Languages come in variants. They have multifarious ways of presenting a state of affairs. We may characterize the very same angry individual as „upset“, „not amused“, „indignant“, „mad“, „exasperated“, „shirty“ or „p.o.'d“ – and this may be a matter of our personal style, or depend on several factors: where we come from, how old or young, serious or easy-going we are, what sort of text (written or spoken) we are producing, what situation we are in etc. The kind of variation which depends on the situation and the function we wish to convey, is called register.

Unsurprisingly, texts (written or spoken) come in many variants, too. Using the large digital language corpora which are at our disposal today, we may map linguistic variation along many dimensions and uncover the system of registers. And conversely, we may use the range of variation to classify texts, identify stylistic and situational parameters, and even automatically detect the author of a text. The CENTRAL autumn school „Variation in Language Corpora“ focuses on how to analyze register variation in digital text corpora.