Exploring Proportionality between Translation Universals & Generic Properties of Scientific Writing in Historical Authored and Contemporary Translated /Non-translated Arabic Corpora

Comparable corpora of translated and non-translated texts are in use since the 1990’s to investigate the variable probabilities of translation universals. These are norms or deformities which become distinct when translations are tested against their originals and source texts. This small-scale monolingual comparable corpus-based study examines existing direct or inverse proportionality between variable translation universals and invariable properties of scientific writing through three selected linguistic features. The theoretical framework of study relies on translation universals of conventionalization and exaggeration (Baker,1993), over/under representation of Unique Items Hypothesis (Tirkonnen-condit, 2004) and interference (Mauranen, 2004).Three Arabic corpora are compiled: Historical Corpus,1,127,020 w, comprising three Abbasid medical and scientific works by Rhazes, Avicenna, and Ibn el-Haitham; Non-translated Corpus,156,258w, including two contemporary authored books in dermatology and physics; and Translated Corpus,375,669w, with three contemporary published E/A translations in medicine, particularly genetics, and physics.  Selected Corpora are subject to a quan-qual analysis by means of Sketch Engine corpus analyzer and a direct eye-search of random samples and cross-sections. Objectives of study are to provide evidenced clues for reconsidering translation universals of over/under representation of emphatic particles, intransitivity, and conventionalized passive voice verb forms in contemporary translated scientific Arabic, and to explore the authentic eligibility of Original Arabic for scientific writing.  An inverse proportionality is proved between the invariables of scientific neutrality, economy of language, the impersonal style, and the variable translation universals of: (1) over-representation of emphatic and infinitive /?i,?anna,?ann/as a unique item in the Translated Corpus (TASC) compared to the Historical ( HASC) and Non-translated (NTASC); (2) lower representation of  multiple Arabic intransitive verb forms in  TASC Random Sample compared to a remarkable higher richness in HASC RS; a ratio of 2:1 between HASC and TASC Rs’s is proved; (3) Conventionalized  avoidance of the Arabic passive verb form, partly due to English Interference, with a lowest frequency monitored in TASC RS. Eligibility of Original historical and contemporary Arabic for scientific elaborations is ascertained. Henceforth, Arabic translators are advised to redirect preferences of the searched linguistic representations in a way directly proportional to generic scientific properties.