In essence, there are arguably three types of proposition, as follows:
(1) analytic a priori propositions;The first two categories are taken directly from epistemology in modern analytic philosophy, and the subcategories of (2) follow the work of Saul Kripke (Kripke 1980).
(2) synthetic a posteriori propositions
These are further divided into two forms:(i) ontologically necessary, synthetic a posteriori propositions;(3) metaphysical propositions.
(ii) ontologically contingent, synthetic a posteriori propositions;
The third category – metaphysical propositions – is inspired by the ideas of Karl Popper and Bertrand Russell.
Let us take category (1) first. Analytic a priori propositions are those like “all bachelors are unmarried”: they are true if and only if they are true solely by virtue of the meanings of terms used (understood as having their standard meanings and applied consistently). An analytic proposition is known a priori: or without appeal to experience. In epistemic terms, it has necessary truth.
Synthetic a posteriori propositions – or category (2) – are those propositions that are not analytic, and are known a posteriori (or by experience, empirical evidence and inductive arguments).
These come in two forms, as follows:
(i) necessary synthetic a posteriori propositionsCategory 3 – metaphysical propositions – is inspired partly by Karl Popper’s idea of an unfalsifiable proposition and Bertrand Russell’s teapot thought experiment, though not identical with either of these.
These are ontologically necessary in that they have either physical or metaphysical necessity. Epistemologically, however, we cannot know with apodictic truth that they are really ontologically necessary.
(ii) contingent synthetic a posteriori
Such statements are not ontologically necessary and are contingent, in that they might have been false.
A metaphysical proposition would be one that it is synthetic in that it is not analytic, and that asserts the real ontological existence of something but not verifiable or falsifiable a posteriori. That is, it would not be verifiable by any means to anyone else except the person who asserts it, e.g., if someone claimed that there was an invisible, incorporeal, undetectable unicorn in his garden, then this would be a “metaphysical proposition.” Furthermore, when pressed, the person would say that the unicorn does not even appear to him in any normal sensory way (say, by vision, hearing, touch, smell), but he knows it exists by some “special,” non-sensory intuition, revelation or supernatural insight.
The person asserting a metaphysical proposition claims some special epistemological access to “knowing” that it is true that defies the standard a priori or a posteriori categories. One could think up any number of mystical, irrational and obviously ridiculous propositions that would belong to this category. But one could also point to certain religious or supernaturalist ideas that are real examples of such metaphysical propositions.
In theory, any proposition could be classified into one of these categories.
But there are also these important types of proposition relevant to economics:
The epistemological status of counterfactual propositions is not straightforward. Any given counterfactual might be either (1) analytic a priori when an abstract statement, or (2) synthetic a posteriori when an empirical statement.
(2) Laws of nature
Laws of nature can be interpreted as synthetic a posteriori propositions. On the traditional “regularity” theory of laws of nature, they contingent synthetic a posteriori.
On a necessitarian interpretation, they are necessary synthetic a posteriori propositions.
(3) Ceteris paribus laws
These could be either (1) analytic a priori when purely abstract or (2) synthetic a posteriori when asserted as true of reality.
Kripke, Saul A. 1980. Naming and Necessity (rev. edn.). Blackwell, Oxford.