Communication in Poetry

Abstract

Communication in Poetry1. Though we may be unable to predict the subjects discussed at a given gathering, we can fairly accurately determine in advance how often individual sounds will occur in the course of a conversation.We can tell the printer approximately how many a's,J's and g's he will need for a 32-page short story.All we need to know for this is the relative frequency of occurrence of the sounds of the language concerned, as established by statistical analysis.Yet, the actual truth always holds a certain element of surprise.It may turn out that the author of the short story or the guests at our party use some sounds more or less often than we would have expected on the basis of previous experience.(Nor will accuracy of our expectation increase significantly if we know what topics will come up in the course of the conversation, or whether the short story has a happy or a tragic end.)The very statistical data on which we base our prediction stemming from the analysis of different sets of materials show a certain divergence.Thus, the relative frequency of /1/ in Hungarian is 5.9% according to T. Tarn6czy (who analyzed Endre Ady's poems), 4.98% according to E. Vertes (who examined the prose of Peter Veres), and 6.27% according to the records of a parliamentary stenographer.! 2. It is somewhat easier to make predictions if we are dealing with poetry.If we know the topic and basic mood of a given poem, we can predict in what direction and approximately in what measure the relative frequency of certain sounds is likely to deviate from an index number based on statistics of the standard language.M. M. Macdermott, for instance, through a statistical analysis of English poems, found that dark vowels are more frequent in lines referring to dark colours, mystic obcurity, or slow and heavy movement, or depicting hatred and struggle. 2 Paul Guiraud analyzed the distribution of sounds in Valery's poems.He found, among other things, that in Les grenades, deliberately done in crude