Colon + Hyphen + Right Paren: At the Origins of Face Semiotics from Smileys to Memes

Abstract

The article engages with the semiotics of “stylized faces” in online communication, focusing on “smileys.” It reconstructs the origins of both emoticons and emojis, outlining how they differ functionally (paralinguistic signs vs. narrative figures) and commenting on their pragmatics, with regard to the issue of literacy as related to generational fruition. A chronology is provided of the first tokens of smileys in written communication, both before and after the Internet. By relying upon the anthropology of the face dating back to prehistory, the issues of iconism and universality are discussed, supporting the view that there is a strong cultural, conventional component in face depiction, varying diachronically (emoticons versus emojis) and diatopically (emoticons versus kaomojis, i.e., Japanese emoticons). Emoticons and emojis are regarded as prominent examples of intermedia, working at the intersection of written word and image. Finally, stylized digital faces are set in the broader framework of Internet memes, thus discussing the dichotomy between structural memes (the focus is on the formula) and iconic memes (the focus is on the image and, thus, the face). Throughout the text, great care is devoted to the philology of sources, some of them being presented in this form for the first time.

The English language cannot fully capture the depth and complexity of my thoughts, so I’m incorporating emojis into my speech to better express myself. Winkie face.

—Gina Linetti, on Brooklyn 9/9 ¹

Faces pervade online communication in many ways. Profile pictures on social networks are almost always images of our face; when they are not, they still signify in opposition to it, providing an alternative to the classic model of the passport photo. Or to the selfie: the photographic self-portrait—self-appointing as such via enunciational means—par excellence. Nonfacial profile pictures testify how the face is basically a syntactic semiotic device: if, instead of my “natural face,” I choose to “wear” the picture of my cat, the LGBTQ+ rainbow flag, a photo of Kim Jong-un, or a monochrome by Yves Klein, I am inviting my social contacts to consider such images as the online mask of my public persona, euphorically or dysphorically as the case may be.

Face as Writing and Vice Versa

Our face identifies us within a digital ecology connoted by the ideology of transparency: this is the logic of the social web developed in the last 20 years as the counterpoint to the original Internet, which featured anonymity (nicknames) and camouflage (avatars) as standard communicational strategies across chat rooms, image boards, forums, and blogs.² Today’s mainstream social platforms may force users with “peculiar” names (detected—more than often due to ethnocentric bias—as pseudonyms or heteronyms) to change their usernames, or the platforms may require the uploading of pictures with no face modification of any kind or even official identity cards, under penalty of blocking the account or disabling registration in the first place. In this light, the face, meant as the metonymy of the subject and the synecdoche of the individual, participates in the original indexical dimension of the photographic medium and incorporates mimetic and identity values. The face resembles and identifies us, in any case—when it is representational (referential function) as much as when it is constructive (utopian function), when it is being displayed naked and when it is being interpolated. The latter case is what we may define as a “projective face”: the projection of the face we would like, even just playfully and ephemerally, rendered through the projection onto the natural face of various “parafacial devices” whose aim is to make up, disguise, accessorize, translate it. Let us think of the Snapchat, FaceApp, Instragram, and TikTok filters that give us freckles or pointed ears, that make us slim or older or change our sex, those that transform a photo into a cartoon or plastically intervene to convey onto it a specific pictorial style.

Writingization and facialization:³ on the one hand, the face has become akin to writing, establishing crystallized formats and grammars (avatars, selfies, filters); on the other hand, writing has become akin to faces by acquiring personae, the ones we use when we want to make the tone of conversation (or the mood with which we are typing a given message) explicit—or when, somehow, we run out of words.⁴

Paralinguistic Smileys

In front of computer screens, we are unable to perceive a whole series of semiotic signals, both voluntary and involuntary, that, in face-to-face interaction, would enable a richer decoding of messages (as well as the situation itself): gaze, facial expressions, and suprasegmental traits such as tone of voice, prosody, inflection, accent. A sentence can be uttered in a neutral way, staidly: this happens when semantics and pragmatics overlap, and the enunciator aims to convey the same message that can be inferred from the mere decoding of the alphanumeric sequence of the text. On the contrary, the same sentence can be uttered by playing within the space of possibility that opens up between semantics and pragmatics—for instance, by resorting to irony, a communicational strategy whose illocutionary force aims to overturn the literal meaning proposed in the text. Smileys were born as “diacritical faces,” namely, as disambiguation metasigns that might compensate for the notorious paralinguistic poverty of computer-mediated communication. In particular, they were meant to recount the emotions—the “thymic and pathemic dimensions,” semioticians would say—involved and yet hard to render explicit from the writing itself due to the pragmatics of the digital environment, which requires real time responsiveness and thus facilitates brachylogy.⁵

