2. Gaps in Compatibilism
Compatibilist arguments in philosophy are intuitively appealing by pointing in effect to a mental role understood empirically and in practice (so called ostensible definition). Yet the analysis does not seem to fully explain the facts, leaving much to imagination, and so begs the question as is the term. Based on Hobbes a noted modern article under the pseudonym Hobart argued choices had been treated without regard to any process or activity to make choices. Emulating engineers more is supposed to clarify the issues, understanding processes. Yet the analysis provided is not very detailed about processes. Indeed, it would seem any kind of method, procedure, faculty involved in choices would raise the specter of predestination and eliminate the magical indeterminism. This classical Compatibilism, as going back to Hobbes, takes as the unique quality of intelligence the ability to solve problems by finding means to ends, and so being goal directed with a uniquely broad scope. Problem solving activity is the relevant process. Only interference with such normal mental activity would constitute a constraint upon exercise of will, with such activity freely proceding in virtue of causes determining it. Hobart even said freedom requires determinism in order to have control (as opposed to going haywire). A.J. Ayer from the early 20th century is known for distinguishing constraint and compulsion from determination of normal mental activity. While appealing in principle, exactly what is so special about normal activity demanding concepts of personal responsibility has not been explained in any detail and something further is needed. Lots of normal mental activity is obviously determined by facts known perfectly, instead of personal activity, correctly recognized with minimal thought, as being determined to leave a room by the door instead of through walls.
One school of thought tries to side step the old assertions of indeterminism by relabeling them, in a linguistic move, as causation by an entity instead of by conditions, as properties of entities with a defining identity, in so called Entity Causation, going back to Aristotle, compared to the scientific concepts associated with David Hume and empirical schools. So a person can be blamed or credited as an entity, in virtue of this causal relation. The only problem is what forms a definable entity is sets of the usual cause and effect relations binding a set of parts into a recognizable whole. Only more obfuscation seems to be the result. It is a bit of having your indeterminism and eating it, too. (Famous Immanual Kant regarded Hobbes' freedom the same way, calling it a "wretched subterfuge". Yet Kant is classified as compatibilist in his own way.)
The most modern treatments such as Frankfurt's, popular in academic discourse, get far removed from anything practical or experienced. The argument procedes by imagining fantastical technology to read minds with a potential for interference in a way making a universal demand for indeterminism false by a counter example. Some more realistic versions can be cooked up but the whole point appears directed at universality of a principle, when in practice special cases can simply be excluded. So it does not apply to some special case, so what? Such works are in line with the dominance of skepticism in Analytical Philosophy, making hay out of obscure issues in logic, exposing faults in older, idealistic philosophy, without any practical application.
Abstract metaphysics is brought in by John Fischer offering a distinction between "guidance control" consistent with determinism versus "regulative control" with the usual magic. The terms are hardly suggestive of any difference, and through innumerable obscure issues discussed exactly what guidance control consists of is not very clear. Here it must be left to the reader to explore if interested. How can something known in experience require such a long and convoluted treatment? Something more direct would be preferable, though not the prevailing style in philosophy.