
    'I will portray us as / while we are having dinner in the garden'
    García Calvo (1990: 124) rejects this construction: '' en español, con las llamadas Primeras Personas, está prohibido construir [...] frases como: Nos voy a retratar cenando en el jardín. [...]''.

e. causative or factual structures

  1. Nos hice perder.
    'I let us lose / we lost because of me'
  2. Me hicimos pasar por mayor de edad.
    'we pretended I was of legal age / we passed me as of legal age'

What these structures have in common is that the unstressed pronoun is not the argument of the verb that marks its case, but the argument of the predicate below in the tree structure.

It is therefore more about structures with exceptional case marking. Partial co-reference is possible with an unstressed pronoun if the unstressed pronoun is not the argument of the verb that marks its case. Another restriction concerns the unstressed pronouns of the third person, which obviously cannot appear in structures of partial co-referentiality.

  1. a. (*) Peri lesi + j hizo un café 'he / she made you a coffee'
    b. (*) Peri + j lei hicieron un café 'they made him / her a coffee'
  2. a. (*)Peri Come oni + j vio en la película 'he / she saw her in the film'
    b. (*) Peri + j loi imaginaban más listos 'they thought he was smarter'

The examples in (25), like those under (9) and (11), are mostly viewed as ungrammatical. The difference between the two examples in (26) is now interesting. In examples (14) to (24), all structures with exceptional case marking and with an unstressed pronoun in the first person singular or plural, the partial koreference does not seem to pose a problem (not to say: not to be ungrammatical), in ( 26), on the other hand, it is completely excluded. Not one of our native speaker informants interpreted the sentence Los vio en la película in the sense of the structure i Come oni + j (+ ...) vio en la película>, but all interpreted the sentence in the sense of the structure i Come onj + k (+ ...) vio en la película>.

In the case of the anaphoric pronouns, too, the partial koreference (as in German, see below) is obviously completely excluded:6

  1. a. *Peri sei + j (+ ...) hizo un café 'he / she made [pl.] a coffee'
    b. *Peri + j (+ ...) sei hicieron un café 'they made themselves [Sg.] a coffee'
  2. a. *Peri sei + j (+ ...) vio en la película 'he / she saw himself [pl.] in the film'
    b. *Peri + j (+ ...) sei imaginaban más listos 'they thought they were smarter'

With the stressed pronouns of the third person, on the other hand, the partial co-reference is again possible:

  1. a. Peri lo compró para ellosi + j (+ ...) 'he / she bought it for her'
    b. Peri + j (+ ...) lo compraron para éli 'they bought it for him'

In German, from the speaker's point of view, the partial koreference is, so to speak, `` completely unproblematic '' and generally not unusual. If a boy has gone to the sports club with his mother in order to be registered by her in the athletics group, he can then report to his sister:

  1. Momj and mei were at the sports club today. Wei + j have mei Registered.
    The sister could answer:
  2. Well finally, youi + j wanted youi register last month!

This is completely inconspicuous and is not rejected as ungrammatical by the native speakers of German. This applies in all possible contexts and is not restricted to certain classes of verbs and is accordingly also possible in cases such as (32).

(32) ii see usi + j in the mirror.

Spanish-speaking learners of German are usually amazed at this structure in German. In German, such constructions are consistently both grammatical and acceptable. So in Spanish it is different; Most of the structures of partial koreference, which are normal or inconspicuous in German, are rejected as unacceptable by Spanish speakers. As we shall see, the prescriptive grammar - recently, it must be said - rejects a number of the structures with partial co-reference as ungrammatical.



In search of explanations for the fact that Spanish native speakers tend to reject partial koreference or that this construction is apparently less acceptable in Spanish than in German or apparently partly ungrammatical, one will hardly find anything in the literature. Muller (1985) treats the phenomenon in French and speaks of one problem of acceptance, of pronoms incompatibles and one blocage the combination of pronouns and NP. So far, only García Calvo in 1973 and 1974 [1990], currently Ramírez Gelbes (2008), have dealt with the phenomenon in Spanish. García Calvo (1973) is particularly worth mentioning because it is probably the first contribution in which the subject of partial co-reference in Spanish is dealt with in detail. Noteworthy is v. a. the attempt to visualize the respective references using examples such as the structures given in (33), (34) and (35) in Figures 3, 4 and 5.

