An attempt is made in this paper to analyze the purely formal nature of information-theoretic concepts. The suggestion follows that such concepts, used to supplement the logical and mathematical structure of the language of science, represent an addition to this language of such a sort as to allow the use of a unitary language for the description of phenomena. (The alternative to this approach must be certain multi-linguistic and mutually untranslatable descriptions of related phenomena, as with the various versions of Complementarity).

This conception is tested for the specific case of Heisenberg's Uncertainty Principle, in order to show that, with the assumption of a suitable and intuitively satisfactory definition of the quantity of information contained in a measurement, the Heisenberg Principle becomes an informational restriction arising from the formal properties of the symbols of a given language rather than as a “law” of nature.