Symbolicity, language, and mediality

1 Introduction

There is a broad variety of phenomena called ‘symbols’ – in everyday conversations, in esoteric beliefs, in artistic discourses and in scholarly theories. Although not always elaborated, these ideas of symbols – from vague notions to detailed concepts – have one thing in common: they are semiotic. Also in ordinary communication and popular speculation, ‘symbols’ are associated with ‘signs’: things that stand for other things. These things or entities may be concrete or abstract and appear in various constellations. The letters ‘thirteen’ may signify or symbolize a certain quantity and this numeral idea may, in turn, signify or symbolize bad luck. Indeed, the etymology of the word ‘symbol,’ stemming from the old Greek symballein, meaning to throw together, puts the idea of a symbol very close to the idea of a sign in general, understood as something that represents something else. Not surprisingly, therefore, symbols and signs have sometimes been (and occasionally still are) understood as exchangeable entities.

At other times, differentiations are made. In modern semiotics, it is common to define symbols as a subspecies of signs in general. Charles Sanders Peirce’s tripartition of signs into icons, indices and symbols has been especially influential (see, for instance, 1906, CP 4.531–532, 4.536, 4.540, 4.544). However, scholars have given less detailed attention to symbolicity than they have to Peirce’s two other foundational types of signification, namely iconicity and indexicality. Furthermore, when it is actually considered in semiotic studies, symbolicity is mainly connected to language, especially verbal language. Linking symbolicity to verbal language, as Peirce did and I will continue to do in this article, is certainly legitimate. However, it is reductive to confine the limits of symbolicity to verbal language. To more fully understand the work of symbols, and thereby better grasp the place of verbal language in the context of semiosis in general, the concept of symbol must be scrutinized from a truly broad perspective and in a methodical way that does not start from but rather ends with verbal language.

Although I advocate a wide-ranging perspective, the aim of this article is not to collect and compare the numerous and only partly overlapping notions termed ‘symbols’ or ‘symbolicity.’ Instead, my starting point will be some vital facets of the concept of symbol, as circumscribed by Peirce. Among other things, I will attempt to demonstrate the far-reaching applicability of the concept, which includes language in general and, more specifically, verbal language. This involves going beyond more narrow conceptualizations that lean mainly on the notion of convention as the ground of symbols, and instead taking Peirce’s notion of habit as the determining ground of symbols (see, for instance, c.1885, CP 1.369, 1.372).

However, even if I do go into some detail, this is not supposed to be a Peirce exegesis. As in earlier publications, where I have investigated media interrelations in the light of iconicity (Elleström 2017) and indexicality (Elleström 2018b), I will work with my own understanding and elaboration of Peirce’s semiotic ideas, with the aim of being able to semiotically conceptualize mediality in general and media interrelations in particular – not to provide an account of Peirce’s ideas that is as accurate and complete as possible. Hence, I will not always reason in terms of semiosis at large, but mainly delimit the perspective to human communication and media interrelations. Human communication is about human minds sharing cognitive import, which is made possible by media – intermediate entities and phenomena that interconnect minds. In my research, mediality is an umbrella concept, involving everything pertaining to media in communication. Hence, mediality includes concepts that are more specific such as media products, media types and intermediality that I will explain and explore subsequently in the article.

The problem to be tackled is the absence of an available method for broadly yet precisely conceptualizing the immensely important role of symbolicity in human communication beyond verbal language, involving the relations between verbal languages, languages in general and other forms of communicative symbolicity in various media types – all of which overlap and interact in complex ways. This lack fragmentizes the overall understanding of communication, makes the relation between language concepts and human communication at large fuzzy, and obscures the mainly multimodal character of symbolicity in communication, including verbal language, which muddles the understanding of media interrelations.

In an effort to solve this problem, I will provide a systematic overview of symbolicity within the context of communication among human minds that brings together an all-encompassing variety of symbol use. The overview is regulated in several vital ways. I use a Peircean symbol concept, adapted to scrutinize human communication. Importantly, the outline includes a multitude of multimodal media types and media with different material, spatiotemporal and sensorial traits. In addition, the overview is structured around two parameters that cross each other: symbols being limited or widespread among people and symbols not being part of or being part of systems (languages) – in other words, degrees of spreading and systematics of symbols.

This framework makes it possible to investigate a row of deeply interrelated issues. Hence, I will delineate language in general and more specifically verbal language in relation to the proposed symbolic parameters and clarify the role of cultures, which I understand as communities that share habits, in different language conceptions. Scrutinizing symbolicity helps disentangling the relation between language and culture because both concepts rely on habits. Building on this, the structured overview also allows me to conceptualize language in relation to mediality and suggest a productive way of discerning differences and similarities between languages and media types, without either conflating or totally separating the two concepts. This conceptualization, which I consider central for the article, will rest on careful observations of different basic media traits that have symbolic functions. Finally, to complete the exploration of interrelations among the three concepts of language, mediality and culture, I will investigate how the dependence of language and media type conceptions on culture affects the idea of intermediality, meaning the interrelations among different forms of media.

Overall, this endeavor will strongly promote a more comprehensive understanding of symbolicity in general and a deeper knowledge of the role of symbolicity in human communication in particular, reaching far beyond verbal language. More specifically, the goal of my explorations is to conceptualize media interrelations in terms of symbolicity and simultaneously to provide a method for broadly yet precisely studying the immensely important role of symbolicity in human communication beyond verbal language. On a larger scale, beyond the symbolicity covered in this article, the goal is to be able to conceptualize semiotically mediality in human communication in general and media interrelations in particular. These are key issues for research within the humanities because of the omnipresence of different media types interacting and being interrelated in complex ways that profoundly affect meaning production in all forms of human communication. No media types work in isolation and all media types intersect. Understanding media interrelations, including the role of symbolicity in various media types, means having a better understanding of issues such as how meaning is transformed when travelling among different media types.

2 Peirce on symbolicity

Because of its wide-ranging nature, Peirce’s semiotics is an ideal starting point for scrutinizing communication involving all imaginable sorts of media. His best-known and most productive categorization of signs is based on three different grounds that connect representamens (signifying entities) to objects (signified entities) creating interpretants (resulting entities, normally understood as mental effects; c.1897, CP 2.228–2.229). Representamens, objects and interpretants are dynamical functions rather than fixed properties. In brief, icons emerge when something comes to stand for something else because of a perceived similarity; indices emerge when there is a ground of contiguity in semiosis, which Peirce often described as real connections; and symbols emerge when habits lead to representations. I submit that these three grounds – perceiving similarity, grasping contiguity and acting on habits – are fundamental cognitive capacities and functions that render meaning making possible.

Here, based on an extensive earlier investigation (Elleström 2014: 105–107), I will briefly present some more details of the Peircean concept of symbol that will make it possible to broadly elaborate symbolicity in human communication beyond the realms of verbal language. Compared to many of Peirce’s other concepts, his ideas of the symbol have remained relatively consistent over the years, although the terminology spans a rather wide register. Peirce primarily circumscribed the ground of symbolicity in terms of association, habit, convention, rule and law. Whereas this terminological range allows for a certain flexibility of the concept of the symbol, its core is clear.

Early on, he stated that a symbolic representamen is “the general name or description which signifies its object by means of an association of ideas or habitual connection between the name and the character signified” (c.1885, CP 1.369, cf. c.1885, CP 1.372). The description of symbols based on “an association of ideas” is, admittedly, rather vague, considering that similarity and contiguity, the grounds of icons and indices, might also be said to cause associations. “[H]abitual connection” is a more precise way of delimiting the specific traits of the symbol. This characterization of the symbol is made in the context of a discussion of verbal language, which continues to be Peirce’s prime example of symbolicity within the realm of human communication. However, like so many other Peircean concepts, habit is inscribed into a much larger frame beyond human nature and culture (see Fernández 2012).

