corpus

The word corpus comes from Latin, originally meaning "body." And that's a helpful way to think about it—a body of language! While we often picture a physical collection of books when we hear the word, a modern corpus is usually digital: a massive database containing millions or even billions of words.

Linguists use corpora (the plural of corpus) to study everything from how frequently certain words are used, to how grammar evolves over time, to regional variations in language. They can analyze patterns that would be impossible to spot by simply reading books or listening to conversations. For example, a linguist might create a corpus of news articles to see how the word "climate" is used in different contexts, revealing shifts in public discourse.

But corpus isn't limited to language research. You might hear about "the corpus of Shakespeare" to mean the full body of his writings. In publishing, the word can refer to the main part of a book or article (as opposed to the title, footnotes, or appendix). In finance, it can mean the principal of an investment—the amount of money you start with, before you count interest or earnings.

When corpus does mean a text collection, it's often built on purpose rather than at random. People may choose what to include (news, fiction, social media, spoken transcripts), and they may add labels like author, date, genre, or region so they can compare parts of the collection. Even computer scientists build corpora to train natural language processing (NLP) systems—the technology behind things like chatbots and machine translation.

You'll also see corpus in some fixed phrases and names, especially in law and medicine, because of that original "body" meaning (for example, corpus delicti in law, or terms like corpus callosum in anatomy). So next time you hear the word corpus, think "body"—a body of texts, a body of work, or the main body or principal part of something.