September 19, 1982, 11:44 a.m.: on the Carnegie Mellon University discussion board on Usenet (a historic precursor of forums), the umpteenth misunderstanding provoked bickering between the users. A 34-year-old MIT-trained computer scientist named Scott E. Fahlman proposed using a (typo)graphic expedient, inserting the “character sequence” :-) (to be read “sideways”) at the end of the sentence, to mark it as a joke, and the sequence :-( “to mark things that are NOT jokes,” which would be “more economical,” given the preponderance of ludic language.⁶

The two polar expressions were immediately baptized “smiley faces” or “smileys,” with a counterintuitive overextension of the euphoric term to the detriment of the dysphoric one (which may suggest a ludic tint even for nonludic sentences);⁷ they would be called “emoticons” widely only beginning in the mid-1990s (some online dictionaries suggest from 1994). Whereas it is widely acknowledged that emoticon is the portmanteau of emotion and icon (an icon to convey emotion), Crystal (2003, 36) suggests another origin, whereby emoticon would derive from emote (or ‘pose’), a method of communication used in MUDs (Multi-User Dungeons, a type of role-play game, mainly of the textual kind) that “allows a player to express a character’s actions, feelings, reactions, gestures, facial expressions, and so on” (2003, 181).⁸ In any case, the term emoticon was already employed in what seems to be one of the first studies encompassing this subject matter: a thesis investigating the paralinguistic strategies used in email messaging through the “multi-level qualitative content analysis” (Asteroff 1987, i) of the correspondence of 16 librarianship students, divided into two groups based on their expertise in the new communication system, which revealed that novices use a greater quantity of paralinguistic elements such as exclamation points and, indeed, smileys than do advanced users.⁹ This study is important because, besides corpus-based examples and analyses, it presents, as appendixes (1987, 221–28), a series of memoranda prepared by institutions and companies whose aim was to familiarize members and employees with the correct decoding and usage of smileys, thus attesting a precise moment in the management of the process of informalization of language; what we use online is, in fact, a sort of interlanguage, a hybrid that simulates some characteristics of verbal speech and face-to-face interaction in written form, by means of an intersemiotic code that appeals to graphic-visual resources.

Emoticons gained immediate success as they responded to a specific, pressing, and widespread need with a solution at everyone’s fingertips. What Fahlman considers his own creature has never been copyrighted (the same was to happen in 2007 with the hashtag, or #, on Twitter), thus making it possible for these signs to be adopted in any context and multiply. The most disparate smileys have been created to convey the most disparate moods, such as amazement (:-O). Emoticons have become institutionalized ontologies in digital communication due to the convergence of codes, platforms, and softwares: the Unicode standard, incorporated in the main word processing programs, such as Microsoft Word, automatically transforms the strings :-) and :-(—which are based on the encoding in ASCII (American Standard Code for Information Interchange)—into ☺ and ☹, respectively. The long series of smileys reached the “reactions” launched by Facebook in 2016 to mark our attitude toward a given content: an evolution of the logic of “likes” already established on various social media (hearts on Twitter and Instagram, thumbs up on Facebook).¹⁰

Narrative Smileys

In the continuum between crystallization and wear and tear, on the one hand, and multiplication and renewal, on the other, emoticons increasingly expanded their range of uses and their right to speak for themselves. Very soon they started being utilized not only to signify abstract feelings as a companion to verbal written text but also to structure and demarcate speech as vicarious punctuation (the smiley face may stand for an exclamation point, as already suggested in Asteroff 1987), to replace words due to their synthetic capacity, and to portray actual characters. I can say, “I fell out of bed this morning: definitely a way to hit the ground running today :-)” to mark the idiomatic and word play, and I can say “Today I feel :-)” not to connote something but to state something plain and simple (“Today I feel happy/good/great”). If I want to allusively tell you that “I would like to give you a kiss,” I can also write, “I would like to give you a :-*,” and if I want to say “Homer Simpson” I may use the string ~(_8^(|).¹¹ Relying on their pictographic value (allowing both paralinguistic and referential functions), smileys quickly evolved from accessory elements with “very limited … semantic role” (Crystal 2011, 23) to elements with full, autonomous meaning, featured with an increasingly logographic value; digital, online writing has progressively incorporated multimodal elements (more similar to handwriting than to modern typography), corroborating the notion of “iconic turn” proposed by Gottfried Boehm.¹²