The definition of the symbol as “a conventional sign” (Peirce 1893, CP 4.56, cf. c.1902, CP 2.275, 2.292–2.294) is close to the idea that it is grounded on a habitual connection between representamen and object. Both of these terms – ‘habit’ and ‘convention’ – constantly recur in Peirce’s definitions of the symbol, together with terms such as ‘rule’ and ‘law,’ whose meanings are perhaps closer to convention than to habit. Peirce emphasized that the symbol is based upon “a rule… a habit, or acquired law” (c.1902, CP 2.292), that it acts “by virtue of a law, usually an association of general ideas” (c.1903, CP 2.249), and that it is “a conventional sign” and “a special habit or rule of interpretation” (c.1903, CP 4.431, cf. 1903, CP 4.447–4.448). He also noted that the term ‘symbol’ is appropriate, as the Greeks used it “very frequently to signify the making of a contract or convention” (c.1895, CP 2.297).

However, I would argue that conventions are special cases of habits. David A. Pharies noted that “individuals can have their own private habits” (1985: 34), and I suggest that a convention could be understood as a regulated habit that is common among many individuals – it is an implicit or explicit agreement. Thus, several people may share a habit, but it may also be individual. Peirce indicated in several texts that the notion of the symbol includes not only collective and culturally determined conventions, but also individual and natural habits. He remarked that the symbol is “a conventional sign, or one depending upon habit (acquired or inborn)” (c.1895, CP 2.297), that it is a sign “whether the habit is natural or conventional” (1902, CP 2.307), and that the symbolic representamen depends upon “a convention, a habit, or a natural disposition of its interpretant or of the field of its interpretant” (1904, CP 8.335, cf. 1906, CP 4.9, 4.531–4.532, 4.536, 4.540, 4.544). The idea that symbols may depend upon natural dispositions or inborn habits makes the demarcation line between icons, indices and symbols fuzzier. However, in the context of human communication, and in this article, I prefer to labor with the notion of acquired habits because of its explanatory values.

Thus, a symbol is a sign type based on (acquired) habits and conventions. As conventions might be said to be habits that are shared by several people and hence regulated to a certain extent, the briefest way to define the symbol would be to say that the relation between its representamen and its object is grounded in a habit and, in this research more specifically, an acquired rather than an inborn habit.

The popular view that the Peircean symbol is defined by arbitrariness, in contrast to the non-arbitrary icon and index, is mainly a deceptive misconception, caused largely by a conflation with Ferdinand de Saussure’s linguistic notion of arbitrary signs (2011 [1916]). In Peirce’s semiotics, all signs are grounded – or, in other words, motivated. To my knowledge, Peirce made few mentions of arbitrariness in the context of symbols, combining it with the predominant notions of rule, convention and, especially, habit (1885, CP 3.360, cf. 1903, CP 4.448). Following the basics of Peirce’s semiotics, a question such as ‘what constitutes the relation between representamens and objects?’ should inform us about the ground of representation – about traits and characteristics that signs do have, not about traits that they do not have. Thus, positive answers to the question are that icons are based on similarity, indices on contiguity and symbols on habits. Arbitrariness is not really a defining trait but rather a lack of traits. While it makes sense to say that an iconic representamen represents its object because of a quality of similarity, and that an indexical representamen represents its object because of a quality of contiguity, it would be preposterous to say that a symbolic representamen represents its object because of a quality of arbitrariness.

It is true that there is a weak form of arbitrariness related to symbols, in the sense that there are aspects of the relation between representamen and object that are not based on the intrinsic qualities of the representamen (as in icons) or the relation between representamen and object (as in indices); however, it is certainly not arbitrariness, but rather habits, that generate symbolic signification. Although anything may, in principle, become a symbol of anything else, such connections may only seem to be arbitrary a posteriori, without knowledge of the origin of semiosis. In reality, the creation of symbols does not happen by pure chance, of course, but because there is a ground between representamen and object. For instance, the reason why very different words in various verbal languages may represent the same object (for example, “butterfly,” “Schmetterling,” “mariposa,” “sommerfugl”) is not that people around the world have thrown dice to select sounds and letters. The reason is that the different words have emerged in dissimilar contexts ruled by divergent cultural and linguistic habits – and in association with different icons and indices focusing on dissimilar properties of, in this case, butterflies.

Therefore, the ground of the symbol is habits, not arbitrariness. Of course, a symbol is not the same as a habit as such. A symbol is an actual result of “a special habit or rule of interpretation” (Peirce c.1903, CP 4.431) being effectuated. Habits may be performed or observed without gaining representative functions. Symbolicity includes an interpretive act in which a potential representamen is established as an actual representamen representing an object to which it is habitually connected. Hence, the complete symbolic sign does not include either specific traits of the representamen triggering perceptions of similarity (as the icon does) or a real connection between the representamen and the object (as the index does), but only an interpreting mind that effectuates the symbolic meaning guided by habits. Peirce accentuated that a symbol, such as an ordinary word, “is applicable to whatever may be found to realize the idea connected with the word; it does not, in itself, identify those things” (c.1895, CP 2.298). The symbol is a sign “which is constituted a sign merely or mainly by the fact that it is used and understood as such, whether the habit is natural or conventional, and without regard to the motives which originally governed its selection” (1902, CP 2.307). In other words, “A symbol is defined as a sign which becomes such by virtue of the fact that it is interpreted as such” (c.1903, NEM 4: 254, cf. c.1902, CP 2.275–2.277, 2.283–2.284, 2.292–2.294, 1903, CP 5.73–5.76, c.1903, CP 2.243, 2.247–2.249, 2.254–2.265, 1908, CP 6.471).

3 Two symbolicity parameters

Following the general overview of Peirce’s concept of symbol, this section drills down into two parameters that I believe are especially useful for outlining symbolicity in human communication and hence in relation to mediality. The first is symbols being limited or widespread among people, and the second is symbols not being part of or being part of systems. Because the two parameters are both related to degrees of extension (from few to many persons and from isolated to interconnected signs; in other words, degrees of spreading and systematics of symbols), they are easily confused. However, as I will demonstrate in the following sections of the article, the two parameters are clearly unalike and must be understood to cross each other. Investigating the differences of the parameters as well as their interactions makes it possible to see various shades of symbolicity in human communication more clearly.

The first parameter, symbols being limited or widespread among people, has already surfaced in our recent discussions of Peirce, and there is more to add from him. One of Peirce’s most quoted and debated passages is, “A symbol, once in being, spreads among the peoples. In use and in experience, its meaning grows” (c. 1895, CP 2.302; see, for instance, Nöth 2010: 86–87). I take this to mean that symbols, like all signs, are parts of chains of semiosis, but also that habits, the grounds of symbols, spread, grow and expand – or indeed decline and become more limited. Habits are voluntary of enforced repetitions, regularities, traditions, rules, norms and perhaps even laws. They may evolve haphazardly or be consciously construed, and they may be triggered by nature or by nurture. Although relatively constant within a certain context, they are adaptable and subject to change, which makes them very useful for communication, but also problematic depending on whether one shares certain contexts.

