Arabic homographs: systemic language difficulties in text perception

The article is devoted to the cognitive difficulties that arise when reading Arabic written text. The developed system of non-recurring morphological markers practically excludes homophones from the Arabic language, however, the features of consonantal Arabic writing give a large number of homographs that form oppositions of two or three. The study parallels current research into Arabic machine translation, the development of which poses similar questions and remains ineffective for now. However, the authors focus on the specifics of human reading of text, as both native and foreign speakers read texts, aiming to understand the original rather than a literal translation. Therefore, on one hand, the research is based on the psychology of perception, which testify to the text’s integrity, structure, constancy and meaningfulness, on the other, we carried out morphothematic analysis, which allows us to determine the constituent parts of words and describe their meanings and functions.The authors analyzed their semantics and grammatical categories and came to the conclusion that most of them, being units of the same word formation chain, are included in the same semantic field, but differ in specific morphological markers, which indicate voice, causation and the method of performing an action. Conclusions were also drawn about which multi-level features of the Arabic language, from graphics and morphology to syntax, contribute to the understanding of the text, and which ones hinder it.The results of the study are important for a relatively new research field of Arabic-Russian translation studies and can also be used in methodological developments for teaching Arabic.