In 1999, 27-year-old designer Shigetaka Kurita developed a set of 176 pictographic symbols, 12 × 12 pixels each, on behalf of NTT DoCoMo, the leading telephone company in Japan, for i-mode, a Web service designed for youngsters to access via mobiles: the idea was to facilitate communication by allowing the production of a rich and still manageable written content in a fast and fun way. Emojis were born: the sun (☀), a cloud (☁), an open umbrella (☂), a snowman (☃), a pair of worried eyes (👀), a smiling kitten (🐱), a clock (⏰), and so on.¹³ The semantic-lexical similarity of the terms emoji and emoticon is purely coincidental (in Japanese e 絵 means “image” and moji 文字 means “character”) so that the popular association of emoji with ‘emotion’ is also due to a folk etymology. If emoticons were created to convey emotions (in the Kuritian canon, on the contrary, the presence of physiognomy is marginal), emojis are cartoons whose purpose is to tell stories: in 2010, Fred Benenson (2010), a computer scientist who was among the founders of Kickstarter, created a pictographic version of Herman Melville’s Moby Dick using the Amazon Mechanical Turk crowdsourcing system. Emoticons are grassroots, bottom-up, communitarian (Fahlman on Usenet); emojis were born branded, top-down, and corporate (Kurita for DoCoMo) and were instantly projected into the world not simply of communication but rather of marketing.¹⁴ Their success, granted by the progressive organic integration of pictograms within different types of digital keyboard sets (starting with Apple’s iOS in 2011), outclassed that of their ancestors (emoticons, which turned out to be hyponyms of the next greater digital phenomenon), thus completing the path of full linguistic affirmation of smiley faces; in 2015, the Oxford English Dictionary selected as word of the year not a lexical element but rather the emoji “face with tears of joy” (😂), considered particularly representative of the contradictory and paradoxical Zeitgeist (as well as of the very rise of emojis in communication),¹⁵ and in 2016 New York’s Museum of Modern Art acquired the original set of emojis designed by Kurita, thus sanctioning the cultural institutionalization of these signs.¹⁶

On the one hand, emojis are generally regarded as the ultimate step in linguistic simplification; for instance, in September 2017, Google Arts & Culture published “Explaining Contemporary Art with Emojis,” taking for granted that emojis are “the simplest form of communication.”¹⁷ On the other hand, emojis proliferate not only with regard to their quantity but also their visual semantics; whereas emoticons originally served to disambiguate, emojis often prove so complex or so abstract that they become ambiguous. A 2016 study conducted by University of Minnesota’s GroupLens Research group showed that it is not uncommon to use emojis without mastering their meaning (at least, the one assigned by graphic designers and crystallized on Emojipedia).¹⁸ A considerable number of the people interviewed were not able to attribute with certainty a neutral, positive, or negative sentiment to some of the most widespread emojis or, on the contrary, assigned an opposite sentiment, as in the case of the—intentionally contradictory—grimace face (😬).¹⁹ Emojis seem simple, but they are not.

Whereas “the category of ‘new literacies’ largely covers what are often referred to as ‘post-typographic’ forms of textual practice [including] semiotic languages (such as … emoticons (‘smileys’) used in email, online chat space or in instant messaging)” (Lankshear and Knobel 2003, 16), today’s so-called functional illiteracy has taken the subtle form of stylistic clumsiness: that is, using not an incorrect emoji but rather an emoji that has “gone out of fashion.” A 2021 study by media agency Perspectus Global, based on a survey completed by 2,000 British people between 16 and 29 years old (the cohort of the so-called Generation Z or Zoomers, born in the late 1990s and early 2000s), showed that youngsters have precise pictographic dos and don’ts; whereas the most traditional ways of saying “ok!” via emojis are being dismissed as outdated and cringeworthy (e.g., the hand making the gesture connecting the thumb and index into a circle, the thumb up—judged as old-fashioned by 24 percent of respondents—or the check mark), emojis such as the celebrated face with tears of joy, the face surrounded by hearts, the zany face, and the fire and the eggplant (both used with sexual connotations) register wide, transversal consensus.²⁰