In any case, habits can vary from occurring sporadically to being widespread among people and vice versa. Habits sometimes grow to reach such a magnitude and importance that they become regulated – more or less strictly, officially or even legally – until they reach a state that Winfried Nöth described as follows: “Since those who interpret the symbol are not free to endow it with any meaning they might wish to associate with it, but have to comply with the meanings associated with them through the habits that determine the interpretation of the symbol, they cannot themselves use the symbols as their mere instruments” (Nöth 2010: 87). However, when investigating symbolicity in general, and without disagreeing with Nöth’s account, I think that this lack of freedom is only applicable to symbolicity based on habits that have reached a certain degree of enforced regularity. Habits shared by only a few people can certainly give rise to symbolicity evolving at liberty. Furthermore, habits are not isolated; they overlap, which means that even strongly regulated habitual systems shared by hosts of individuals, such as those underpinning verbal languages, allow for certain degrees of freedom without falling apart. As is well known, so-called language rules change slowly but constantly.

With the mention of verbal language, we have now already entered the issue of systematics and the second parameter: symbols not being part of or being part of systems. Although Peirce did not scrutinize this topic directly (it is only indirectly present in his examples of both systematized and only loosely interconnected habits), it is central for mapping various forms of symbolicity. Granted that there is a strong tendency in semiotic research in general to think of signs as always being parts of ‘sign systems,’ I believe that vital shades of semiosis disappear under such a view. It is clear that all signs are always interconnected in various ways and cannot be understood, either in practice or in theory, without a broad perspective on context and a multitude of interdependencies. However, this is not the same as always being part of systems.

In line with standard definitions of the concept, I loosely delineate a system as an organized accumulation of interrelated and interdependent parts, meaning that no part can fully function in isolation from the other parts of the system. Thus, in a symbol system, the signification of its constitutive parts cannot be exhaustively established without relating them to other symbols in the system and to the system at large. Here, ‘relating’ does not mean just vaguely connecting, but rather interconnecting in a mutual creation of meaning where the different symbols partly determine each other’s meaning. Hence, I understand systems to be more than the sum of their parts, which is characteristic of what are sometimes called complex systems. In system theory, one routinely reasons in terms of emergent traits of complex systems: qualities of systems that cannot be traced to their individual components but result from their complex interdependencies.

Another standard feature of system theory is the idea that systems are surrounded and often influenced by ‘environments.’ This is clearly the case for symbol systems. However, symbols that are not part of systems also exist in surroundings – contexts and situations – with which they often have complex relations, and all symbols definitely appear in sign chains as semiosis develops. This, I re-emphasize, is not the same as being dependent on other components within a system. At least on the level of observation that I find productive for disentangling the broad variety of semiosis in human communication, one cannot explain all symbolicity in terms of sign systems. Compared to icons and indices, and based on the flexible and highly adjustable nature of habits, symbols are undoubtedly prone to take part of ordered sign conglomerates – systems – because systematization as such is a form of habit. Nevertheless, as I will demonstrate in the following sections of this article, much symbolicity occurs outside the realms of systems.

However, I must emphasize that there are no clear borders between relatively free-floating symbols, loosely organized symbols and symbols that are part of more solid systems, or between more simple and more complex systems. Furthermore, symbol systems are prone to partly cover each other, fuse and rely on each other in various ways. Therefore, a stricter definition of symbol systems on my part would be counterproductive. Instead, I recognize that, depending on one’s objectives, there are several legitimate ways of delimiting the concept of system more narrowly or more broadly. The purpose of my second parameter, symbols being parts of systems or not, is to demonstrate clearly different modes of existence for symbols – but also to make clear the floating borders between these modes of existence.

To sum up, my overview of symbolicity in communication will be structured around these two crossing parameters. First, is the use of the symbols limited or widespread among people (that is, are the symbols delimited to more personal semiosis, grounded on habits among only few individuals, or extended to common semiosis, grounded on broadly shared habits: conventions)? Second, are the symbols parts of systems (do they work in relative isolation or rather in strong interaction with several interdependent symbols)? Importantly, both these parameters must be understood as differences in degrees, not as bipolar factors. Combining them makes possible a schematic overview of symbolicity in four main parts:

1. Limited symbols that are not part of symbol systems

2. Widespread symbols that are not part of symbol systems

3. Limited symbols that are part of symbol systems

4. Widespread symbols that are part of symbol systems

4 Limited symbols that are not part of symbol systems

The most elementary form of symbolicity is grounded on personal habits that work in isolation. They do not spread among people or form larger structures. For instance, a person may have developed the habit of drawing a flower in her calendar to remind herself to call the dentist: for that person, who likes flowers and is apt to drawing them, but perhaps hates going to the dentist, the flower becomes a habitual sign for the dentist. This is a case of autocommunication, where a mind connects to a later version of the same mind through a media product that is visual, static and iconic (because the drawing resembles a flower). However, because it has the habitual function of a reminder to go to the dentist, the drawing is also symbolic. Another example of very limited symbolicity delimited to only a few people would be the habit of a family to communicate internally through jars of cat food. A jar placed outside the front door might mean, ‘’I have already fed the cat,’ while putting the jar inside the front door might mean, ‘Please feed the cat before you leave.’ As a media product, the jar is certainly a sort of index, because it has a real, physical connection with its contents, which is supposed to be given to the cat, but it also gains a symbolic function because of the family’s internal habits of using it to communicate elementary cognitive import.

5 Widespread symbols that are not part of symbol systems

Personal habits and habits that are delimited to only few people may or may not spread and prepare the ground for shared symbolicity among many people – within cultures, subcultures, political and religious groups, nations or geographic areas – and these habits may then perhaps decline or entirely disappear. Yet, if these spreading habits are to form a ground for more stable symbolicity, which may sometimes be desirable, they must be regulated somehow. At that stage, which may be entered slowly and unnoticed or quickly and clearly, the habits become conventions. Although conventions are born out of agreements about how to regulate habits, those agreements could be everything from silent understandings (leaving the door open for misunderstandings) to socially enforced contracts. However, regulating symbolicity in this way, to make it work among a multitude of people, is not the same as creating symbol systems, which should be understood as organizing interdependent symbols into wholes where each symbol has a symbiotic relation to the other symbols in the system.

Thus, I focus here on symbols that are widespread among people without forming systems. Examples are everyday gestures, such as thumbs up or thumbs down, that symbolize approval and disapproval, respectively, in many cultures. Although these two signs are in a binary relation, they do not form a system in the sense of creating emergent meaning. Of course, such gestures can be successfully transformed among various media types: we may see actual gestures in time and space, see depicted gestures in static images like paintings and photographs or hear someone saying ‘thumbs up for that!’ The so-called V sign is another well-known gesture representing victory based on habits that emerged during and after the Second World War. This symbol may appear visually, for instance written on walls or signed with two fingers, and auditorily, in the form of the sound of the Morse code sign for V ‘… –’ (resembling the famous introduction to Beethoven’s Fifth Symphony, which could hence be used as a symbol for victory for those familiar with the convention). Although the symbolicity in these examples depends on a certain knowledge of verbal sign systems (the look and sound of letters and the spelling of words), I do not think that the V sign, or rather V signs, with their specific meaning, are parts of a symbol system because they are not methodically interrelated with other symbols with which they form significance extending the sum of the parts. It should also be noted that there are insulting versions of the manually displayed V sign. Different habits can be manifested in physically similar ways and result in ambiguity, which often becomes painfully or amusingly clear in encounters between different cultures.

Further examples of widespread symbols that are not part of elaborated symbol systems include solitary or only elementary interconnected symbols for pubs, shops, manufacturers, companies, ownership, products, environmental control of various products, services, organizations, political parties and public authorities. In modern cultures, most of these symbols are visual. However, auditory symbols standing for, say, a specific program, are common on the radio, and ice cream vans may play certain melodies to symbolize their arrival in the neighborhood. Symbols like these may also be transformed among different media types. The original symbol for the Swedish Postal Company was the sound of a horn blown by a man arriving on a horse to deliver the mail, which was eventually replaced by a visual depiction of such a horn: an icon (based on similarity with a horn) including a former index (the sound of the horn representing the real presence of the mailman) and together making up a well-established symbol – ultimately based on a convention.

