Hedging has emerged in the literature in academic writing as an important mechanism by which academic writers mitigate the force of their claims (Hyland, 1996). Despite its significant status in academic society, however, interpreting and using hedges appropriately have repeatedly been reported to be difficult for learners, especially in academic writing where EFL writers tend to unfold a collection of facts in a direct and impersonal manner (Bloor Bloor, 1991; Dudley-Evans, 1991; Hyland, 2000a; Kamimura Oi, 2006; Oh, 2007; Skelton, 1988). The present study was motivated to examine the distribution of forms and functions of hedging in discussion section of academic research articles (RAs) in two languages (English and Persian), and three disciplines (engineering, natural sciences, and social sciences) and thus aims to provide suggestions for ways to guide academic writers toward the skillful use of hedging. The data used for research comprised 150 English articles and 150 Persian articles. From each discipline, 50 articles were chosen. After collecting the discussion parts of the articles, they were loaded into AntConc 3.4.4, the instrument used in this study to identify lexical verbs in English and Persian. After determining the hedges out of the identified verbs, they were classified into four verb categories based on Hyland's(1998) classification and were analyzed in terms of both frequency and functions . The functional framework chosen for this study was organized around three broad functional categories based on Hyland's functional taxonomy of hedges in academic writing(1998)(accuracy-oriented,writer-oriented,and reader oriented).Findings showed that the English RAs were 61 % more hedged than Farsi RAs. Moreover, the distribution of hedging devices was shown to be different across disciplines. The results also showed that the English and Farsi RAs contained more writer-oriented hedges and fewer reader-oriented hedges. Accuracy-oriented hedges were the second frequent function across English and Farsi. The findings suggest that hedges are used differently across languages and disciplines. Iranian writers used less hedges than English writers, which could be due to their unfamiliarity with the concept of hedging and its use. Also, both English and Iranian writers used writer-oriented hedges which suggests that they both prefer to distance themselves from the preposition and protect themselves from unwanted challenges. Moreover, the analysis of hedging in three disciplines revealed that the use of hedges depended much on the nature of data in each discipline. Keywords: hedges, lexical verbs, metadiscourse markers, epistemic modality, discipline, discussion sections