In this paper, I want to discuss both the history of ideology as a semiotic phenomenon, and its history as a concept in the social sciences. In the first part of my talk, I will argue, following Vološinov (1929), that language is always ideological, and that, following Halliday (e.g. 2003), the ideological efficacy of language is a function of its architecture, ie, of its metafunctional, realizational and stratificational complexity. These features of language emerge as part of the extended moment in our evolutionary history which can be considered the "semiotic big-bang" (Lukin, in press). I will also consider the reasons why it has been argued that the modern period has witnessed a rise in the proliferation and power of ideology (Malešević 2010, 2017).

In the second half, I turn to a history of the study of ideology, which turns out to be tied up with the history of linguistic relativity. Though this idea is typically associated with Whorf and linguistic debates in the 20th century, its lineage goes back to Condillac's Essay on the Origin of Human Knowledge, which presents the first coherent account of the sign-mediated nature of sensory experience (Aarslef, 1982). From Condillac, Enlightenment philosopher Destutt de Tracy took his concept of the study of ideas, for which he coined the term "idéologie". I conclude with a brief overview of how different linguistic schools approach the study of ideology.

War emerged as a consequence of the agricultural revolution, as humans settled into farming and herding in villages, around 5000 years ago (Eckhardt 1992). But war in the modern (ie post-Enlightenment period) has become more and more lethal, and as such, requires more and more semiotic energy for its legitimation (Malešević 2010). This paper examines the lexical item war, to show how its centrality to modern society is reflected in its linguistic characteristics. I consider the lexical item with respect to its dictionary definition and thesaurus location, its absolute and relative frequencies (relative to related concepts, such as violence) in both multi- and mono-generic corpora, its typical collocations, and some of its colligational affordances (for instance, its preference for an intransitive semantics). The findings (set out in Lukin, in press) confirm the ideological saturation of lexis (as argued for by e.g. Vološinov, Firth, Halliday), and the coercive nature of microgrammar. In other words, when speakers “choose” the lexical item war, they invite into their discourse a host of semantic tendencies, which largely affirm our ongoing commitment to war as a legitimate form of human action, despite the evidence of its brutal, inhuman consequences.