Other familiar widespread symbols that do not mainly gain meaning through systems, but rather on the ground of solitary conventions, are the idea of lions, which are represented in many kinds of media modes to mean courage, and the idea of doves, again represented in many kinds of media modes to mean peace. Although widespread, these conventions are far from universal. Symbolicity only works within certain communities and contexts, however large they may be. As always, they are not arbitrary but grounded on habits – and habits do not evolve haphazardly. For various reasons, people associate certain traits with certain animals in a particular community. Thus, the same sort of animal may work as different symbols representing divergent traits. For example, the snake represents evil and devastation but also wisdom and the power of healing. Whereas a dragon mainly symbolizes destruction and vice in the Western hemisphere, it stands for creation and benevolence in the Eastern hemisphere.

Moving into the spiritual sphere, symbols are found virtually everywhere. Even though religions can probably be understood as systems of belief, they harbor plenty of widespread symbols that are not really part of systems in a stricter sense. One example is the yin yang circle, representing the two contrary but complementary forces in Daoism. There is much visual iconicity in this circle, fusing two contrary shapes in an elegant way, making the visual form a potent icon for a mental concept. Ultimately, however, a widespread habit specifies its meaning for those familiar with the convention. Another example is the Christian cross. Like many other symbols, it is historically motivated in a solid way. Whether represented by spoken or written words, by still images, movies, sculptures, etc., the cross, within the Christian context, symbolizes Christ because of the crucifixion – one of the most central events in the stories about Christ. However, it is a convention that makes the cross so central compared to other potential symbols of Christianity, and despite so many other non-Christians being crucified. Naturally, the cross does not stand in splendid isolation in Christianity; one could also mention symbols for evangelists and apostles, for example. Because these symbols are vaguely interdependent, they can possibly be said to be part of quite untied symbol systems, but I believe they are rather part of loose assemblies of symbols working within a strong and elaborated context of religious belief and referring to robustly interconnected religious narratives, without forming emergent meaning through their internal interactions.

However, my intention here is not really to argue for or against the existence of systematics in various symbol aggregations. On the contrary, my overarching aim is to demonstrate the sliding scales between symbols that are widespread and limited, and between symbols that are and are not part of systems – not to divide symbol use into distinct categories. The fact that the more definitive status of several of the illustrating examples that I here offer can be disputed actually strengthens my main point about gradability. This is because the grades exist in a hermeneutic sphere and cannot be objectively determined or measured.

6 Limited symbol systems

Even though there is no definite border between symbols that are not part of systems and symbols that are part of systems, it is now time to move on to more clear-cut examples of systematically ordered symbols. In doing so, it should be remembered that communication always depends not only on the specific properties of media products, but also on the background knowledge of the communicating minds and on contextual factors, which means that no symbols or other signs work in isolation. However, as I have already emphasized, functioning in a context is not the same as belonging to a system. Symbol systems, as they are circumscribed in this article, appear when the signification of different symbols are not only added to and generally affected by each other, but when the senses of the various symbols are also specified in relation to each other. Thus, a system of symbols is a collection of interdependent symbols functioning together to create emergent meaning and enhance communication. No such systems are truly universal; they are created and perform their service in certain historical, geographical and cultural contexts (see, for instance, Abdel-Raheem 2020). Of course, just as individual signs generally involve facets of iconicity, indexicality and symbolicity in combination, symbol systems also unavoidably comprise iconic and indexical features, even though it is the symbols that, because of their ground on habits, make thorough systematization possible. Symbol systems are invented and relatively stable structures that are, at the same time, relatively flexible and can be reinvented.

Following these circumscriptions, and in line with quite common practice, I suggest that languages might be comprehended as communicative sign systems. As systems are organized, communicative sign systems must rely on habits that are robustly interrelated and form what are sometimes called codes. In terms of Charles Sanders Peirce’s semiotic vocabulary (c. 1895, CP 2.297), as explained earlier, this reliance suggests that communicative sign systems are symbolic sign systems – because systems consist of interrelated habits. Therefore, I delineate languages as systems of symbols used for communication. In the present research, communication is furthermore delimited to communication among human minds (although several other animal species also communicate through at least rudimentary languages). Given such a definition, it is clear that there are several forms of language. To start with, they may be limited to only one or a few individuals, or shared by larger communities through conventions, understood as habits regulated on a larger scale. Here I will focus on a few brief examples of communicative symbol systems delimited to only a few people, which could be called private languages, before discussing languages shared by larger communities in the following section.

Private languages may be construed to avoid certain knowledge being spread, or they may remain private simply because of lack of interest from others. In an otherwise traditional diary, written with conventional verbal language symbols understood by a lot of people, other symbols of a private nature that are designed for autocommunication may be inserted to communicate all sorts of things that cannot be understood by others – because the code, well organized into a system by the diary writer, is unknown to others. Likewise, children or adults may want to be able to communicate secrets within a limited group of persons. For such a purpose, a symbol system that cannot be decoded by others may be construed based on media products with any material, spatiotemporal and sensorial qualities (cf. Elleström 2020). For instance, the media products may be corporeal or external to the body (produced by body movements or materialities such as paper or stone), static or temporal (items in fixed positions or with changing internal arrangement) and perceived by various senses (such as combinations of touches, sounds or visual impressions). Secret languages used to delimit insight – in popular parlance called ‘codes’ (a term that within semiotics normally means symbolicity grounded on systematic habits in general, secret or not) – are naturally more or less extensive.

In other cases, the private nature of languages is involuntary and a drawback rather than an advantage. A scientist who construes an elaborate system of visual but not word-based symbols for efficiently identifying the most salient properties of bacteria – symbols that can be combined and recombined to reveal properties that are more complex – has created a language. However, if the system is never spread because it is poorly received, or perhaps refused publication, it remains with its producer as a private language – it remains unconventional. Likewise, a choreographer who invents a novel method for representing all kinds of body movements on stage, made up of a system combining visual and auditory symbols creating emergency through interaction, may or may not be successful. If the devised language is too complicated, too vague, too difficult to learn or perhaps simply cumbersome and inefficient, it is likely to exist only for a short time among very few people before it vanishes completely.

7 Widespread symbol systems

Other languages, communicative symbol systems, grow and become parts of common semiosis, grounded on conventions: far-reaching habits shared by many people. To be spread efficiently, languages must rely on robustly interrelated habits. However, not only limited but also widespread languages may consist of anything from relatively few symbols that are roughly interconnected through simple principles, to thousands of symbols that are strongly organized via elaborate principles. Thus, the interrelatedness of language symbols varies, from bordering to being merely an unattached collection of symbols to being robustly interdependent symbols that deeply affect each other and make the sum considerably larger than its added parts. These language qualities are not necessarily constant. Some languages not only spread, in the sense that their grounding habits are conventionalized, but also develop. They do not only grow in general importance but also evolve, becoming more complex and fine-tuned. Naturally, because languages may progress in this way, they may also regress, become less complex and fine-tuned because of vanishing principles. In brief, languages change, and these changes are not in any clear-cut way connected to them becoming more or less widespread.

Apart from so-called verbal languages, which I will later investigate at some length, humans have developed myriad widespread symbol systems based on a broad variety of signs with representamens consisting of different modes of materiality, spatiotemporality and sensoriality. Here I will demonstrate this breadth by moving, roughly, from examples of symbol systems that can barely be called systems, because they are rather loose gatherings of symbols, to instances of increasingly interdependent symbols that profoundly affect each other in systemic interaction – languages that at various points in history have become complex and fine-tuned.

