Towards an Etymological Map of Romanian

Proceedings of Recent Advances in Natural Language Processing, pages 315–323 Sep 1–3, 2021. https://doi.org/10.26615/978-954-452-072-4_037 315 Towards an Etymological Map of Romanian Alina Maria Cristea,1 Anca Dinu,1 Liviu P. Dinu,1 Simona Georgescu,1 Ana Sabina Uban,1,2 Laurentiu Zoicas1 1University of Bucharest 2Universitat Polit`ecnica de Val`encia alina.cristea@fmi.unibuc.ro, ancaddinu@gmail.com liviu.p.dinu@gmail.com, simona.georgescu@lls.unibuc.ro ana.uban@gmail.com, laurentiu.zoicas@lls.unibuc.ro Abstract In this paper we investigate the etymology of Romanian words. We start from the Roma- nian lexicon and automatically extract infor- mation from multiple etymological dictionar- ies. We evaluate the results and perform exten- sive quantitative and qualitative analyses with the goal of building an etymological map of the language. 1 Introduction Located at crossroads between East and West, the Romanian language presents a kaleidoscopic ety- mological picture. Originated from Latin, it suf- fered the influence of many cultures with which the other Romance languages did not have (much or any) contact, hence its physiognomy became, from a certain point on, different from that of its cog- nate languages (cf. Niculescu (1965, 1978, 1999, 2003)). The Romanian lexicographers, having to deal with this miscellaneous etymological structure of the language, must perform a fairly complicated task which not rarely ends up by giving in to the difficulty of identifying a word’s origin. Our analysis, based on a computational system- atization of the origins of words, aims to evaluate quantitatively and qualitatively Romanian’s etymo- logical composition. We propose a socio-cultural interpretation of the semantic domains most perme- able to borrowings from the various languages with which Romanian had a stronger contact, consider- ing that a systematic perspective on the lexicon’s etymological structure, doubled by statistics on the permeability and needs of the various onomasio- logical fields, may provide clues for future research concerning still unknown etymologies. 1.1 Preliminaries. Peculiarity of Romanian vs. the Other Romance Languages Apart from its genetically belonging to the Ro- mance linguistic family, the Romanian language shares certain phonological, morpho-syntactic and lexical features with the Balkan languages, as a consequence of its geographical position. For this reason, it was also included in the so-called “Balkansprachbund” or “Balkan language area” (cf. Rosetti (1968)) , together with Eastern South Slavic languages (Bulgarian, Macedonian and Serbian), Albanian and Greek. There are two significant differences between Romanian and the other Romance languages: 1. According to Sala et al. (1988), the fundamen- tal lexical core of Romanian comprises less words inherited from Latin than the other Romance lan- guages (Ro. 30% vs. It. 44%, Fr. 36%, Sp. 40%, Pt. 45%). 2. At the same time, while the Italo-Occidental Romance languages make use, in their basic lexi- con, of at least 25% loanwords from Latin (It. 28%, Fr. 27%, Sp. 27%, Pt. 25%), the Romanian lan- guage only counts little more than 1% words bor- rowed directly form Latin; thus, even if we add the 8% Latin words borrowed via French and Italian, the most Eastern Romance language still does not reach the Occidental proportion of the “cultural superstrate” (cf. Reinheimer Ripeanu (2004)). By combining these two components (inherited and borrowed words from Latin), …

Content truncated to 500 words