Archeosmileys

Traditionally, the paternity of emoticons has always been assigned to Usenet, Fahlman, and paralinguistics. Over time, however, as with any proper mythology, the origins of the smiley face have been backdated ad lib. The first step is backdating within the digital world. As reported in a now-offline article published on “E-mail & More” (a spin-off of the celebrated specialty magazine Smart Computing), which seems to be the original and only reliable source on this subject matter,

On April 12, 1979, Kevin MacKenzie suggested a way to spice up dry and emotionless e-mail. In an e-mail to the MsgGroup (which was an electronic discussion list and one of the earliest mailing lists), he suggested adding emotion to e-mail using certain visual symbols and punctuation, such as -) for a tongue-in-cheek sentence. Thus, emoticons were born. (Shaw 2000)

A further backdating points at the early Internet platform PLATO. As summarized by specialist Brian Dear,

The PLATO system was created in 1960 at the University of Illinois. Initially it ran as a one-terminal system connected to the ILLIAC computer. By 1963, the system was running on a CDC 1604 with multiple simultaneous users. By 1972, the system had expanded to run a thousand simultaneous users on a CDC CYBER mainframe. Control Data Corporation began marketing PLATO commercially in 1976, resulting in PLATO system installations in dozens of cities around the world. Many of these systems were interconnected, enabling email and remote logins through the network. For nearly ten years, there were more users on PLATO than there were on ARPANET, the precursor to the Internet.²¹

PLATO was designed as a computer-based educational system but, surprisingly, it became

the first online community, and the original incubator for social computing: instant messaging, chat rooms, message forums, the world’s first online newspaper, interactive fiction, emoticons, animations, virtual goods and virtual economies, a thriving developer community, MUDs (multi-user dungeons), personal publishing, screen savers.²²

Since the mid-1970s, perhaps as early as 1972, PLATO had allowed character configurations that, thanks to the system’s unique specifics (one could superimpose several letters, numbers, and other signs one onto the other), made it possible to create complex figurations: actual characters (some of which would be given names by users deriving from the combinations necessary to generate them; for example, Wobtax, ),²³ the protagonists of games and stories; these were the precursors, therefore, not of emoticons, but of emojis. It is conceivable that these smiley faces did not make the headlines because PLATO was a niche experience reserved for the early afficionados of home computing; moreover, the website that made the system’s pioneering story available to everyone has been online only since 2000, and the book that collects the research of its curator was published only in 2017.²⁴ Backdating, then, jumps outside the digital world into the world of typography and handwriting.

It is common opinion—acknowledged even by Fahlman²⁵—that Vladimir Nabokov had anticipated emoticons in an interview published in the New York Times (1969). In fact, Nabokov comments on that interview in a collection of miscellaneous writings (1973), specifying that it was only partially published. The exchange between journalist Alden Whitman and Nabokov, in which the writer contemplated the need for a sort of emotional punctuation—and this very passage was not published in the Times—reads as follows:

Whitman: How do you rank yourself among writers (living) and of the immediate past?

Nabokov: I often think there should exist a special typographical sign for a smile—some sort of concave mark, a supine round bracket, which I would now like to trace in reply to your question.

In short, Nabokov imagined something like this: .

In 2007, on the twenty-fifth anniversary of the Fahlmanian emoticons, debunking journalist Barbara Mikkelson of Snopes.com published proof that, in private letters, it was already common practice to use graphic signs to render paralinguistic elements, such as the tone with which to read a certain statement; the source is a quote by journalist Ralph Reppert from the Baltimore Sunday Sun published in the Reader’s Digest of May 1967. Again, pungent irony is on the table:

