• Doleschal, Ursula. 1992. Movierung im Deutschen. Eine Darstellung der Bildung und Verwendung weiblicher Personenbezeichnungen. Unterschleissheim/München: Lincom Europa.
• Kopf, Kristin. 2023. Normalfall Movierung: Geschichte und Gegenwart des generischen Maskulinums in Prädikativkonstruktionen. Jahrbuch für Germanistische Sprachgeschichte 14(1). 196–226.
• Szczepaniak, Renata. 2023. “Die Universität ist nicht nur Studienstätte, sondern auch Arbeitgeberin” - Zur Rekontextualisierung des in-Suffixes. In Igor Trost (ed.), Remotivierung -- Von der Morphologie bis zur Pragmatik, 169–195. Berlin/Boston: de Gruyter.

Feminization is the morphological process whereby the semantic feature of female sex is marked on a personal noun (Doleschal 1992: 9). It exists in all Germanic languages (e.g., Dutch journalist-e and German Journalist-in ‘female journalist’), but its usage frequency varies greatly between languages and varieties. Sentences such as the example in (1), in which the referent is female, but the personal noun (liefhebber) is non-feminized, are not a rarity in Dutch.

(1) Ze [Prinses Beatrix] is een liefhebber van zijn werk en was ook bij de opening van het Armando Museum in 1998. ‘
She (Princess Beatrix) is an admirer of his work and attended the opening of the Armando Museum in 1998.’ (Source: http://rss.fok.nl/feeds/nieuws)

By contrast, the process has been the historical default in German (Kopf 2023), and it has gained even more ground in recent decades. Moreover, feminization is attested in German even outside of its regular scope of personal nouns with female referents. Its use in reference to grammatically feminine concepts and objects (e.g., die Partei ist die Gewinner-in der Wahl ‘the party is the winner(fem) of the elections’) has been linked with the indexical value of the feminizing suffix -in as a marker of gender-fair and hypercorrect language use (cf. Szczepaniak 2023).

In this paper, I explore the limits of feminization in Dutch, Flemish, and German news media, both in terms of reduction and in terms of expansion. Using corpus data from 18 different news media sources from the Netherlands, Flanders, and Germany, covering nearly two centuries (1828-2020), the development of feminization in predicative constructions, such as in the above examples, will be traced. The aim is to explain the above-described developments in the Dutch and German feminization systems. It is hypothesized that the factors grammatical gender, referentiality (as a pragmatic factor), and language policies prohibiting or enhancing the use of feminizing morphology condition the (non-)use of feminizing morphology. Data analysis demonstrates that this can be empirically confirmed, and motivations to feminize, which were not predicted in advance, may apply as well: e.g., in Belgian Dutch, the suffix -e is significantly more productive than in Northern Dutch; feminization is generally less easily blocked by language policies when the personal noun is an inhabitant name or a sports name.

Concerning expansion of feminization, data from German demonstrate that the process is nearly unlimited in reference to human beings, and it is expanding in non-human reference. In recent decades, the number of feminized items with female human referents has risen to nearly 100% in German news media; in reference to inanimates (Partei >> Gewinnerin), feminization takes place in over half of contexts. The data demonstrate that this marking of feminizing morphology is still semantically motivated: it occurs in contexts in which a collective noun (such as Partei) is an antecedent to the personal noun, or in personified contexts.