My first example of a weak system of symbols is national flags. Although the collected flags of all the world’s countries clearly belong together in that they fulfil the same basic communicative function – to represent specific nations – most of them work fine without being systematically interrelated. Yet, groups of national flags can be said to work in more collaborative ways. For instance, the Nordic flags are all based on large crosses that look fairly similar, differing in terms of the colors and a few other characteristics. Thus, they are clearly more interrelated with each other than with any other random flag, even though they are not the only ones containing crosses. They constitute a group of national flags that are similar enough to represent a larger geographical and cultural area and different enough to simultaneously represent different nations and regions within this area. Therefore, with some hesitation, I think they can be said to gain meaning in relation to each other, although the symbol system is weakly developed.

Other weakly developed languages, normally fused with robust iconicity, are international symbols in public spaces, airports and hotels; currencies such as systems of coins, bills and stamps representing pecuniary value; military grades on uniforms; and laundry symbols. Although limited in scope and complexity, these sorts of symbol systems communicate efficiently, at least partly because they comprise recognizably interrelated symbols that, to some extent, enhance each other’s meaning when combined. Probably more sophisticated and capable of more nuanced communication are the various symbol systems in the form of gestures that are used within some professions (such as divers and croupiers) and within certain cultures (during hunting, religious ceremonies, etc.) when speaking is impossible or not advised. Whereas these languages are composed of visual representamens produced by bodies, it was common for tramps to use visual representamens displayed on solid surfaces close to houses to communicate with each other. In systematic ways, these non-verbal signs, used within specific communities, conveyed the basics of how one should behave and how one could be expected to be treated at specific places.

Road signs and seamarks, where different shapes and colors clearly bind subcategories of symbols together, are fairly developed language types consisting of mostly interdependent symbols. Because of overlapping but nevertheless divergent conventions, they are largely but certainly not universally valid around the globe. Approximately the same applies for map symbols and, say, alchemical and astrological symbols, whereas modern chemical and astronomical symbol systems tend towards worldwide standardization. Although not global, heraldry is highly developed in its details and rules for representing characteristic features. Originating from Europe in the Middle Ages and still used almost exclusively in Europe, heraldry is the systematic use of visual symbols to identify mainly individuals, families and territories. So-called heraldic arms are composed of conventional parts (such as shields) and rules for combining traits (resulting in, for instance, color contrasts) and symbolize a delimited range of characteristics belonging to the respective individuals, families or territories. They also regularly include obvious visual icons, often of animals – icons that gain their full representative value because they normally also work indexically (having real connections to, say, the territory) and indeed symbolically (being parts of strict conventions). Heraldry, like all forms of semiosis, involves chains of signs; a visual sensory configuration that iconically represents an animal may well lead to a symbolic interpretation of the animal, standing through convention for some specific traits or ideas.

Further examples of developed languages producing clearly emergent meaning are numeral systems of thoroughly interrelated visual symbols. Among them, the Arabic numbering system has proven to be the most efficient for communication and now dominates the world. Combined with developed principles for arranging the numbers spatially on a flat surface, together with plenty of other especially designed visual symbols, advanced mathematical calculations can be efficiently performed and communicated to anyone who has learned to master the intricate mathematical habits. Mathematical symbol systems are advanced in the sense that they consist of a multitude of symbols that together, through their various combinations and interactions, create cognitive import that far surpasses the meaning of each symbol in isolation – to a greater extent than languages such as heraldry. Logical systems are normally related and connected, in various ways, to the principles that have become established within mathematics.

Moving from science to the arts, musical notation systems are, as a rule, profoundly different from mathematical schemes. However, they may be equally complex and indeed advanced in terms of possible interrelations among the individual symbols, representing musical parameters such as pitch, dynamics, rhythm, tempo and instrument type – interrelations that boost the total outcome of semiosis. The most advanced mathematical, logical and musical notation systems enable an infinite number of new combinations. They are also markedly iconic. The spatial arrangements of a multitude of individual symbols contribute to the semiosis not only because of strongly developed habits (combinatory rules) but also because of similarity relations between representamens and objects; the visual, spatial structures partly resemble the structures that they represent, whether they are cognitive structures in mathematics and logics or auditory structures in music.

Other examples of sophisticated languages connected to art are systems of body movements, gestures and facial expressions in theatre. Although the conventions are often quite dissimilar in different cultures, these languages exist or have existed in large parts of the world to symbolize primarily feelings and emotions of the represented persons on stage. Again, as in the earlier examples, these symbols are visual. Despite being based on different material and spatiotemporal modes, non-verbal languages seem to generally rely on vision, a sense faculty that lends itself to spatial and temporal arrangements of all sorts of materiality to create detailed discrimination among perceived impressions – and hence to efficiently creating comprehensive systems for communication.

However, visuality may be combined with other sensory modes in non-verbal language to form multimodal symbolicity. Based on Turkey’s selam, a flower language used in harems, a detailed and advanced system of symbolic flower arrangements emerged in 18th and 19th century Europe, especially among British upper-class women during the Victorian era. Thus, those who were familiar with the codes could use flower arrangements to communicate efficiently. Several books on the subject matter were published, describing and proposing conventions: the symbolicity of individual flower species, leaves and berries, what it meant to include or not include thorns, the significance of various colors, smells and textures, and, not least, the meaning of how everything was arranged together in a system of deeply interrelated symbols.

The most advanced symbol systems that have been developed are probably those called verbal languages, incorporating spoken, written and other verbal languages. I briefly circumscribe verbal languages in line with standard definitions, stating that a verbal language consists of words and a grammar that controls the use of these words. Words are communicative symbols consisting of speech sounds – or symbols manifested in other sensory modes (in effect visual or tactile symbols) that represent speech sounds. A verbal grammar is a system for the inflection and syntax of words. Words are inflected to indicate characteristics such as gender, number and tense distinctions and are put together syntactically to form clauses and phrases. Words can represent all kinds of concrete and abstract things, phenomena and ideas, and grammar makes it possible to combine words in a virtually infinite number of ways, enabling truly emergent meaning making. However, even though verbal languages constitute extensive, detailed and complex systems capable of fine-tuned communication, the verbal features do not automatically disambiguate communication.

I introduce the notion of verbal languages under the heading “widespread symbol systems” because verbal languages typically result from broadly shared conventions. In principle, however, they may be limited to the habits of only one or a few individuals. Like other symbol systems, verbal languages appear at a certain place and time, and even the most widespread languages are likely to sometime decline or even disappear – either through gradual transformation or because only a reduced number of people master them. Less widespread verbal languages die regularly. Verbal languages can also be limited because they never spread widely from the beginning, such as the languages that J. R. R. Tolkien invented based on his profound knowledge of historical verbal languages.

Specific verbal languages are clearly based on habits that are acquired, not inborn. A transnationally adopted infant starts speaking the language around her, not the language in the country where she was born. However, the capacity to think in linguistic structures and learn to master any verbal language seems to be inborn (Chomsky 2016). Perhaps this can be aligned with Peirce’s broad idea that habits can be inborn as well as acquired (Peirce 1895, CP 2.297). Given that the habits of all actual verbal languages are acquired, verbal languages, like other languages and symbols not belonging to systems, are bound to not only spread or shrink in terms of how many people are familiar with their conventions, but also to be constantly modified. However strictly controlled by rules verbal languages may seem to be in grammar, word-meaning etc., they evolve and progress – or regress – because the underlying habits are inevitably subject to change.

