Encyclopedia of Evolutionary Psychological Science

Living Edition
| Editors: Todd K. Shackelford, Viviana A. Weekes-Shackelford

Semiotic Square

  • Donald FavareauEmail author
Living reference work entry
DOI: https://doi.org/10.1007/978-3-319-16999-6_3831-1


The semiotic square is a conceptual tool that was first developed within the field of semiotics to depict the logical relations of opposition and implication inherent in any given sign or sign system. It has more recently been refashioned for use in the study of the evolution of language.


The Semiotic Square is an analytic tool developed by linguist and semiotician Algirdas Julien Greimas (1917–1962) to depict how complex, nuanced, and even contradictory meanings can emerge from more primitive sets of oppositions (e.g., prescribed/forbidden) and their logical negations (e.g., not prescribed/not forbidden) in any given sign or sign system.

A graphic representation of the “possibility space of meaning” bounded by the four corners of opposition and implication as depicted below, the semiotic square is intended by Greimas to reveal “the elementary structure” of signification whereby terms and values are defined at least in part with respect to what they are not (Greimas [1966] 1983). For Greimas, this system of logical oppositions and their implications constitutes the “deep structure” of any coherent system of meaning, from the rules of a society to the narrative logic of a literary text. Accordingly, cultural theorist Fredric Jameson describes Greimas’ semiotic square as “a virtual map of conceptual closure, or better still, of the closure of ideology itself” (Jameson 1976) (Fig. 1).
Fig. 1

A simplified version of Greimas’ semiotic square, wherein any given “meaning” that can be assigned to a sign can be plotted along the coordinates of these axes (adapted from Greimas and Rastier 1968). In this framework, S1 and S2 represent a pair of signs holding contrary meanings (such as male/female or happy/sad) that necessarily imply the existence of their logical negations, −S1 and −S2. The resulting square depicts the field of meanings which can then be derived from logical relations of contrariety (dashed arrows), contradiction (solid arrows), and implication (dotted arrows)

According to Greimas, individual actions, ideas, and identities arise from, and can only be meaningfully understood in relation to, the possibilities defined by the enclosing logical structure (Greimas and Rastier 1968) – and it is this insight that several researchers in the fields of Artificial Intelligence and Evolution of Language studies have been recently trying to operationalize.

The Relevance of the Semiotic Square in Language Evolution Studies

Roboticist Luc Steels and his collaborators have developed a modified version of Greimas’s semiotic square in their experiments designed to have robotic agents create and “evolve” a shared language based on their interactions with the environment, which are then further codified by their interactions with one another (Gantelluci and Steels 2008; Steels 2011, 2015). In Steels’ conceptualization, the four corners of the square represent what he claims are the four mutually “scaffolded” linguistic entities through which shared meaning emerges and is sustained in a community of symbol users. These conceptual entities he calls the REFERENT, the IMAGE SEGMENT, the MEANING, and the UTTERANCE, all of which mutually influence each other in an ever on-going dynamic of perception, conceptualization, verbalization, and interpretation, as illustrated below.

Steels’ version of the semiotic square differs in several important aspects from Greimas’. For whereas Greimas’s square is a structuralist and formalist attempt at diagramming the topography of logical relations inherent in a given sign or sign system, Steels’ square is an attempt to diagram the evolutionary process by which a sign’s “meaning” is first bestowed and then recursively fine-tuned and employed within a group of interacting agents (Fig. 2).
Fig. 2

Luc Steels and colleagues’ version of semiotic square (adapted from Steels 2015), depicting the process by which any given sign derives its meaning, as explained in the text below

As part of the “4E” (embodied, embedded, extended, and enactive) turn in cognitive science, Steels’ model finds agents’ initial perceptions of the world “grounded” in their embodied, sensorimotor interactions within the world. No objects in the external world are pregiven REFERENTS per se in Steels’ semiotic processing framework. Rather, the plenum of available sensory input enabled by the agent’s biology (or machinery, in the case of artificial agents) must be selectively organized by the agent (or by the agent’s evolutionary history) so as to result in reliably successful interaction in an as-yet unlabeled world. The resulting perceptual IMAGE SEGMENT is so named to acknowledge that cognition deals with only relevant and actively selected aspects of external objects, as those aspects are implicated in action sequences relevant to the agent, and not to a pictorial “representation” of the fullness of the external world per se (Steels 2006, 2011, 2015).

These perceptual IMAGE SEGMENTS then become conceptualized and given meaning within contexts of interaction via a selectionist dynamic. Categorical groupings and the acquisition of task-appropriate distinctions to be preserved for use in future similar interactions are the result of exuberant possibility-testing in the assignment of image segments to action repertoires, followed by the Darwinian pruning and “entrenchment of particular solutions by positive [and negative] feedback loops” not only over evolutionary time, but ontogenetically, as well (Steels 2015).

Both the verbalization and the interpretation of linguistic signals, likewise, are fine-tuned by the feedback based on the successful and unsuccessful interactions with other agents in using those symbols – and as these feedback cycles are ever-ongoing, generative, and recursive “semiotic networks” grow in the space defined by the interaction of these four mutually scaffolded processes of meaning (Steels 2004, 2006, 2011).


The disciplines of semiotics and of evolutionary psychology use term “semiotic square” somewhat differently. In semiotics, the square is used as an analytical tool to depict “the elementary structure of signification, marking off the oppositional logic that is at the heart of both narrative progression and semantic, thematic, or symbolic content” (Felluga 2015). In evolutionary psychology, the semiotic square is used to examine the real-world conditions under which sign-based meanings are created and evolve among a community of agents.


  1. Felluga, D. F. (2015). Critical theory: The key concepts. London: Routledge.Google Scholar
  2. Galantucci, B., & Steels, L. (2008). The emergence of embodied communication in artificial agents and humans. In I. Wachsmuth, M. Lenzen, & G. Knoblich (Eds.), Embodied communication in humans and machines. Oxford: Oxford University Press.Google Scholar
  3. Greimas, A. J. ([1966] 1983). Structural semantics. Lincoln: University of Nebraska Press.Google Scholar
  4. Greimas, A. J., & Rastier, F. (1968). The interaction of semiotic constraints. Yale French Studies, 41, 86–105.CrossRefGoogle Scholar
  5. Jameson, F. (1976). Foreword. In A. J. Greimas (Ed.), On meaning: Selected writings in semiotic theory (pp. vi–xxii). Minneapolis: University of Minnesota Press.Google Scholar
  6. Steels, L. (2004). The evolution of communication systems by adaptive agents. In E. Alonso, D. Kudenko, & D. Kazakov (Eds.), Adaptive agents and multi-agent systems (pp. 125–140). Berlin: Springer.Google Scholar
  7. Steels, L. (2006). How to do experiments in artificial language evolution and why. In A. Cangelosi, A. D. M. Smith, & K. Smith (Eds.), The evolution of language (pp. 323–322). Singapore: World Scientific Publishing.CrossRefGoogle Scholar
  8. Steels, L. (2011). Modeling the cultural evolution of language. Physics of Life Reviews, 8, 339–356.CrossRefPubMedGoogle Scholar
  9. Steels, L. (2015). The talking heads experiment: Origins of words and meanings, Computational models of language evolution (Vol. 1). Berlin: Language Science Press.CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  1. 1.National University of SingaporeSingaporeSingapore

Section editors and affiliations

  • Christopher D. Watkins
    • 1
  1. 1.Division of Psychology, School of Social and Health SciencesAbertay UniversityDundeeUK