Prefixació / Prefixation

Evidence from the NEOXOC project confirms that trends related to prefixation in Catalan are in line with the expectations derived from previous research. Between 2008 and 2010, the Project collected 1.890 neologic lexical units coined by use of prefixation, 16.20% of the total neologisms collected, which positions prefixation as the second most productive word formation process after suffixation. The set of prefixes – in terms of number, inventory and productivity – also follows the guidelines indicated in previous studies. No one prefix could be associated with a specific node, and therefore considered characteristic of a given territorial area. We can most likely, then, deduce that language in the media register is very homogenous across the linguistic domain. Although samples with regional specificity can be found, this specificity is based on the lexical component. Thus, among the neologisms, the words pseudoensaïmada, vicebatle, antimallorquí, and interinsular clearly have a Balearic flavor. The neologisms interparroquial and interpirinenc sound like they are from Andorra, while the word hemidelta evokes the Terres de l’Ebre. Words like the adjective antilínia (Girona, Barcelona, Perpignan), the verb desestacionalitzar (Andorra, Tarragona), and the words sobreexplotar and sobreoferta (Balearic Islands, Alacant) suggest specific territorial realities, as well. The comparative analysis of the data from each of the nodes suggests heterogeneity in the frequency of prefixes and of grammatical categories. Regarding the weight of prefixation among the whole set of word formation processes, data from Girona contrasts with the project globally considered: figures range from 5.24% (Girona) to 14.34% (Lleida, the next), and peaks at 23.94% (Alacant). The global distribution of data from Tarragona is also surprising: this node obtains very similar figures for nouns, adjectives and verbs, which is outstanding when compared to data from all the other nodes. Productivity of different prefixes in every node compared to the global result also shows some unpredictable mismatches. At present we have no explanation for these contrasts. They could be traces of specific trends in each area, or they could be due to our analysis of the data still being too limited, that is, that we still have not done enough analysis to be able to project the results to the whole language. For this reason the continuity of the research project seems necessary. The systematic and extensive collection of data for several years would allow us to reach some firm conclusions.