Because symbol systems tend to overlap, merge and be interdependent in various ways, there is no hard limit between verbal and non-verbal symbol systems, just like there is no hard limit between symbol systems and more loosely coordinated symbols. For instance, the mathematical languages that I discussed earlier are intricately entangled with verbal languages. Some verbal languages are more limited in terms of inherent vocabulary and grammar and rely entirely on the large systems that we normally refer to as spoken and written verbal language. For instance, maritime signal flags form a symbol system based on several different principles that all depend on verbality ‘outside’ the symbol system itself. It includes a limited number of flags that stand for certain words or ideas, but also flags that stand for individual letters and numbers. Furthermore, combinations of flags can represent meaning that is only known to a few collaborating people. Likewise, the semaphore system, whether mediated by flags held by human arms in various positions or moving manufactured arms, includes a limited number of symbols representing letters and words. In a similar manner, the now obsolete optical telegraph could communicate verbally through telescopes and movable arms or hatches mounted on towers whose various positions acted as signs for mainly letters and words. Morse code is a symbol system that displays letters visually and aurally through combinations of dots and dashes, and is thus entirely dependent on pre-existing verbal systems. Like most symbol systems, Morse code can be mediated by a multitude of technical media, even though only a few of them have practical and historical importance. Thus, from the mid-nineteenth century, Morse code mainly worked through electric telegraphy and eventually came to be most efficiently applied through wireless telegraphy from the early-twentieth century onwards.

I recently wrote that the Morse code system is entirely dependent on pre-existing verbal systems. In fact, one could argue that it is simply a specific variation of visual and auditory verbal symbol systems building on the alphabetic principle. The Morse code system is an alphabet designed to be mediated powerfully with the aid of certain technologies to communicate over great distances. Thus, when using Morse code we are close to alphabetic writing (the use of visual symbols that in combination form words) and, in fact, also to speech (the use of auditory symbols that in combination form words), even though the sounds of Morse code symbols are quite different from the sounds that are normally produced by our vocal cords. However, it would be easy to speak in Morse code if one wished to do so and, in principle, humans might develop a widespread convention to use the Morse alphabet for communication through the vocal cords – although for practical and cultural reasons this is probably quite an absurd idea.

We have now come to what, for many, appears to be the core of verbal communication, which in a simplified way is usually described in precisely these terms: speech (that we hear) and writing (that we see). Compared to languages in general, which seem to be largely visual, the most advanced verbal languages are clearly also universally, and, some would argue, even primarily materialized through sound. Therefore, verbal language is not really one single system, even if one disregards ‘parasitic’ symbol systems such as semaphore language and the Morse alphabet, and also disregards the crucial differences between language forms such as Hindi, Spanish, and Afrikaans. Some decades ago, Jan Mulder argued in detail for the notion that written and spoken languages constitute separate semiotic systems. Based on the simple but far-reaching observation that what is seen (written words) is different from what is heard (spoken words), Mulder argued that it is “terribly wrong” to see written and spoken languages as “mere variants of the same thing”; they are “entirely different semiotic systems” (Mulder 1994: 43). Gunther Kress held a similar position when he argued that “collapsing speech and writing with their entirely different materiality into one category, thereby joining and blurring over the distinct logics of time and space, of sequence and simultaneity, exposes the implausibility of a mode called ‘language.’ It is difficult to see what principles of coherence might serve to unify all these features. So I take speech and writing to be distinct modes” (Kress 2010: 86). I think these ideas are correct, even though the two verbal symbol systems have been developed to allow for advanced mutual transmediation: changing from writing and reading to speaking and listening is often a fairly efficient procedure, but definitely not a seamless one. Furthermore, writing not only represents objects external to verbal language such as things, events and ideas that exist in the world. As noted, writing also represents speech, so when one reads the visual symbols they simultaneously stand for the sounds of speech, whether one utters those sounds aloud or not.

Nevertheless, without denying the deep historical, cultural, practical and semiotic interdependencies between written and spoken verbal languages, they are profoundly separated in terms of mediality. Distinguishing between writing and speaking is essential and, in effect, only a starting point to unravel the complicated nature of verbal language. For instance, it must be added that spoken language cannot be understood as only one kind of symbol system. When mediated by human voices that are separated from other sensations created by a communicating body, like when one hears someone speaking from behind a wall, the symbol system is auditory. However, in true face-to-face communication, speech is sensorially multimodal: one perceives the sound of the voice but also visual impressions of body movements, gestures and facial expressions. Touching may also be a vital component of people talking to each other. Therefore, instead of using the dichotomy between speech and writing, it might be more precise to describe the differences in terms of visual, auditory and various forms of sensorially multimodal verbal sign systems.

Even if one settles with a basic distinction between speech and writing, there are at least three different advanced verbal symbol systems. Sign language also interconnects strongly with spoken and written verbal language, even though it is clearly a relatively self-sufficient visual symbol system and not only a variant, like the Morse system. Furthermore, each of these three or more systems may be subdivided into even more specific systems, such as written Swedish, spoken English or Chinese sign language.

Having made these observations of language in general and verbal language in particular, I am now in a position to examine some fundamental features of language concepts in relation to culture and media types and their effects on perceived media interrelations.

8 Cultural differences in conceptualizing language

Although the definition of languages as systems of symbolic signs used for communication captures the core of languages, it is not an exhaustive description of how language can be conceptualized. To exemplify the intricate nature of language, I will briefly illustrate the bearing of culture on written verbal languages – systems of visual symbols involving features such as words and grammar – whether they are composed of phonographic alphabets or character sets of pictograms and ideograms. There are myriad symbol systems for verbal writing that differ in fundamental ways, and one of these ways concerns the inclusion of sign types other than symbols.

In accordance with Peirce’s notion of sign types always being mixed (c. 1895, CP 2.302), one should expect that there are always at least some elements of iconicity and indexicality in semiosis dominated by symbolicity. Depending on factors such as history and culture, laypersons and scholars may see these other sign types as more or less inherent or alien facets of what is understood to be normal written verbal language within a certain community. Even though written verbal language can be given a transhistorical and transcultural core definition in terms of visual symbolic sign systems, it is clear that the surroundings of this core vary considerably in relation to time and space. As already noted, habits are probably never fully universal, and the various ways of realizing written verbal language demonstrate the impact of cultural contexts. To acknowledge this influence is not to add yet another complex factor in the forming of symbol systems, but in effect only to reformulate the fundamental role of habits for language and pinpoint it on another level. This is because cultures are communities that share habits (including habits such as traditions, rules, norms and laws); cultures harbor conventions that essentially influence, among many other things, language conceptions.

Thus, the presence of iconicity in written verbal language varies depending on culture. Iconic meaning making is not grounded on habits or conventions, as is the case of symbolic meaning making. Iconicity is based on perceived similarities among sensory perceptions and cognitive structures. In the domain of written verbal language, seen from a Western perspective, this means that symbols such as letters, words and punctuation, as well as the spatial arrangements of these, may produce meaning, not only because of learned, habitual connections between the visual impressions and what one takes them to mean, but also because of perceived similarities between what one sees and what the visual impressions are understood to signify. For instance, visual empty spaces between words and sentences suggest semantic spaces or differences. A more blatant example can be provided with a sentence like: “I can see two mOOns.”

Most Western scholars have been strongly inclined to dismiss iconicity as a peripheral part of verbal language, both oral and written. After the spreading of Saussure’s linguistic dogma of the arbitrariness of language a century ago (2011 [1916]), belief in the existence of iconicity or its relevance for verbal language has sometimes been considered naïve and has even been subject to some scorn and ridicule.