  1. Se muerde la cola.
    'he bites his tail fin'
  2. Le muerde la cola.
    'he bites his tail fin'
  3. Se tired la cola.
    'they bite his [and at the same time he himself] in the tail fins' etc.

It is noteworthy that he also takes the 3rd person into account here, who behaves very differently from the series of the 1st and 2nd person singular and plural (in the 3rd person is due to the coincidence of le and se present a particular problem; see below). García Calvo (1973) lists the different possible combinations of subject in the singular or plural and personal or. Reflexive pronouns, but only illustrates the situation in the 3rd person singular and plural.7 In terms of frequency, the pronouns of the first and second person, also known as dialogue pronouns (Muller 1985), which refer to the people involved in an act of conversation, are more interesting (see Sinner 2005).

The pictures are cute, but they only illustrate what is difficult to put into words. Native speakers of German talking to native Spanish speakers about the difference between I'll make us a coffee and I'll make a coffee for us want to entertain, will find it difficult to understand the lack of understanding on the part of the Spanish native speakers, as they know the structure so well and cannot even imagine doing without it; Conversely, it is sometimes difficult for native speakers of Spanish to understand where the differences lie when it comes to the inclusion and exclusion of a person in or from a group - or, as illustrated here, a fish in or out of a school of fish .





As has already been shown elsewhere (Sinner 2005), García Calvo does not provide any authentic Spanish evidence and explicitly admits that he did not search for them systematically (1973: 293). It is very telling that the only non-constructed examples used by the author to prove the infracciones - Standard violations - of the type *nos amo 'I love us' comes from French! On the one hand, it is absurd to have the grammaticality or acceptability of a structure in one Language with evidence from a other Illustrate language (see below). On the other hand, Muller (1985, 1986) showed that partial koreference in French can also be found in literature8 finds and is by no means fundamentally rejected by the speakers, but rather certain structures - and precisely below je nous fais ... - are at least partially regarded as acceptable (see below). The judgments about the supposed ungrammaticality - García Calvo speaks of '' peculiar interdicción '' (1973: 270), '' prohibición '' and '' sintagmas prohibidos '' or '' inadmisibles '' (1973: 271) - are therefore subjective Speaker judgments that can only be explained with the introspection and intuition of the author and with the language awareness he has acquired (Sinner 2005).

In a '' descripción gramatical de los hechos '' (García Calvo [1974] 1990: 123) that followed in 1974, one expects a discussion of structures such as me amamos 'we love me' in Spanish, but it is essentially a discussion of French and English examples. It is all the more remarkable, then, that in the most important Spanish grammar of the last decade, in the Gramática descriptiva de la Lengua Española from 1999, the author of the only article in which the partial koreference is taken into account, Sánchez López (1999), cites García Calvo ([1974] 1990) as the only Spanish source - alongside a reference to Lasnik9 who deals with Asian languages ​​- if she determines that it is apparently a universal one10 Principle of grammar to exclude the possibility that the predicate contains an expression that is partially co-referential with the subject. She illustrates the latter with the examples "* Nos compré las entradas para el teatro." And "* Vosotros siempre te habéis admirado mucho." (Sánchez López 1999: 1072-1073). Apart from the fact that the second example is pragmatically nonsensical because you may be conceited, but not yourself admire can, this article is also a concise example of how absurd the transfer of examples of a certain conception of grammar from one language to other languages ​​can be, because the Lasnik quoted by Sánchez López works on Vietnamese and Thai and continues with the as ungrammatically discarded English We like me apart.

Lasnik (1989ay1989b) saw English sentences likeWelovemeas ungrammatical and unacceptable, but the author takes into account sentences such as I make us coffee Not. The rejection seems to be mainly semantically motivated, because sentences like in English are also I make us coffee apparently no exception, although García Calvo ([1974] 1990: 130) believes it is a 'strange' usage because he could not find more than two examples.

García Calvo was not yet able to use electronic corpora as is possible today. Today, before you start looking in corpora, you can, without any effort, z. B. at Google check whether someone has ever used a particular structure on the Internet. On the Internet and in electronic corpora, hundreds or thousands of records for sentences such as I'll make us some coffee, I'll fix us a nice cup of tea etc., which clearly show that judgments about the impossibility of certain structures in times of giant corpora and the Internet must be handled with care.

