Human science makes a claim natural science never has to: that behaviour, belief, and social life can be studied the way a chemical reaction or a rock formation can, through observation, hypothesis, and test. But people are not inert. A classification that names a kind of person, the securely attached infant, the child viewer of television, changes the behaviour of the people it describes, in a way no chemical compound ever has. The same experimental method that revealed how far ordinary people would go in obeying an authority could not, in the end, settle whether its own participants believed what they were being told. Human science studies knowers using knowers, and a tool built to measure one thing does not always survive being pointed at the other.

​

The four pages of this theme examine what it means to study people using methods built to study anything else. The Scope page asks how human science can classify its own nine-odd disciplines when the act of studying people may not be one method at all. The Perspective page asks whose sample, which classification, and whose economic model gets to stand in for humans in general. The Methods and Tools page asks what reliability and validity actually require once the person being measured can read the measurement and respond to it. The Ethics page asks what a human scientist owes the person being studied, and what the resulting knowledge is later allowed to do.