However, this has not been the case for Eastern scholars. According to Ersu Ding, Saussure’s quick dismissal of iconic signs in language as marginal and linguistically uninteresting has been met with much greater skepticism in China (Ding 2014: 121; cf. Hu 2010). As Ding succinctly put it, “his holistic scheme of sign formation simply could not accommodate anything other than a systemic pairing of the signifier and the signified as a result of structural differentiation” (Ding 2014: 125) – and languages do not develop arbitrarily; they are grounded. This is to say that symbol systems are not only based on system-internal structuring habits, so to speak; they also habitually incorporate significant amounts of iconicity and indexicality that are always already parts of the symbol systems, to a greater or lesser degree. For instance, iconicity can be a forming principle in the development of verbal language: “When a sign was going to be established, it always presupposed that one proposed signifier was better than another proposed signifier, otherwise it could never be a sign” (Hu 2010: 53). Here, being “better” means that the sign works more efficiently, and an iconic element may have precisely such a function.

Thus, In China and other Eastern countries such as Japan, the basic concept of verbal language, including written verbal language, has been more inclusive. Scholars largely agree that Chinese characters significantly rely on iconicity and “have characteristics of both pictorial signs and linguistic signs” (Liu 2013: 144), even though the idea of language as systems of symbolic signs used for communication has not, to the best of my knowledge, been denied. Thus, one could dare to say that the concept of written verbal language in the West and the East has largely been the same, but not really the same. In any case, there is now massive support in empirical research from both the East and the West that iconicity is, to varying degrees, a solid component of verbal languages from all over the world.

9 Language in relation to the concept of medium

Having noted language differences in terms of sign types, we can now focus on differences in the physical mediation of languages, which leads us to examine the relation between the two concepts of language and medium. I define a media product as a material, or more broadly a physical, intermediate entity that makes communication among human minds possible (Elleström 2018a, 2020: 9–24). Thus, media products are material objects, actions or processes that connect human beings mentally. Consequently, they must be conceptualized in terms of both physicality and cognitive capacities. Therefore, I submit that media products have both various presemiotic traits (they necessarily have material, spatiotemporal and sensorial features) and semiotic traits (they manage to trigger meaning-production in various ways based on their material, spatiotemporal and sensorial features).

Laypersons and scholars alike tend to group media products into categories. Sometimes one will mainly pay attention to the most basic features of media products and classify them according to their most salient material, spatiotemporal, sensorial and semiotic properties. For instance, one thinks in terms of still images (most often understood as tangible, flat, static, visual and iconic media products). Still images are an example of what I call basic media types (basic types of media products). Basic media types are relatively solid because of their perennial fundamental traits. At other times, one pays attention not only to the basic features of media products, but also to their historical and cultural context and what and how one expects them to communicate, and classifies them according to such qualifying features. This results in what I call qualified media types (qualified types of media products). Qualified media types include categories such as music, painting, television programs, news articles, visual art, and email. Because the qualifying aspects are vaguer than the basic media features, the limits of qualified media types are bound to be ambivalent, debated and changed much more than the limits of basic media types (Elleström 2020: 54–64).

In contrast to media types, which involve both presemiotic and semiotic traits, languages are normally (as in this article) comprehended more narrowly as systems of symbolic signs. From these dissimilar definitions, it follows that languages cannot simply be equated with media types. However, languages (types of symbol systems) and media types are clearly interconnected. One way of putting it is to say that languages can be vital semiotic parts of qualified media types. For instance, verbal language is an essential part of email, telephone conversations and stand-up comedy. In an equivalent manner, nonverbal systems of theatrical body movements, gestures and facial expressions are parts of theatre.

Another way of putting it is to emphasize, as Heather Lotherington did, that even though language is traditionally “described as a medium” it is actually “an abstract until it is materialized: mediated physically, in speech and signed conversations, and technologically, in printed documents, social media sites, roadside signs, movies, games, and suchlike” (Lotherington 2020: 219). In other words, if one specifies the understanding of languages as abstract symbol systems to the concrete use of a symbol systems – moving from the immaterial langue in the direction of the factual parole, in rough analogy to Saussure’s terminology (2011 [1916]) – a language comes closer to being understood as a media type. If one fully concretizes how a symbol system is mediated – that is, how it is realized in specific material, spatiotemporal and sensorial ways – a symbol system becomes equivalent to a media type (more specifically a type of physically existing paroles).

This equivalence is actually partly achieved in some of the different verbal sign systems mentioned earlier: written, spoken and signed verbal language. The notion of written verbal language definitely includes a specific sensorial mode – visuality – and often a spatiotemporal mode: the visual signs being realized on a flat surface. Similarly, the notion of spoken verbal language (understood as voice sound only) definitely includes a specific sensorial mode – audibility – which requires materiality capable of transmitting sound waves and temporal extension. In effect, the notion of signed verbal language fully covers the criteria for being understood as a media type: sign language is mediated by human bodies (not only hands and arms) acting in time and three-dimensional space to produce movements forming visual signs; these signs are clearly part of a verbal symbol system involving strong facets of iconicity and indexicality. Sign languages are relatively autonomous in relation to written and spoken verbal languages in the sense that they cannot be transmediated on a ‘word-to-word level,’ which is largely possible between written and spoken languages (between written and spoken Farsi, for instance). Considering that sign language is also designed specifically for face-to-face communication with and among people with hearing impairments, it is definitely a qualified media type rather than a basic media type. Braille is another verbal sign system that could be added to these three and also count as a full-fledged qualified media type because of its settled presemiotic and semiotic traits and communicative functions. Braille is designed for tactile communication with people with visual deficiencies. Although Braille is largely verbal because most of its signs stand for single letters or words, it also includes tactile graphs working iconically. Therefore, it is more than an alternative alphabetic system such as Morse.

However, written verbal language based on vision, which I discussed in some detail earlier, can hardly be understood as a single qualified media type. Rather, I would say that this category includes several closely related basic media types. As mentioned, written verbal language is definitely visual and normally realized on flat surfaces, but it can also occur on rounded and irregular surfaces, and letters and words may also be three-dimensional in themselves. Written verbal language is most conveniently mediated through solid and inorganic materiality, but it is fully possible to write in gas and liquids or with the use of organic material such as living bodies. Thus, written verbal language is a rather open category that includes a variety of basic media types, whether one considers the more exclusive Western or more inclusive Eastern varieties.

Apart from ‘languages forming media types,’ as described above, one might also argue for the relevance of ‘media types forming languages’ understood as a metaperspective: ‘media types forming languages (possibly including other languages).’ My point here is that as soon as media products are categorized into qualified media types, they tend to function as systems; they are inclined to be compared with each other and gain (sometimes quite advanced) significance in relation to each other. This also applies for qualified media types that are dominated by iconicity and indexicality. Qualified media types, whether it is the television news, dinner conversations or oil paintings, are operationally qualified in the sense that certain communicative functions are tied to them, which means that individual media products are met with certain expectations because of their classification. The whole qualified media type, as it appears historically and culturally, forms a dynamic pattern that affects the significance of each individual media product belonging to it or being newly inscribed into it (whether in its center or its periphery). As the qualified media type is formed, the habits that it is grounded upon may become fine-tuned and intertwined in complex ways so that it may be said to form a language – albeit not a verbal language (even if it may include verbal language). As with all symbol systems, languages formed by qualified media types are created by producing and perceiving minds that may try to stabilize as well as destabilize the system – following or breaking expectations. Some qualified media types, such as newspaper advertisements, are controlled relatively weakly. Because of the relatively flexible habits of the media type, individual newspaper advertisements can be formed quite freely. Other qualified media types, such as legal proceedings, are controlled strongly. The habits of the media type are not especially flexible, so the form of individual legal proceedings are expected and required to be largely predictable. Like other qualified media types, legal proceedings can be seen as a language: a symbol system that is built of a multitude of interacting individual proceedings (comprehensive media products) that, in turn, comprise complex mixes of media products and sign types, including verbal language.