In relative terms, it must be said that you can actually find 'everything' on the Internet. On the other hand, the fact that an element cannot be found in a corpus does not allow us to conclude that the structure is ungrammatical or even non-existent. Even if a corpus cannot provide information about what is possible or impossible in a language, it can nevertheless provide information about what is actually implemented in a language and how it is common practice (Levin / Song / Atkinson 1997: 25, p . Sinner 2004: 117). This, in turn, is an important point of reference for the position of the speaker, which is not necessarily to be found in grammar - however we understand it, as a prescriptive construct or as a grammar model. García Calvo (1973: 276–278) uses his own competence as a speaker to show the more or less pronounced ungrammaticality of the structures of partial co-referentiality (which, however, is not yet called that for him). For Spanish, too, you can find evidence such as (36) and (37) and (38) on the Internet, which definitely come from native Spanish speakers:

  1. Nauzet: `` Nos veo capacitados para ganar en el Heliodoro '' (Nauzet Alemán Viera, born in Las Palmas de Gran Canaria) ( 1240333098.html)
  2. no es la forma, y ​​definitivamente, no es el momento. pero yo nos creia libres !!! libresss !!! como nadie en el mundo entero. muniekitarotah, 6. 2.2008 (,)
  3. Entre espejos nos veo. (Maurico Ocampo, Pag de escritos, críticas y todo lo que se les ocurra que pueda generar rebeldía; March 13, 2006) ( veo.html)

The first and only reference to the ungrammaticality of the partial koreference in Spanish in a prescriptive work can be found in the Diccionario Panhispánico de Dudas of the Real Academia of 2005. It says there under 7. c) in the section on unstressed personal pronouns:

No puede haber correferencia parcial entre el clítico y el sujeto del verbo; por ello no es posible una oración como * Nos hice una cena riquísima (el referente '' yo '' del sujeto es solo una parte del referente '' nosotros '' del complemento indirecto). Los referentes han de ser, o bien totalmente distintos, o bien totalmente coincidentes: os hice una cena riquísima (sujeto '' yo '' y complemento indirecto '' vosotros ''); Nos hicimos una cena riquísima (the referente del sujeto y del complemento indirecto es `` nosotros '').


This is now the state of prescriptive grammar; However, the descriptive approximation via the corpus analysis and in particular the consideration of the speaker's attitude in the form of acceptability and grammatical judgments gives a somewhat different picture.



As part of a study on the Spanish of Catalonia (Sinner 2004), cases of partial co-reference were taken into account. Among other things, statements by speakers from Barcelona and Madrid on a number of lexical, syntactic, morphological and pragmatic phenomena were compared in order to determine regional differences in the use and understanding of certain linguistic phenomena. In the case of acceptability and grammaticality studies, the test sentences must be constructed in such a way that it is ensured that approval or rejection is actually also related to the examined element and not to other phenomena in the structures to be evaluated. It must be checked very carefully whether the test kits show more than one problematic aspect and, if necessary, questions must be asked for clarification. The necessity of diversionary questions, which should lead the informants away from the actual goal of the study, made it close in the contrastive investigation to Catalan Spanish to include other questionable phenomena in addition to supposed peculiarities of Catalan Spanish. The acceptability of the following cases of partial koreference was also examined (for details: Sinner 2005):

  1. Si quieres nos hago un cafetín, ¿te apetece? 'If you want, I'll make us a cup of coffee, do you like?'
  2. Yo nos creo superiores a nuestros adversarios. 'I consider us superior to our opponents'

The first sentence shows cafetín a diminutive form which only in some regions of Spain - v. a. in Asturias - in this way is common. Worked for speakers of other dialects of Spanish -in as a miniaturization of the place - 'a small café' - strange in the given context and thus provided the desired additional topic of conversation. The use of the verb was of further interest hacer 'make', which is more common in Spanish in the Catalan-speaking areas than in Spanish in other regions and is often used with a different meaning or in constructions where it is otherwise (in other areas) impossible, such as, for example hacer un caféwhat is not used as 'making a coffee' but in the sense of 'drinking a coffee'. In addition to test kits for desemantized hacer sentences were also required in which hacer really means 'doing' in the narrower sense, as is the case in (39). In fact, almost all test subjects commented extensively on the sentence, which, due to the partial co-reference, they almost consistently rejected as '' totalmente imposible '' 'completely impossible', '' fea '' 'ugly' etc., without further elaborating on the verb to express; also cafetín was just en passant classified as curious.

