[Exerpted from a classic of modern magical theory]
The project language:
A magical change is a logical change brought by the subject who thinks the difference in logic with sufficient clarity to capitalize on reciprocity. This clarity has two parts: accuracy in the understanding of the logical change desired, and determination of the object to which this logic is to apply. Clarity about logical distinctions (transcendent and ordinary) can be catalyzed by the formulation of the distinction in a logical language. The current understanding of magic makes it possible to construct a language in which differences in the logic of cognition can be formulated, and these formula can by understood to refer to an object specified. This language will have these characteristics:
Formulae of the language will express the logic governing the application of cognizing concepts (geometry, causality, substance, etc.). Specifically, they will express a logical distinction between two modes of cognition: the ordinary and the magical. It must represent this difference with regard
a) to the various concepts. Each concept gives a logical modality, that is, describes a set of possible propertied, relative to the logic of the concept.
b) the mode of possible intuition. In the present generation of the language, there will be only one such mode: sensibility.
c) the mode of possible thought. As we saw, the logic of cognition determines the mode of thought – what is thinkable and what is not – when cognition varies, thought varies also. It may be that certain logical distinctions cannot be made under the usual logic of thought, but by first varying cognition, thought is allowed to enter new modes where the distinction can be made. This seems to be the case with magical perception, for example. The usual logic does not allow an objects causal continuum to be treated as an object, but by ‘bootstrapping’ changes, thought is brought to a mode where it can.
The distinction formulated is a paradox, but it cannot be the sort of paradox which is epidemic to the logical system. Such cases can be handled with paraconsisten logics, where type of paradoxes are distinguished, and the logical function of these types is specified to contain them, while preserving their paradoxical meaning. This is done by making the rules of logical inference in the language differ from type to type. So, this is really another layer of modality.
The logical side of the language will be specified through the modalities described above. The other side is reference – what object does a formula refer to? It would be possible to make symbols for particular objects (i.e. words) part of the language, but this would lead to a multiplicity of logically equivalent formulae, one for each object about which a given distinction is to be made. To preserve the elegance of the language, and its universal applicability, it is desirable to instead specify objects by ostention – that is, by reference to experience of the object. Thus the language will contain variables which range over all possible objects (the universe) and are given particular meaning by reference to actual objects.
This sketches the form of the language – the rest is mathematical busywork.
“The consciousness of self (apperception) is the simple representation of the ‘I’, and if all that is manifold in the subject were given by the activity of the self, the inner intuition would be intellectual”
To step outside sensible intuition, into intellectual intuition, is to end the experience of self as an agent, and replace it with the flux of person-less thought. Thus the manipulation of cognition, if pushed into intuition, leads to the dissolution of the self.