This paper aims at studying the speech act of congratulation in Persian and English with regardto semantic formulas. To gather the semantic formulas related to congratulation, 100 movies (50 in Persian and 50 in English) were chosen as the research data. The only model of cross-cultural comparison was related to that of Elwood (2004).Therefore, we used Elwood’s model as the yardstick to design a new classification for coding congratulation utterances. The new model which is designed in this study covers the classification of the Elwood’s model but omits the overlapping strategies which were observed in that model.The new model enjoys more clarity and is more clear-cut. Also, the non-congratulatory categories of Elwood’s model are deleted in the new design. Six generic categories were recognized as the model of analysis: (1) offering congratulation; (2) mentioning the occasion; (3) blessing wish; (4) expressing feeling; (5) divine statement; and (6) complimenting. In order to investigate the significant of differences between Persian and English categories, Chi-square formula was employed. The results of the study revealed that: a) there is a significant difference among various semantic formulas in each language and b) Persian and English congratulation utterances are significantly different with regard to mentioning the occasion, expressing feeling, and divine statement. However, no significant difference was observed with regard tocomplimenting,offering congratulation and blessing wish. To link theory into practice, pedagogical implications are discussed in the context of EFL.

Keywords