(40) was also constructed, based on an example sentence from Moliner (1991: s.v. creer). The sentence '' No le creo tan inteligente como dicen '' 'I do not consider him to be as intelligent as is claimed' has been adapted to the needs of the study by inserting a partial co-reference.

The control groups in Barcelona and Madrid, who otherwise differed greatly in their judgment, rated the sentences relatively positively with regard to grammaticality (formal acceptability); A good half of the informants rated the sentences as grammatically correct, so they found no reasons for rejection due to the structure. However, both groups rate the sentences very negatively in terms of their acceptability (intuitive acceptability):11 On a seven-point scale from very good (7) to good to bad to very bad (1), the first of the two sentences only reaches a rating between 2.2 and 3.1.12 In the case of nos hago Statements like '' sueña extraño '' (5 informants), '' es vulgar '' (1), '' una barbaridad '' (1), '' es redundante '' (1), '' se trata de un dativo ético '' (2), '' es un dativo afectivo '' (1) '' es como te cantas algo'' (1) or '' no encaja [el pronombre en plural con el verbo en singular] '' (4). And in the case of nos creo: '' sueña extraño '' (3), '' no significa nada '' (1), '' la construcción no existe '' (1), '' poco frecuente '' (1), '' suena a traducción anglogermánica '' (1), '' suena inglés '' (1), '' sólo la diría un extranjero '' (1), '' no es idiomática '' (1). It was particularly noteworthy that none of the informants, all of them people who work professionally with the Spanish language or through Spanish, could explain what was wrong with the sentence, or the explanations were absurd. Constructions like me hago, te hago, os hago, nos hacemos or nos tomamos, Strike out the pronoun or replace it hacer para nosotros O hacer para los dos or. creo que somos superiores, considero que somos, nos creemos given. In one case it was held nos creo form nos considero suggested, which would keep the partial co-reference. It is noticeable that even with the strongest rejection, the speakers do not know what is 'wrong' from the point of view of the prescriptive grammar; the non-use of the structure, its negligible frequency is ultimately the only explanation for its rejection by the speakers (see Sinner 2005): What is not used is unusual, is assessed as such, or rejected or questioned. This also explains why it is judged that the sentence sounds strange or like a translation. The informants did not know how to justify their intuitive rejection of the structures. It should be noted, however, that a good half of the informants did not reject the examples as ungrammatical. Obviously there is a frequency problem: the construction is in the parole not present, even if the langue the partial koreference apparently at least partially enables problem-free. It is obviously important, as is always the case with acceptability judgments, how the structures are contextualized.

Ramírez Gelbes (2008) now believes that acceptability is related to the aspectuality of the respective predicate. In her study, based on a survey of 38 students from Buenos Aires, she found that the structures with an inclusive co-reference (singular speaker is included in the plural speaker) are classified as acceptable if they meet two conditions: 1) The singular speaker dominates the plural speaker (stands i.e. in the tree structure above the plural referent) and 2) the predicate is telic, the verb thus describes processes that require an end point, which, however, `` does not necessarily have to occur '' (Meibauer 2007: 196). Ramírez Gelbes (2008) even suggests including some static (i.e. atelic) verbs in the group of telic verbs, as they allow inclusive co-reference, but this seems to indicate that their argument that telicity is one of the conditions for possibility partial co-reference, is just not applicable. Their results are interesting with regard to the acceptability of the different examples of partial co-reference; The following is the acceptance rate determined by Ramírez Gelbes (2008) in brackets:

  1. Dado que mucha gente me conoce, creo que nos conviene candidatearme. (24%)
  2. Aunque los miembros del otro partido me ofrecen algunos privilegios si los voto, todavía creo que me conviene votarnos: a vos para secretario y a mí para vocal. (32%)
  3. No hay problema: yo nos inscribo. (71%)
  4. Juan y yo teníamos que estar de guardia y yo sabía que lo único que nos mantiene despiertos es oír alguna voz; de modo que nos hablé, a Juan y a mí, toda la noche sin parar. (21%)
  5. Vos me lo dijiste y yo me lo dije a mí mismo, pero en ese momento reaccioné como no tenía que reaccionar: me hablamos en vano. (0%)
  6. Aunque la foto estaba borrosa, esos éramos nosotros: nos conozco de memoria. (65%)
  7. Como no debíamos dormirnos, ni él ni yo, y él estaba callado como una momia, decidí leernos historias divertidas. (33%)
  8. Yo estaba tan aburrida que nos tejí un pulóver a cada uno. (48%)
  9. Como moríamos de hambre, nos preparé algo con lo poco que había en la heladera. (55%)
  10. Nosotros no habríamos reaccionado así: yo nos sé muy bien. (5%)