Many people write letters with strong expression in them, but my Aunt Ev is the only person I know who can write a facial expression. Aunt Ev’s expression is a symbol that looks like this: —) It represents her tongue stuck in her cheek. Here’s the way she used it in her last letter: “Your Cousin Vernie is a natural blonde again —) Will Wamsley is the new superintendent over at the factory. Marge Pinkleman says they tried to get her husband to take the job —) but he told them he couldn’t accept less that $12,000 a year —). (Mikkelson 2007)

Going further back in time, leaving the world of verbal communication and entering that of visual communication and design, it is impossible to ignore the contribution of Harvey Ball to the global success of smileys. In 1963, Ball designed a print ad featuring a round yellow smiling face on behalf of an insurance company in Worcester, Massachusetts (which, according to legend, paid him 45 dollars). The smiley quickly became associated with the phrase “Have a happy day” (a case where the verbal text provides the paratext to the visual one) and served as one of the prominent symbols of the psychedelic movement both during the first (1967) and the second (1988) Summer of Love.²⁶ Ball never copyrighted the yellow smiley; this was later done by French journalist and entrepreneur Franck Loufrani to launch the Smiley Company, following an ante litteram viral campaign that appeared in several newspapers, starting with the first page of France Soir on January 1, 1972, where the smiley was accompanied by the motto “Prenez le temps de sourire” (Take time to smile).²⁷

In his short “Prehistory of Emoticons” (2007), lexicographer Ben Zimmer referred for the first time to an essay by Ambrose Bierce, “For Brevity and Clarity,” written in 1887, in which Bierce half seriously proposed language reform that would use punctuation to underline when a sentence is to be understood as a joke; Bierce proposed using a sign in all respects similar to the “concave mark” suggested by Nabokov, anticipating more typographico the classic Internet jargon of acronyms such as LOL (Laughing Out Loud) and ROTFL/LMAO (Rolling On The Floor Laughing/Laughing My Ass Off).

While reforming the language I crave leave to introduce an improvement in punctuation—the snigger point, or note of cachinnation. It is written thus and represents, as nearly as may be, a smiling mouth. It is to be appended, with the full stop, to every jocular or ironical sentence; or, without the stop, to every jocular or ironical clause of a sentence otherwise serious—thus: “Mr. Edward Bok is the noblest work of God .” “Our respected and esteemed contemporary, Mr. Slyvester Vierick, whom for his virtues we revere and for his success envy , is going to the devil as fast as his two heels can carry him.” “Deacon Harvey, a truly good man , is self-made in the largest sense of the term; for although he was born great, wise and rich, the deflection of his nose is the work of his own coat-sleeve.”

(Bierce 1912, 386–87)

In 2007, Wikipedia reproduced a page from issue 212 of the satirical magazine Puck, thus backdating the creative use of typography to draw stylized faces and their emotions to 1881:

TYPOGRAPHICAL ART. We wish it to be distinctly understood that the letterpress department of this paper is not going to be trampled on by any tyrannical crowd of artists in existence. We mean to let the public see that we can lay out, in our own typographical line, all the cartoonists that ever walked. For fear of startling the public we will give only a small specimen of the artistic achievements within our grasp, by way of a first instalment. The following are from Studies in Passions and Emotions. No copyright. Joy, Melancholy, Indifference, Astonishment. ²⁸

The search for the first token, a sign announcing that a new communicational type would be needed, eventually leads to the perceptive conspiracy of pareidolia. In 2009, City Room, a New York Times blog devoted to historical research, offered a philological discussion on the discovery of a proto-emoticon in a speech by US president Abraham Lincoln published in the newspaper on August 7, 1862:

FELLOW-CITIZENS: I believe there is no precedent for my appearing before you on this occasion, [applause] but it is also true that there is no precedent for your being here yourselves, (applause and laughter ;) and I offer, in justification of myself and of you, that, upon examination, I have found nothing in the Constitution against. [Renewed applause.]²⁹

The sequence “semicolon + right paren,” meant as winkie face, is supposed to have indicated that, at that specific point, the public had cheered and laughed; it is, in fact, a hapax in the corpus of Lincoln’s speeches, as well as in the journalistic conventions of the time, and so should be considered a typo.³⁰ For the same reasons, a similar conclusion should be drawn with regard to the (pseudo) proto-emoticon discovered by Levi Stahl (2014), a literary enthusiast (and an employee of the University of Chicago Press), in Robert Herrick’s poem “To Fortune,” included in the collection Hesperides, published in 1648; a couplet, stylized as follows in some editions (not in all of them), apparently includes a smiley face: “Tumble me down, and I will sit / Upon my ruins, (smiling yet :).”

