Recently, there has been renewed interest in the role of iconicity in human communication (Fay, Ellison, & Garrod, Reference Fay, Ellison and Garrod2014; Perlman, Dale, & Lupyan, Reference Perlman, Dale and Lupyan2015). Although arbitrariness has historically been considered a central design feature of language (Hockett, Reference Hockett1960), sound symbolism appears more prevalent than first acknowledged (Dingemanse, Blasi, Lupyan, Christiansen, & Monaghan, Reference Dingemanse, Blasi, Lupyan, Christiansen and Monaghan2015; Monaghan, Shillcock, Christiansen, & Kirby, Reference Monaghan, Shillcock, Christiansen and Kirby2014). Similarly, new studies on the possible evolutionary roots of language (whether spoken, signed, or written) suggest a stage of iconically grounded reference in, for instance, pantomime, vocalisations, or figurative depiction (Garrod, Fay, Lee, Oberlander, & MacLeod, Reference Garrod, Fay, Lee, Oberlander and MacLeod2007; Nölle, Staib, Fusaroli, & Tylén, Reference Nölle, Staib, Fusaroli and Tylén2028; Perlman & Lupyan, Reference Perlman and Lupyan2018; Zlatev, Żywiczyński, & Wacewicz, Reference Zlatev, Żywiczyński and Wacewicz2020).

Indeed, preceding any forms of conventionalised writing, our Palaeolithic ancestors used iconic depictions in parietal and portable art to denote animals, humans, and perhaps even narrative scenes (Aubert et al., Reference Aubert, Lebe, Oktaviana, Tang, Burhan, Hamrullah and Brumm2019). From an ontogenetic perspective, iconicity facilitates early-visual communication as children typically learn to draw before they write, and pictorial narratives appear intuitively accessible even to preschool children (Jolley, Reference Jolley2009). Given these observations, it is puzzling that no generalised ideographic systems have evolved to be fully dependent on iconicity.

Morin suggests that the reason these apparently fundamental forms of human signification never evolve to become generalised ideographies is because of the lack of procedures for standardisation. We agree that any generalised form of communication will depend on the continuous contextualised negotiation of meaning (Dideriksen, Christiansen, Tylén, Dingemanse, & Fusaroli, Reference Dideriksen, Christiansen, Tylén, Dingemanse and Fusaroli2022). However, we also suggest that there might be semiotic factors intrinsic to the materiality of certain media that challenge the evolution of ideographies – in particular, to the extent they depend on iconicity. Heraldic signs, coins, or commodity brands work well in their specialised systems for marking identities (families, cities, values, or products), but they have no means of representing the course of multiple events; they cannot tell a story.

In written language, meaning is built-up as new words are added in a continuous linear succession. Any new detail or event manifests as a spatial prolongation of the text. In an iconic depiction, the addition of new detail will instead correspond to a transformation of the depiction itself (Lotman, Reference Lotman1975). Thus, a purely iconicity-based code faces the choice of three possibilities: (1) Presenting a single-static depiction of a salient moment from unfolding events to support the reader's inferences about immediately preceding and succeeding events; (2) using the linearisation principle from glottography to present a series of depictions representing a succession of events (like a comics strip); or (3) if the medium allows, animating the image to create a continuous transformation of the icon (like a cartoon).

The first option is what we appear to find already within examples of early or contemporary rock art. However, these depictions are often limited to a single event (e.g., a hunting scene; Aubert et al., Reference Aubert, Lebe, Oktaviana, Tang, Burhan, Hamrullah and Brumm2019) or require a preexisting knowledge of the narrative being communicated. In both Australian Aboriginal and San rock art, depicted scenes represent complex narratives related to how the world was created (e.g., the Dreamtime; McDonald, Reference McDonald2013; Tacon, Reference Tacon1989) or spiritual encounters and trance states. However, without an understanding of associated mythologies or certain cultural behaviours, the art cannot be decoded. This is highlighted by the misinterpretation of San rock art by Western ethnographers, where poor translations of oral traditions have confused concepts embedded in the art (Challis, Hollmann, & McGranaghan, Reference Challis, Hollmann and McGranaghan2013; McGranaghan & Challis, Reference McGranaghan and Challis2016). Thus, the potential for static depictions to serve as ideographies is limited, despite icons sometimes having standardised elements (e.g., X-ray depictions in Australian art represent living animals, solid-filled depictions represent dead animals; Tacon, Reference Tacon1989). Rather, these iconic depictions can be considered mnemonic devices that do not communicate independently.

The second option, the comics principle, has numerous historical instantiations (e.g., the famous Bayeux Tapestry; Brilliant, Reference Brilliant1991). Although it overcomes some of the outlined limitations with respect to representing complex sequences of events and has established itself as a rich genre of graphic storytelling (Cohn, Reference Cohn2013; Stjernfelt & Østergaard, Reference Stjernfelt, Østergaard, Pedri and Petit2013), it has not evolved into conventionalised ideographic codes. Among the reasons, as discussed by Morin, is probably that it resists the compression and standardisation needed to become an economic medium of communication. For instance, comics do not come with a functional system of anaphora by which a character or object mentioned in an earlier scene can be referred to with a shorthand “she” or “it.” Rather, reappearing characters are redrawn in new configurations in every panel. Bliss symbols could be considered a solution, combining schematic icons with some arbitrary elements, grammatical categories, and the linearisation principle from verbal language (Nawar, Reference Nawar2012). However, as a kind of “creole,” it depends on language-like conventions for composing meaning, bringing back issues of standardisation and the need for an oral gloss.

The third option – the animated cartoon – may be the solution most true to the inherent semiotic nature of iconicity (not piggybacking on discretisation and linearisation principles from language; Lotman, Reference Lotman1975). A hypothetical cartoon language would have several advantages for effective communication. On the receptive side, it would be highly intuitive and accessible, thus overcoming language barriers and possibly requiring minimal formal training (Berney & Bétrancourt, Reference Berney and Bétrancourt2016). Although such an ideographic language has not (yet) evolved, it is probably not because of cognitive factors pertaining, for instance, to the architecture of our visual system or working memory. It is also not because of social factors alone, because such a communication system would potentially require fewer conventions to standardise. The main reason is probably instead the lack of a suitable medium that would support such a code, which makes it an unfeasible solution for general communication. If the effortless production, transmission, and reception of cartoons could be effectively supported by available material media, this could hypothetically constitute Morin's missing ideographic code. The historical success of glottographic codes is thus not only a matter of the way it relies on standardisation processes from natural language, but also how it is supported by the available technological solutions and material media that historically has included clay tablets, pen, paper, and print, and is currently evolving with digital media.