Elliptical Zero-Subject Constructions (Zero Pronouns)

The Spanish language allows for the omission of the pronominal subject of the sentences. These omitted pronouns are usually called zero pronouns. Whereas in other languages (e.g., in Japanese), zero pronouns may appear in either the subject's or the object's grammatical position, in Spanish texts zero pronouns only appear in the position of the subject.

In MT systems, the correct detection and resolution of zero pronouns in the source language is of crucial importance if these pronouns are compulsory in the target language. In the following example, a Spanish sentence that contains a zero pronoun and its translation into English with the equivalent compulsory pronoun are shown.

• (S) [Ese hombre]$_i$ era un boxeador profesional. Ø$_i$ Perdió únicamente dos combates.
  
  
• (E) [That man]$_i$ was a professional boxer. He$_i$ only lost two fights.
  
  

We should remark that zero pronouns can also occur in English, although they appear less frequently, since they usually are used in coordinated sentences in which the zero pronoun usually refers to the subject of the sentence. Although zero pronouns have already been studied in other languages, such as Japanese--with a resolution percentage of 78% in the work of [Okumura & Tamura, 1996], they have not yet been studied in Spanish texts. [Ferrández & Peral, 2000] has presented the first algorithm for Spanish zero-pronoun resolution. Basically, in order to translate Spanish zero pronouns into English, they must first be located in the text (ellipsis detection) and then resolved (anaphora resolution) [Peral & Ferrández, 2000b]:

• Zero-pronoun detection. In order to detect zero pronouns, sentences should be divided into clauses, since the subject can only appear between the clause constituents. After that, a noun-phrase (NP) or a pronoun is sought, for each clause, through the clause constituents on the lefthand side of the verb, unless it is imperative or impersonal. Such an NP or pronoun must agree in person and number with the verb of the clause.
  
  
• Zero-pronoun resolution. After the zero pronoun has been detected, our computational system inserts the pronoun in the position in which it has been omitted. This pronoun will be detected and resolved in the following module of anaphora resolution. Person and number information is obtained from the clause verb. Sometimes, in Spanish, the gender information of the pronoun can be obtained from the object when the verb is copulative. In these cases, the subject must agree in gender and number with its object whenever the object can have either a masculine or feminine linguistic form.