Thermal and Copenhagen Interpretations
The Copenhagen interpretation requires, according to Bohr, “establishing a boundary between the quantum system one is going to investigate and the classical apparatus of measurement with which one will describe it.” http://theory.gsi.de/~vanhees/faq/epr/node4.html
To describe something classically means macroscopically, with thermodynamical methods. The only thermodynamical variables which exist are, according to statistical mechanics, expectation values of slowly altering microscopic operators.
These expectation values (for example, the average position of a pointer tip) will thus be measured in an experiment, and conclusions about the features of the associated quantum system will be drawn from it.
The primary goal of the thermal interpretation is to rethink this requirement of the Copenhagen interpretation in order to eliminate the division of the world into classical and quantum objects.
In this way, the probability structure arises out of the deterministic dynamics of the universe by a reduction of complexity to the variables relevant to a particular model. This is achieved by using the traditional methods of statistical mechanics (projection operator formalism) in the same way that the stochastic Brownian motion of a colloid particle in a classical fluid arises out of the deterministic Hamiltonian many particle dynamics.
Arnold Neumaier (Arnold.Neumaier@univie.ac.at) A theoretical physics FAQ