The informants could, however, only between the test sets presented to them correctly and incorrect decide; It is also clearly problematic that the author determines the acceptability values ​​without determining what exactly was rejected in the sentences, although, as we have seen, this is to be seen as an indispensable requirement. Your sentence 10, for example, is in my opinion because of the verb drool and was not rated as acceptable by only 5% of the students surveyed because of the co-reference, while the practically identical structure in 6, with the verb conocer, Receives 65% approval rating;13 with the wording yo nos conozco muy bien the acceptability of 10 would certainly have turned out completely different, as my own studies on the acceptability of these two constructions show in a direct comparison. When comparing the two sentences in (41) to check these results, 21 adult informants from Buenos Aires have the first one with them drool 'Know' was also rejected by almost 95%, while the second with conocer 70% 'know' as acceptable has been classified.

  1. a. Nosotros no habríamos reaccionado así: yo nos sé muy bien.
    'we would not have reacted like this: I know [!?] us very well'
    b. Nosotros no habríamos reaccionado así: yo nos conozco muy bien.
    'we would not have reacted like this: I know each other very well'

Obviously, this is not about the question of telisch or atelicAs the author thinks, it is about semantics and pragmatics: A detailed analysis of the example sentences reveals a number of problems that would have required a more detailed questioning of the rejection. The high acceptability of nos inscribí must also be seen in connection with the fact that inscribirse the reflexive pronoun is incorporated; Ramírez Gelbes (2008) does not even go into this fact.



But how can the question of partial koreference be explained from a linguistic point of view and what possibilities are there to explain its restriction in Spanish?

With regard to the fundamental problem of partial co-reference with Klitika - I do not want to say impossibility, as has repeatedly been the case in the debates with colleagues, also because the corpus analyzes and the acceptability studies in this regard, as we have seen, partly speak a different language - the following explanatory approach is conceivable. If one assumes, as Anderson (2004: 11) suggests, that clitica is a form of congruence–agreement - represent,14 thus it becomes apparent that its limits with regard to the partial co-reference correspond to a certain extent to the restrictions to be found between the noun inflection of the verb and the lexical subject, as can be seen in (42):

  1. a. * Yo no hablaremos.
    'I won't speak'
    b. * Nosotros no hablaré.
    'we won't speak'

Since clitics, unlike inflection, are syntactically relatively independent, they could be considered a other Grasp the shape of the flexion. One might assume that the Spanish verb has three types of nominal congruence, one im casus rectus and two in casus obliquus: The congruence in the casus rectus is achieved through the verb morphology in the finite forms of the verb. The congruence in casus obliquus usually appears optional - but in some cases also mandatory - with an unstressed pronoun and is compatible with both the finite and the infinite forms. in the casus rectus there must be congruence with the person and number of the subject. The congruences in casus obliquus relate to the person and number of the other additions (for example the accusative object).

The congruences in casus obliquus can deviate from the congruence of the nominative with regard to their reference or be co-referent; this means that overlapping reference is impossible. So if one of the congruences in the casus obliquus refers to an individual that is contained in the reference of the nominative congruence, it must agree with this. This is a syntactic and not a semantic constraint that occurs when the clitic matches the argument of the verb, but is not given when it Not to match.

The congruences in casus obliquus can deviate from the congruence of the nominative with regard to their reference or be co-referent; this means that overlapping reference is impossible. So if one of the congruences in the casus obliquus refers to an individual that is contained in the reference of the nominative congruence, it must agree with this. This is a syntactic and not a semantic constraint that occurs when the clitic matches the argument of the verb, but is not given when it Not to match.

The already mentioned noticeably higher acceptability of structures of the form SINGULARi PLURALi + j (+ ...) as in (43) as of structures with the sequence PLURALi + j (+ ...) SINGULARi as in (44) is also due to the fact that the pronoun in Obliquus agrees with the nominative and not vice versa, and that the possibility of congruence z. B. is greater for the first person plural than for the first person singular.

