Extentionality

It was Wittenstein who introduced the theory of extentionality. He used the theory as a tool for, among other things, to show that there is no such thing as a soul, personality, and more. An ideal language is said to be truly functional when it is constructed in such a way that the truth of a complex sentence only depends on the truth of the sentences that are it is composed of, not of their meaning. Where the truth is a simple sentence, it is only on the extensions (extent) of the predicate, not the intention (the content) that is important.
The construction of a truth functional language means, such as when two sentences are used, a and b, that they have the same truth-value. We can then substitute a for b all contexts where the latter exists, and vice versa, without changing the truth-value of the content. If the truth value is changed, we will not succeed in constructing an extentional language.
How should we then deal with questions about faith? There are certain phrases of the form, "A says that p" or "A believes p". Sentences like these are among the most constitutive elements of social science - and in daily life. It is claimed that psychology and sociology should use extentional language and not be concerned with intentionality. If belief sentences can not be transformed into an extentional language, then they do not belong in science. This leads to a radical reform in these sciences. It means that we must reject the "scientific" works of James, Freud, Weber and others, all opinion and attitude studies, sociology of knowledge, intelligence (IQ) tests and more. 

 But all science aims to understand what the subject is saying. Science needs to translate the subject's message to its own language. We can only interpret as we have understood. This means that the extension (extent) and intention (content) belong together. Or put another way: quantitative and qualitative methods belong together and are not contradictory scientific methods. The best scientific studies I have read are able to combine these methods and show that studies of extent and content belong naturally together .