Following this logic, we may end up dating smileys back to the Stone Age, which is what both mainstream media and specialized scholars have tried to do from time to time. The media have learned how to thematize paleoanthropological discoveries related to the first graphic expressions created by human hand by linking them to digital pictograms; for instance, in February 2021, the Jerusalem Post proposed a recently discovered bone fragment with a series of vertical carvings, dated to 120,000 years ago, as “the oldest emoji” (Tercatin 2021). Scholars have followed a similar phylogenetic path; for instance, while presenting her research on the geometric signs diffuse across Ice Age Europe to a wider audience, paleoanthropologist Genevieve von Petzinger (2016a) explained how “humans have been using symbols to communicate and convey emotion since the Stone Age,” experimenting with “visual marks that paved the way for the development of writing—and, more recently, the creation of modern symbols, including emoji.” Statements like these are interesting as they lead us to reflect on the fundamental dichotomy between mimetic signs, of iconic nature, and symbolic, conventional signs (partially justified, though, by a process of analogical abstraction). Nonrepresentational signs, von Petzinger suggests, often “outnumber the animals and humans—and yet they have not received the same amount of attention as their figurative counterparts,” with notable exceptions such as the studies by André Leroi-Gourhan and Georges Sauvet, “who did recognise the potential of the geometric signs.” The “first signs”—which is the title of her scientific monograph—studied by von Petzinger (2016b) seem to be more closely related to emojis rather than emoticons.

Stylization and Universality

The first faces ever traced by human hand look surprisingly familiar to us: the 17,000-year-old ones engraved in the Marsoulas cave in the Pyrenees (studied in the early 1970s by Leroi-Gourhan) remind us of Picasso and Matisse; the 10,000-year-old ones dug in the rock at Lene Hara, near East Timor, recall the round sculptures of Henry Moore; the 27,000-year-old ones (the oldest so far), painted in black over a red background in the caves around Angoulême, bring to mind Brancusi. Contemplation of the “first faces” makes us reflect on two closely interrelated issues: the iconic nature of these faces and their relationship with time (diachronicity) and space (diatopy)—their stylization, on the one hand, and their apparent universality, on the other.

Traditionally, emoticons are considered iconic signs. As such, they participate in the general functioning of the iconic sign; namely, they should address the referent by mimesis, in terms of similarity. Iconism, or—according to the critical vocation of semiotics (in the sense of Kant’s critical methodology)—the reconstruction of the conditions of possibility of the iconic sign, is a classic problem of semiotic theory, especially in the 1960s and 1970s works by Umberto Eco. The semiotician wonders: What does it really mean to say that something “resembles,” “is similar to,” “has the same properties as” or “the same shape as” the object it stands for? The semiotician replies that considering the problem in these terms “prevents us from analyzing the iconic sign as a social product, that is, as subject to convention. And, therefore, it prevents us from really seeing its history, from exercising control over it, from highlighting its ideological stratification” (Eco 1970, my translation). Iconism is, at the same time, a perceptological, gnoseological, and aesthetological problem; in our terms, we may reduce it to the linguistic issue of pertinence (as defined by Luis Prieto): the economic representation of the face proposed in signs such as smileys, emoticons and emojis, passes through a process of “amplification through simplification” (McLoud 1993, 30) which focuses on the relevant distinctive features selected among the many possible ones included in our “face mask.” When, eventually, do we have something we can properly call “a face”? And which face is it? It is clear that some discontinuity has to be highlighted in the triadic system composed by eyes, nose, and mouth, in terms of binary oppositions. The ones identified by linguist Michele Zappavigna (2012)—with regard to eyes, mouth, nose, forehead, accessories, repetition (of some elements), appearance (the direction to be assigned to the pictogram in order to correctly read it)—provide a fitting and yet involuntary application of the semiotics of the plastic arts (or, for brevity, plastic semiotics) proposed by Algirdas Greimas and his pupils (especially Jean-Marie Floch); the eidetic (relating to forms), chromatic (lights and colors), and topological (disposition of the constituents) categories are articulated through plastic traits and formants. In the Zappavigna model (2012, 71–82), for instance, what we would define as the “plastic formant of mouth” seems to be particularly rich; the mouth can be completely or partially open or it can be closed and, if so, it can be flat or curved, and in the latter case it can curve upward, downward, or in a mixed fashion. Any junction marks a potential difference in meaning.

