The concept of covariance which describes the way different observers' measurements of the same phenomenon are related. More generally the concept that the rel

The concept of covariance which describes the way different observers' measurements of the same phenomenon are related. More generally the concept that the relationships between observers are a fundamental aspect of the systems they measure. In special relativity for example this is the Poincaré group which relates observers in different inertial reference frames. Another example is entropy in statistical mechanics (although I'm being weird by casting it in this light): entropy depends on the knowledge and sensitivity of the observer to the microstates of the system they are measuring. The relation to two observers here is considerably more complicated than a group but it is there and a fundamental aspect of the subject. I believe this concept is fundamental to every science but the "softer" you get, the harder it is to model. It is there implicitly whenever we make an observation.