Grammatical gender should, first of all, be differentiated from both social/cultural gender and biological sex (see Wikipedia's page on gender for the difference).

Grammatical gender is a way of describing classes of nouns within a language that code a variety of information such as animacy and participate in morphosyntactic phenomena such as agreement.

Although grammatical gender has played a particularly significant role in prior investigations of Sumerian grammar, comprehensive statements on its place in the overall system have been few and far between. In Michalowski's recent summary of Sumerian grammar, he provides a useful overview of the nature of grammatical gender in Sumerian (2004, 35). Michalowski's summary is particularly concise and is quite representative of the conventional opinion.

: Sumerian has two genders, animate and inanimate. The animate class covers humans and divinities, everything else is inanimate; perhaps one should use the terms personal and impersonal. Gender is not marked directly on the noun, but only surfaces in cross-reference, in pronouns, which are dominated by animates, and verbal concord" (Michalowski 2004, 35).

As Michalowski makes clear, there is no morphologically segmentable group of gender markers in Sumerian, but reflexes of the personal and impersonal genders show up in a number of places within the grammar: wh-words such as a.ba "who" and a.na "what"; third person possessive pronouns such as -a.ni "his/her" and -bi "its" and, crucially, as the lynchpin of a system of verbal agreement.

Perhaps more importantly, however, grammatical gender occupies a place within an animacy hierarchy or, to be more precise, a hierarchy of inherent lexical content. Sumerian is a so-called split ergative language and one of the hallmark's of many split ergative systems is that animacy or inherent lexical content plays a role in determining whether the case-marking on a particular nominal phrase is ergative-absolutive or accusative-nominative.

One of the somewhat unusual features of previous studies of Sumerian grammar and the research traditions that these previous studies have spawned is the pervasive role of the opposition between the impersonal gender and the personal gender not only as a method of description, but also as a heuristic device.

• Aikhenvald, Alexandra Y. 2000. Classifiers: A Typology of Noun Categorization Devices. Oxford University Press.
• Corbett, Greville G. 1991. Gender. Cambridge University Press.
• Ibrahim, Muhammad Hasan. 1973. Grammatical Gender: Its Origin and Development. The Hague: Mouton.
• Lakoff, George. 1987. Women, Fire, and Dangerous Things: What Categories Reveal about the Mind. University of Chicago.
• Michalowski, Piotr. 2004. Sumerian. In The Cambridge Encyclopedia of the World's Ancient Languages.
• Senft, Gunter, ed. 2000. Systems of Nominal Classification. Cambridge University Press.
• Silverstein, Michael. 1985. Language and the Culture of Gender: At the Intersection of Structure, Usage and Ideology. In Blount ed. LCS, pg. 513-551