This article is a revised version of the keynote presented at LAK '16 in Edinburgh. The article investigates some of the assumptions of learning analytics, notably those related to behaviourism. Building on the work of Ivan Pavlov, Herbert Simon, and James Gibson as ways of "learning as a machine," the article then develops two levels of investigation (processing of personal data and profiling based on machine learning) to assess how data driven education affects privacy, non-discrimination, and the presumption of innocence. Finally, the article discusses how data minimization and profile transparency will contribute to the methodological integrity of learning analytics, while protecting the fundamental rights and freedoms of human learners thus safeguarding the creativity, humour, and contestability of human learning. [Paper presented at the LAK '16--International Conference on Learning Analytics and Knowledge (6th, Edinburgh, United Kingdom, April 25-29, 2016).]