Hetʾem min-mispar ba-ʿIvrit ha-Yiśreʼelit ha-meduberet

Gender-number agreement in Spoken Israeli Hebrew.

The rules of gender and number agreement in Spoken Israeli Hebrew are referred to in this paper. They are based on an analysis of a corpus, consisting of spontaneous conversations of Israeli native speakers, and cover verbal, nominal and adverbial speech units.
As opposed to common Modern Hebrew analysis, many cases of partial or lack of agreement are observed in Israeli Hebrew (IH) speech. Full, partial or lack of agreement in IH is consistent and predictable, and it works the same way in both the verbal and nominal systems. This consistency is maintained within a sequence of elements too. Agreement rules in IH can be divided into two major groups:
1. Full or partial agreement of an element with a preceding element
2. Lack of agreement of an element with a sequential element
My findings show that agreement works one-way, and agreement rules are conditioned by previous information: there is usually some degree of agreement with a preceding element, but not with a sequential one. Therefore, when an element refers to a sequential element - and therefore to yet unknown information - there is no agreement between the two elements, and the first element is used in its unmarked form, usually the masculine singular.
IH speakers today tend to make no distinction between masculine and feminine plural forms, and use the masculine plural forms to represent both masculine and feminine. This trend is a general process, and is not typical to the agreement rules only. But as a result, agreement with a former element can be partial. Therefore, agreement with a preceding element can be only in number, but not in gender. This can happen when the first element represents the feminine plural; then, the second element would agree with the former element fully (in gender and number) or partially (in number only).
Usually, partial agreement is observed in verbal speech units, whereas in nominal and adverbial speech units full agreement is maintained; yet, as seen in the above example, partial agreement in nominal and adverbial speech units can be also observed.
Full agreement with sequential elements was observed in the corpus only among educated, literate, people, who studied traditional Hebrew rules at school or in some academic institution, and only on formal occasions.