We can definitely identify universal facial traits; still, the way in which the paradigm of possible prominences is organized and substantiated changes across time and space.³¹ As pointed out by Francesco Galofaro (2002), Japanese emoticons (kaomojis, literally “face characters”), a mid-1980s direct outcome of the so-called ASCII art fad, propose a different selection and organization of relevant traits, a different valorization (e.g., in terms of gender representation), as well as a different segmentation of the continuum of emotional states. Japanese smileys “laugh with their eyes,” not with their mouth, and, unlike Western emoticons (which have to be read sideways), are horizontal (like the characters in PLATO). Kaomojis allow the viewer to distinguish between a male (^_^) and a female face (^.^) thanks to the shape of the mouth. They are particularly refined in conveying emotions whose articulation is complex and crucial in Japanese culture (this is the case of embarrassment, for which there are different graphic signs, each capable of expressing a different “degree” of this feeling).³²

A typological study, that is, a comparison between different communities and cultural systems, appears to be an unavoidable path to get to a comprehensive semiotic understanding of these phenomena. In non-Western writing systems, such as Chinese, for instance, symbolic and iconic components may generate fertile attrition: ideograms, single units that are internally articulated and yet signify entire concepts, are used in spite of their literal meaning; the ideogram 囧 (jiong), which originally meant “decorated window” or “shine,” but whose shape resembles a face with bulging eyes and wide open mouth, thus becomes the emoticon expressing “annoyance” and the web character saying, “I don’t want [to do this!].”³³

The ideographic example is perfect for underlining a key component of digital pictograms; their intermediality to be understood in the original terms proposed by Fluxus artist and theorist Dick Higgins (1966). The intermedium is a medium in-between, which simulates another to the extent of creating a third one that is neither the first nor the second but both at the same time. This is the case—a verbal text outlining an image, an image drawn by means of a verbal text—of the visual poems of Ancient Greece, the calligrams of Apollinaire, the point d’amour (point of love) introduced by novelist Hervé Bazin (1966). Marcel Danesi defines emojis as “picture-words” and links them to the same utopian lineage of artificial languages as Leibnitz’s pictograms and Blissymbolics (2016, 157–70).³⁴

The universalistic élan of such emoticons and emojis is mitigated not only by the different pertinence and stylization ideologies at work in different communities (e.g., Western versus Eastern smileys) but also by issues related to ethnocentrism or ageism; different minority groups and communities have become vocal, demanding the correction of the established representation ideologies by means of a progressive integration of figures, in a postcolonial and inclusive perspective. Emojis’ standard smiling little yellow man (👨) (this being an unnatural and, supposedly, neutral color) was joined by versions with different skin colors, equipped with various accessories related to specific identity values; an example is the female emoji wearing the hijab (🧕) promoted by Rayouf Alhumedhi, a 15-year-old Saudi student residing in Germany, in 2016 (Eddy 2016). Emojis of this kind represent a visual homologue to phonetic allophones: the sounds and the signs change, but, across different communities, the function and value remain the same. In the same year, 56-year-old Diane Hill, responding to a BBC Open Doors call, advocated for the introduction of “emOLDji,” a self-ironic series of emojis designed by the elderly for the elderly.³⁵ Emojis of this kind may be conceived as a form of code mixing (such as slang intrusions in a broader context of koine) or industry jargon (which, in the latter case, we could call “gerontosemiotics”).

Faces and Memes

Considering emoticons a sort of primitive Internet memes, as first proposed by media historian Patrick Davison (2012), enables a twofold in-depth understanding: it allows us to properly set emoticons within the contemporary media ecology (i.e., a digital media ecology dominated by the dynamics of so-called virality) and to highlight the dialectic between structures and figures in memes.

