I. Semiotics as a cognitive science
1. Semiotics and the division of human knowledge
The divisions among the forms of human knowledge introduce a disciplinary hierarchy which
not only defines courses of study, but also raises epistemological and even gnoseological concern.
For the Ancients, at least from the second century onward, disciplines were ordered as
follows: "Ethics ensures the initial purification of the heart; physics reveals that the world has a
transcendent cause and thus invites to seek incorporeal realities; metaphysics or theology [...]
ultimately brings the contemplation of God" (Hadot 1996, p.238). Among the Moderns, Locke seems
to be the first to explicitly reshuffle this threefold separation, and that happens at the very moment
when he names and defines semiotics.1
Distinguishing among the three kinds of Science where
All that can fall within the compass of human understanding [...] may be divided
properly into three sorts:
First, The knowledge of things as they are in their own proper beings, their
constitutions, properties and operations [...] This, in a little more enlarged sense of the word, I call
[Physics], or natural philosophy. [...] Secondly, [Practice], the skill of right
applying our own powers and actions. The most considerable under this head is ethics.
[...] The third branch may be called [Semeiotics], or the doctrine of signs [...], it is aptly enough
termed also [Logic] (Locke 1993, p.414-415)
Comparing his to the ancient tripartition, one notes that Locke reverses the places of physics
and ethics __ thereby giving up the initiatory character of the course of study,
which required to start out with the moral education of beginners. He is an objectivist: things first
__ things as they can be known "in their own proper beings". Lastly, Locke
replaces metaphysics with semiotics, an intent that can be attributed to his nominalism, and we can
no longer be perplexed by the metaphysical leanings of many semioticians (for example, Peirce's
semiotics is explicitly inseparable from his metaphysics).
This division of the sciences is by no means a historical oddity. As a case in point
nowadays, it is clear that Popper's theory of the three worlds derives from it. If World 1 is the
physical world, World 2, that of subjective thoughts, and World 3, in the image of Frege's third
kingdom, that of idealities, Popper notably populates this last world with statements per se, then
scientific theories, then books, newspapers and works of art, while justifiably ending up doubting
whether one can ever give order to the "pot-pourri". Our objections (argued Rastier 1991) could be
summarized as follows:
(i) World 3 is quite simply the semiotic world, which, because of a realistic illusion that
undoubtedly owes to his baggage of logical positivism, Popper could not distinguish as such.
(ii) The defining feature of human cognition consists precisely in the mediating function of
the semiotic level between the physical world (World 1) and the world of (re)presentations (World
2 according to Popper).
(iii) The worlds can be distinguished, but not be kept separate in the least: our knowledge
of the physical world depends on the two other "worlds".
(iv) Let us underline that the question has moved from sciences on to the "layers of Being"
and from epistemology on to gnoseology. However, it would be simplistic to base the distinctions
between sciences on ontological distinctions. Indeed, any cognitive activity is a practice, one that
brings into play several "levels of Being", all pertaining to the three worlds. In addition, the objective
of the sciences is not simply to describe such or such a "world", but to describe and articulate the
relationships between the various "layers of Being", by using various technical and semiotic
1. For an elementary presentation of this discipline, see