  1. a. (yo) nos hago un café
    b. (yo) nos creía superiores a nuestros adversarios
  2. a. (nosotros) me hacemos un café
    b. (nosotros) me creíamos superiores a mi adversario
    The following comparison clearly shows this:
  3. a. {Yo / * esta persona / * el que habla} creo ...
    {I / this person / the one who speaks ...} believe ...
    b. {Nosotros / la familia / los alumnos ...} creemos ...
    {we / the family / the students ...} believe ...
Seen in this way, the fact that sentences like under (43) are classified as more acceptable than sentences like under (44), with a greater scope of interpretation or flexibility to explain the first person plural compared to the first person singular. So the frequency certainly also plays an important role in terms of the differences between first and second person as well as singular and plural.15 Muller (1986) has already established this for French. Muller proves that the constructions are in two acceptance and complementary tests with student informants whose mother tongue is French tu vous and nous me the least acceptable constructions are that the negation does not affect acceptability and that the sentences with a subject in the first person singular and an object in the plural are by far the preferred or most likely to be rated as acceptable (Muller 1986: 75, cf. . Sinner 2005). The constructions in which the singular referent is contained in the plural referent dominated by him are to be expected more frequently from a purely semantic and pragmatic point of view than the other way round, since speakers speak for themselves more often than for a group.

Finally, the fact that partial koreference is possible with clitics of the 1st and 2nd person, but never in the 3rd person, depends on a fundamental morphological difference between the clitics of 1st and 2nd person and those of the 3rd person in Spanish together, since anaphoric and pronominal forms only differ in their morphology in the 3rd person (see Table 1).



The clitica of the third person cannot appear in partially co-referential structures because this is simply impossible due to the morphological differences. In contrast, in the case of the 1st and 2nd person, it is actually possible under certain conditions, as has been shown, since there is no morphological distinction here. This lack of distinction explains why stressed pronouns can appear in partially co-referential structures.

This can be explained by the fact that anaphers are nominal elements that are referentially linked to another nominal element in a syntactic domain, while pronouns are nominal elements that cannot have co-reference with another nominal element in a syntactic domain. Correspondingly, pronouns corresponding to morphological anaphors or pronouns cannot enter into co-referential relationships in which both appearances are given. In cases like (46) a pronoun in a relation of partial co-reference would obviously be bound contrary to Principle B of attachment theory - according to which a non-reflexive pronoun must be free in its binding domain.

  1. (*) ellosi + j (+ ...) lei hicieron un café
    An anaphor, on the other hand, would be bound under the same conditions, but anaphors require the reference to match (47).
  2. (*) ellosi + j (+ ...) sei hicieron un café

The only elements that can appear in this way are (a) pronouns that do not have the morphology of anaphor / pronoun, i.e. only the stressed forms (without the reinforcing mismowhich automatically turns them into anaphors) and (b) the unstressed forms of the 1st and 2nd person. These unstressed forms behave like inflections in certain cases, which excludes the possibility of partial co-reference. According to rule and attachment theory, 1st and 2nd person clitics behave like anaphors or like pronouns when they are the argument of the verb marking their case, but not when they are the argument of another verb. This behavior can be explained if one assumes that unstressed pronouns are only syntactically bound to the verb if they are the argument of the verb. If the prepositions are understood as predicates, (a) and (b) can be combined into one case: Partial co-reference is possible in Spanish if the referentially-dependent element is an argument of a predicate that does not match the predicate that has the thematic role of the element marked on which it depends. Similarly, the elements that can be in partial co-reference are unstressed pronouns if they are arguments of a predicate that does not match the verb that determines its case, and stressed pronouns after preposition.



Now that the possible reasons for the presence or blocking of partial koreference have been dealt with, it is time to return to the attitude of the RAE and thus to the question of how the view of the Real Academia Española compatible with what the speakers actually do or not do.

It is noteworthy that the linguists of the Departamento de Español Urgente the news agency efe, which also consulted a lot Manual de Español Urgente publishes, classified cases of partial co-reference about which they were consulted as `` gramaticales y aceptables '' in 1997 (http: // lists. ; visited and archived on August 3, 2003). With regard to the understanding of norms, this can certainly also be explained by the fact that until the publication of the dpd (2005) there was no prescriptive grammar that allowed the use of constructions such as the Departamento de Español Urgente submitted for evaluation '' yo nos veo a punto '' 'I think we are close / done', '' tú os tienes en muy alta estima '' 'you think a lot of you', '' nosotros me vemos en una foto '' 'we see me in a photo', '' vosotros te estáis haciendo mucho daño '' 'you hurt you very much / you hurt you very much' would have refused.

