S is like the forefather of modern statistical programming. Developed in the 1970s by researchers at Bell Labs, it was built from the ground up to extract meaning from large amounts of data. If you like R, you can thank S - R is essentially a free reimplementation and evolution of S.
With S, researchers and statisticians could finally script their analyses instead of clicking through menus. This gave a new freedom to process, model, and visualize data exactly as they wished. Although S itself is less common today, its spirit lives on and thrives throughout the statistics community.
So next time you run a regression or plot a nice chart, spare a thought for the language that paved the way for the era of data science.