43 Humanities Circle [ISSN 2321-8010], Vol. 5, No. 2, Pp. 43-61, 2017. Linguistic Tasks on Bilingual Translation Corpus for Developing Resources for Machine Translation Niladri Sekhar Dash Linguistic Research Unit Indian Statistical Institute, Kolkata Email: ns_dash@yahoo.com Abstract In this paper, we attempt to discuss the linguistic activities necessary to be carried out on bilingual translation corpus (BTC) to develop the linguistic resources and tools required in machine translation (MT). Although attempts are made in last few decades for developing BTC and models for MT, attention is hardly paid to some basic linguistic tasks indispensable for achieving success in the area. Even though we know that generation of BTC is an essential part of MT, we have not tried to understand how these BTC are going to be used in the work. To overcome this deficiency we intend to focus on some of the basic linguistic activities on BTC relating to analysis of BTC, extraction of translational equivalents, developing bilingual dictionaries, generation of terminology databank, selection of appropriate lexical items from the target language, dissolving lexical ambiguities, and generating a network of grammatical mapping with reference to lexical mapping, pragmatic information, and sentential information. In our view, an MT system will be more robust if it is powered with linguistic resources developed from linguistic tasks carried out on BTC. The activities we propose here are not only suitable for MT from English to Bangla, but also for any two language pair. These are also applicable for most of the Indian languages engaged in developing BTC between English and the Indian languages. The linguistic resources generated from analysis of BTC can also be used in language teaching, electronic dictionary compilation, machine learning, grammar development, and language cognition. Keywords: Bilingual translation corpus, translational equivalents, bilingual dictionaries, terminology databank, lexical items, source language, target language, lexical ambiguity, grammatical mapping, lexical mapping