10 Media interrelations and culture

It is now possible to tie up the investigation of symbolicity, language, culture, and media types by briefly examining the impact of culture on media interrelations. Cultures, understood as habit communities, play a decisive role in qualifying media types contextually. Based on a multitude of communicational functions and expectations, qualified media types in general emerge, grow and decline within cultural frames. This wide-ranging principle is here exemplified with the influence of cultural differences in conceptualizing language on the construction of qualified media types that include such differently conceptualized languages. Attaching to our previous example, this means that Eastern–Western differences in conceptualizing written verbal language – including or excluding visual iconicity – influence how qualified media types, including written verbal language, are construed, which has a strong bearing on how one perceives intermedial relations.

Regarding media types, one can distinguish between two kinds of intermedial relations. On one hand, there are relations among dissimilar basic media types (media that are materially, spatiotemporally, sensorially or semiotically different). This can be understood as intermediality in a narrow sense, which is less culture-dependent and more universal because of general cognitive capacities among people. On the other hand, there are relations among dissimilar qualified media types (which does not necessarily involve basic presemiotic and semiotic differences). This is intermediality in a broad sense, which is more culture-dependent and less universal because qualified media types are less stable than basic media types (Elleström 2020: 66–73). Thus, even though the analytical perspective of intermedial relations in a narrow sense (among basic media types) is globally applicable, the perceived phenomena of intermedial relations in a broad sense (among qualified media types) may vary greatly through history and across cultures. For instance, what appears to be a perfectly normal way of speaking in one cultural context may be perceived as an intermedial mixture of speech and song in another cultural context. Therefore, the cultural aspect, the fact that habits and hence symbols are often confined to certain communities, can hardly be avoided in intermedial studies.

I will end by illustrating this conception with a simple example that consists not of a qualified media type including written verbal language, and not even a media product in the form of a whole text consisting of written verbal language, but only one Chinese ideogram (Figure 1).

Figure 1:

Chinese ideogram.

This ideogram is clearly part of and derives substantial significance from a symbol system that includes words and grammar when interacting with other symbol system components. Thus, it is part of a visual, written verbal language and potentially part of all those media types that include such written verbal language. I call it an ideogram because it not only symbolizes certain sounds that form spoken words (as letters do), but it also represents a range of concepts because of visual qualities that are independent of speech sounds. One of those concepts is written language.

However, this signification is far from solely a matter of symbolicity – representation on the ground of habits or conventions. Like many other Chinese writing signs, the ideogram is also permeated with visual iconicity – representation based on visual similarity – although the similarity to the object has been abstracted through history. The iconic object is a man with a painted or tattooed chest, hence the connection to written language. The resemblance can be seen more clearly in earlier variations of the sign (Figure 2).

Figure 2:

Earlier variations of the Chinese ideogram (from Lindqvist 1991 [1989]: 309).

From a Western perspective, these ideograms, and the qualified media types that contain ideograms of the same type, are likely to be construed as something like an integration of written verbal language (defined by symbolicity) and visual still image (defined by iconicity). This integration makes them appear as something intermedial, comparable to unorthodox Western poetry characterized by visual iconicity, such as “Forsythia” by Mary Ellen Solt (Figure 3).

Figure 3:

Poem by Mary Ellen Solt in Flowers in Concrete (1966: unpag.).

This poem mainly consists of letters, standard verbal language symbols representing speech sounds that together form words, among which the central one, “FORSYTHIA,” represents a bush that blossoms with bright yellow flowers in early springtime. However, the letters are organized so that the poem’s visual form closely resembles precisely such a bush, rendering it into an overarching icon made up of symbol components. In Western research, poems like this are typically described in terms of integration of written verbal language and visual still image and used as standard examples of intermediality.

The fundamental medial resemblance between the ideogram and the poem is clear. They are both flat, static and visual media products triggering significance dominated by iconicity and symbolicity. Hence, they belong to the same basic media type. However, they belong to different qualified media types involving different conceptions of verbal language. Whereas the poem’s salient iconic elements are normally perceived to be alien to written Western language, the ideogram’s iconic elements are integral parts of written Chinese language. Thus, from an Eastern perspective, in contrast to the Western habitual view, Chinese ideograms are simply parts of standard written language that is semiotically multimodal but not an integration of media types, and hence not intermedial. To conclude, this demonstrates that symbolicity and habits in general and, on a broader scale, conventions that flourish in communities, such as culturally different ways of thinking about verbal language, affect how we construe intermedial relations among qualified media types; that is, intermediality in a broad sense.

11 Summary

With the far-reaching goal to be able to conceptualize semiotically human communication and mediality in general and media interrelations in particular, I started this article with an elaboration of Peirce’s ideas of symbolicity. Rejecting the widespread idea that symbols are arbitrary signs, I found that Peirce’s most useful notion for my purposes is that symbols are grounded on habits that connect representamens to objects. To demonstrate the accuracy of that concept, and to shed light on the wealth of multifaceted use of symbolicity in human communication beyond verbal language, I provided a broad and systematic exemplification of symbols with different presemiotic (material, spatiotemporal and sensorial) traits.

The overview was structured around two parameters crossing each other. The first parameter was symbols being limited or widespread among people. In the case of widespread symbols, the grounding habits have the status of conventions – a special case of habits in general. The second parameter was symbols not being part of or being part of systems. Comprehending languages as communicative symbol systems, the second parameter can also be described as symbols not belonging to or belonging to languages. Thus, the two parameters concern degrees of spreading and systematics of symbols.

The systematic overview also comprised examples of verbal languages, understood as highly developed languages based on features such as words and grammar, demonstrating their conceptual place in the large area of symbolicity in communication. Revealing the complexity of how verbal languages are mediated – understood as being manifested materially, spatiotemporally and sensorially – I argued that even a distinction between spoken, written and signed languages is only a starting point for illuminating the issue of verbal language diversity. I also highlighted the role of culture in language conceptions, exemplifying with cultural differences in including or excluding iconicity in written verbal language. As cultures can be understood as habit communities – meaning that cultures harbor conventions that essentially influence, among other things, language conceptions – they are essential for much symbolicity.

Following these detailed investigations of symbolicity and language, I proposed a conceptual solution to the question how media types and languages are interrelated, avoiding either conflating the two concepts or separating them fully. I argued that while media are characterized by both presemiotic (material, spatiotemporal and sensorial) and semiotic traits, languages should essentially be understood as symbol systems, which means that they can be vital semiotic parts of qualified media types. The more a certain language is also circumscribed by presemiotic traits – that is, the more it is seen as a concrete communicative device rather than an abstract sign system – the closer it comes to being a qualified media type.

Finally, I briefly explored how the cultural aspect of language and media type conceptions affect the idea of intermediality, meaning the interrelations among different forms of media. Again, using the example of written verbal languages, I concluded that since verbal language is circumscribed differently in different cultures, media interrelations might be perceived differently. In other words, some media borders depend on cultural dissimilarities between, for instance, language conceptions.

All in all, this article has demonstrated the multi-faceted role of symbolicity in human communication with the aid of a systematic employment of parameters, suggested a way to theorize language in general and verbal language in particular in relation to symbolicity in communication, offered a nuanced conception of language in relation to mediality, and highlighted the vital cultural impact on media interrelations.