We provide an overview of forty years of work with language corpora by the research group that started in 1972 as the Norwegian Computing Centre for the Humanities. A brief history highlights major corpora and tools that have been developed in numerous collaborations, including corpora of literature, dialect recordings, learner language, parallel texts, newspaper articles, blog posts and tweets. Current activities are also described, with a focus on corpus analysis tools, treebanks and social media analysis.