Today few can deny the importance corpora have acquired in recent years as a language resource. A corpus is an indispensable tool for examining a language and is used in many fields: lexicography, syntax, semantics, discourse analysis, etc. For example, the word corpus never fails to appear in lexicographical work currently being conducted all over the world.
Whether as a complement to dictionary compilation, or as the raw material and starting point for compiling the dictionary itself, the corpus is being more and more frequently regarded as the tool and support for dictionary compilers. So much so that it is difficult to recognise that a dictionary not based on a corpus or not designed at least with the help of one has accuracy, rigour, reliability and in general, quality.
It goes without saying that corpora are not used exclusively in lexicography, because corpus linguistics is a discipline its own right. Corpora contain linguistic data and are a valuable tool if one wants to analyse real language use. Even though the conducting of empirical analyses of this kind remained outside mainstream linguistics for many years, these tools have become more and more appreciated in recent years, not just as an alternative to generative grammar, or as an opponent of it, but something to contribute another point of view and evidence. Moreover, large amounts of data compiled in corpora make it possible to respond differently to some of the needs and tasks in the language technology field (disambiguation techniques based on statistical processes, translation memories, etc.)