Closed conceptual domains in Hungarian canonical poetry

A corpus linguistic approach




closed conceptual domain, poetry, corpus linguistics, correlation, MI-score, time of day, season, color


The paper shows how the distribution of different concepts within a closed conceptual domain can be used for quantitative analysis of poetic corpora. The method is presented in three case studies based on the complete poems of 23 canonical Hungarian poets. The first case study analyzes the frequency of the concept NIGHT, which is part of the conceptual domain TIMES OF DAY . The second case study analyzes the frequencies of the four season concepts, and the third case study assesses the frequencies of color concepts. The change in the frequencies of the concepts analyzed seems to reflect the well-known poetic change in Hungarian poetry at the beginning of the 20 th century. The paper also demonstrates that in canonical Hungarian authors’ poetry, there is a strong positive correlation between the frequencies of the three conceptual domains, which may mean that referring to these domains is part of the same poetic toolkit aiming to highlight sensory impressions of the physical setting of poems. Finally, the paper shows which concepts from the three different conceptual
domains co-occur in the same poems with a higher mutual information score.