The speakers do not seem to see the matter as clearly as the RAE

Even after the publication of the dpd, the rAe - d. H. the consultation point Department of Espaà ± ol al día - but in an answer to a query on October 27, 2009, according to the quoted passage from the dpd, a ( hlight = consulta + rae ):

Parece, de todas formas, que a pesar de que puede sonar también extraño, es sentido de manera menos forzada el caso en el que el sujeto es singular y el objeto es plural (Nos veo, nos oigo, nos noto), especialmente cuando se trata de verbos de percepción (ver, oír, notar), frente al caso contrario (Me vemos, me oímos, me notamos), que es claramente agramatical. Por tanto, se entiende que la combinación del objeto plural con el sujeto singular, dependiendo del contexto en que se halle inserta, podría considerarse más o menos justificada en su empleo, aunque, desde luego, siempre algo forzada, propia del lenguaje hablado familiar o coloquial, y menos adecuada cualquier otro modo de expresar lo mismo.

Existe, además, otro tipo de enunciados similar (Nos veo en un par de años casados ​​y con hijos), que responden a una construcción concreta y peculiar de nuestro idioma. Con verbos de percepción (ver, notary, sentir, etc.) y con los verbos denominados `` creadores de mundos '' (imaginar, suponer, ver 'en el sentido de imaginar', presentir, considerar, tener por, juzgar, etc.) y la presencia de un complemento predicativo en la frase, normalmente un adjetivo o un participio que concuerda en número con el pronombre átono en función de complemento directo, es corriente esta construcción en la que el sujeto es normalmente una primera persona del singular que se incluye en el plural del pronombre átono mencionado.








Otra opción es la de que el predicativo sea un gerundio:



O un adverbio:



O un complemento circunstancial:


Téngase en cuenta además que la particularidad de esta construcción estriba en que solo puede plantearse desde el punto de vista del que está hablando, es decir, siendo el sujeto singular e incluyéndose en un conjunto de personas representadas por el pronombre átono plural, y no al revés:


Con una segunda persona del singular como sujeto aunque resulta infrecuente en extremo, podría darse, por ejemplo en el estilo indirecto:



Y por último, con una tercera persona del singular resultaría también impropio:


Noteworthy are the statements about the forced use `` propia del lenguaje hablado familiar o coloquial ''.

Not only on the Internet but also in the historical corpus corde and in the corpus of modern Spanish creA der rAe itself there is evidence that shows that partial koreference is not completely uncommon - albeit by no means common - that it goes back a long way in the history of Spanish nor is it necessarily limited to the ranges indicated by the rAe. For example, the following documents can be found from publications published in Spain (see Sinner 2005):

  1. La mucha tristeza mia / que causo vuestro desseo / ni de noche ni de dia / quando estoy donde nos veo / no oluida mi compañia (Cancionero de Garci Sánchez de Badajoz, 1500-1520; CORDE)
  2. Marigreja No sea yo Marigreja, / si nos hago castigar / assi, don canicular / percodido, / como çapato roydo, / viejo ruyn y rapaz! (Farsa llamada Rosiela, 1528; CORDE)16
  3. [...] mientras Pilar declama, explica, mueve los brazos para hacer más ostensible su indignación, su comprensión o su pena, yo Enrique la veo, nos veo sin forma, desdibujados en un marco que me resulta familiar como si fuera Pilar un micrófono y yo un receptor de datos que [...] me veía, me veo obligado a registrar contestando a Pilar [...]. (Ortiz, Lourdes, Luz de la memoria, 1976; creA)
  4. Como esto siga así, nos veo cavando trincheras en Axdir. (Silva, Lorenzo, Del Rif al Yebala. Viaje al sueño y la pesadilla de Marruecos, 2001; CREA)

In addition to this evidence from the literature, an oral language evidence should also be mentioned, which appears particularly meaningful. During the opening of the International Congress of ALFAL - the Association of Linguists and Philologists of Latin America - in Montevideo on August 18, 2008, one of the opening speeches could be heard:

  1. The organización del congreso no fue fácil, no fue un camino de rosas. Después, nos veo aquí reunidos, y podemos estar contentos.

All of this evidence makes the testimony of Español al día It is questionable that the use of structures of partial conferences is peculiar to colloquial or familial orality and otherwise less adequate than alternative forms of expression: They can also be found in literature and in formal spoken language.