By Internet memes we mean the media fragments that underpin our everyday online interactions: texts, images, and videos (almost always captioned) that in semiotic terms—according to the classic typology of transtextuality proposed by narratologist Gérard Genette—are defined as ludic transformative hypertexts or, for brevity, parodies. Memes are generally regarded as the viral content—spreading “like wildfire”—par excellence; they are the cultural catchphrases that we engage with in order to talk about ourselves and our semiosphere, with the excuse of conversing about the current events everybody else is talking about.³⁶

Davison considers emoticons the prototype of meme: they are simple, funny, available and usable by everyone and, as they were invented but not patented, they present what he calls “nonattribution … metameme” (2012, 132–33); namely, as already playfully pointed out in the Puck discussion of typographical art, they are anonymous and under copyleft, which makes it possible for this semiotic resource to spread and gain longevity.³⁷ In Nelson Goodman’s terminology, emoticons represent a perfect example of allographic sign, calling for replication and multiplication; this happens because on closer scrutiny emoticons identify a form of expression rather than a specific substance (whereas simple virality concerns one token and develops by means of copy, memeticity concerns the type that is being established after the token and develops by means of modification or imitation); let us think of how each operating system (Apple’s iOS versus Android), platform (Facebook versus Twitter) or browser (Mozilla versus Chrome) proposes its own customized version of the same face. Superficial differences are on the figurative level, while the essential plastic configurations remain the same.

Smileys are the grandparents of memes. But smileys and memes are interrelated also for other reasons. One of the very first Internet phenomena to be properly defined as a meme was 22-year-old Eric Wu’s 1998 website “Eric Conveys an Emotion,” which recreated, at users’ request, facial expressions from emoticons.³⁸ Smiley faces and, actually, faces of any kind—more or less synthetic, more or less cartoonish, more or less grotesque—have been filling the memetic mediascape from the very beginning. As a matter of fact, the rage faces and the sagas dedicated to NPCs (Non Playable Characters), Chads, and Wojaks (dozens of types, each suited to conveying a mood, personality, situation), all of them sketchily drawn on Microsoft Paint and spread via nerds den 4chan, are nothing but “crazy” smileys. Stylized faces used as ready-mades for commenting on digital content derive from a process of emoticonization (transformation into emoticon-like signs); let us think of the proverbially exaggerated expressions of Nicolas Cage (“You don’t say!”) or Jackie Chan (“What the fuck!”).

Faces and Formulas

There is a wide range of facial memes, centered on faces as they focus on the characters who “wear” them; these are comic, cartoonish, and funny memes as they portray characters whose facial expressiveness is comic, cartoonish, and funny, serving as what Roland Barthes would call the punctum of the picture. But there are also “barefaced” memes, with no focus on faces or characters; they are also “barefaced” in that they are potentially suitable for any context that would respect their basic rules of felicity. These memes coincide with their own structure, being self-sufficient, as they do not convoke any knowledge external to them. In order to laugh at a meme depicting Chuck Norris or Kim Jong-un, I need to recognize them as referents (a competence that Eco would define as “encyclopedic”), whereas, in order to laugh at a label meme such as the Distracted Boyfriend, where a guy turns to look at a girl while his girlfriend glares at him, I just need the pragmatic competence to recognize the allegorical value of the template manifested in the image, which is suitable for signifying any possible narrative where a given actant (the Subject) is attracted by a second one (the Object of value) to the detriment of a third one (the Opponent).

Digging under pervasive communicative phenomena such as memes would allow us to obliquely pursue, within a radically different scenario, the structuralist project to reconstruct the forms of culture, starting from their superficial manifestations. Beyond the etiological obsession with origins and birthrights (that we have followed with the archeosmileys), beyond the typological obsession that more than often accompanies the semiotician (hopefully leading to a crosscultural hermeneutics), it seems revealing that almost all the luxuriant hypertextual visual universe that provides the common currency for phatic exchange in contemporary social media interactions can be modeled starting from a double genealogical line predating the Internet (in particular, coinciding with the outbreak of the Second World War). On the one hand, we have the strength and effectiveness of the formula, whose precursor is the catchphrase “Keep calm and carry on,” the prototype of the half-empty structure (“Keep calm and ___”) to be indefinitely completed and resemantized. On the other hand, we have the strength and effectiveness of the icon, whose precursor is the funny and—not by coincidence—subtly ambiguous face of Kilroy peeping over a wall to claim his presence (“Kilroy was here”). Further studies in this direction will allow us to understand with ever greater detail the role of the face as writing and, vice versa, in online communication.