The prescription clearly does not start until 2005; the future will tell how it affects use. It is clear that the speakers use the structures at least in part, as the many forums show, in which about the pros and cons and also the exact meanings of the constructions, the differences between constructions such as nos compro and compro para nosotros etc. are debated (see e.g. and



The fact that there is nothing in the grammars of Spanish on the problem of partial koreference should not come as a surprise, at least when looking at the works that are not fundamentally intended for foreign language acquisition (see Sinner 2005). It is noteworthy that the question of partial co-reference, which has been neglected for a long time in Hispanic studies - apart from a few hardly noticed publications by a single author - has not yet been taken into account in contrastive linguistics and foreign language teaching, although it is But now we are dealing with an aspect that is a potential source of interference for speakers in whose mother tongue this structure is common or both grammatical and acceptable.

In the Contrastive Grammar Spanish-German by Gauger / Cartagena (1989), for example, there are no references to the structure in German or its restrictive use and low frequency in Spanish, and the same picture emerges in grammars and textbooks.

Through an analysis of different Spanish textbooks used in school and adult education in Germany, it should now be determined whether the problematic use of connected object pronouns in Spanish in the form `` nos hago '', `` me inscribimos '', `` os puedes '' etc. It turned out that in the representations of the grammar and the example texts, none of the textbooks consulted shows such a use of the connected object pronouns,17 but that no reference is made to the fact that some of these structures are not possible in Spanish. On the other hand, there is a freely accessible online textbook for Spanish and German on the Internet18, in which sentences like `` Yo nos veo '' and `` Yo nos regalo un libro '' are given as examples (see Figures 6 and 7).


http- / / grammatik / kapitel4 / spanisch_lernen_online_kapitel_4_5.htm


The examples can be found on two pages with identical content, one of which is intended as a Spanish textbook for Germans and one as a German textbook for Spaniards. Therefore, at least the possibility must be taken into account that the grammar part of the Spanish textbook was translated from a German grammar or that it was created by the German operator of the website (infos24 GmbH).


http- / / grammatik / kapitel_10 / 10_2_declinacion_de_los_pronombres_personales.htm


In terms of translation, the partial co-reference poses a potential difficulty19 for the translation into Spanish Evidence of this in texts translated from German into Spanish such as in Figure 8, where a statement by footballer Ballack from 2008, certainly a formulation like `` I don't see us as favorites '', directly as `` No nos veo como favoritos '' was translated from German. As we have now seen, this is not wrong; but it is probably unusual. However, the question of frequency, of customization, plays a special role in the translation, because it is not just about writing something that you say or write in the target language of the translation canbut about writing what you are most likely to say or write would.


From a contact linguistic point of view, it is clear that the structure is more likely to be accepted if it is known from another language, and that is for German students with German as L2, but also for students of German with Spanish as their mother tongue, at least if they have a deeper knowledge of the Germans certainly do. Ever since I became aware of the structure myself, I have noticed it again and again in non-native speakers of Spanish, and some of the evidence that can be found on the Internet can also be ascribed to non-native speakers. For bilingual or multilingual people who have an L1 or a well-mastered L2, in which the partial koreference is not only possible, but also common, I was able to prove in acceptability tests that they have the partial koreference in Spanish (as well as in Italian ) also approve constructions that are rejected by native speakers or monolingual native speakers (see also Sinner 2005). The necessity to take the problem of partial co-reference into account in the teaching of translation is obvious, given the almost complete adoption of the structure by non-native speakers and its increased acceptability with bilingual speakers.

Since, as has been shown, with regard to the partial co-reference for Spanish, there are neither traditions of description nor traditions of prescription that could have influenced the speakers in the past, and the prescription does not start until 2005, but through the Internet consultations of the Real Academia Española is already being undermined, it remains to be seen how use in Spanish will develop. With the prescription in a normative dictionary of Spanish, the perspective of consideration also opens up in German textbooks of Spanish, which indeed comply with the norms of the Real Academia Española usually make the basis of their representations of Spanish grammar.



1. Anderson, S. R. (2004). Some points of agreement. Retrieved on March 26, 2005 at [Links]

2. Chomsky, N. (1957). Syntactic Structures. London, UK etc .: Mouton. [ Left ]

3. Chomsky, N. (1970). Aspectos de la teoría de la sintaxis. Madrid, Spain: Aguilar. [ Left ]

4. Chomsky, N. (1981). Lectures on Government and Binding. Dordrecht, Netherlands: Foris.