Although the identification of scripts’ distinctive visual properties, as a prerequisite to emulation, is constitutive of typographic mimicry, it has not yet been studied in detail from a structural, i.e. graphetic point of view (for the field of graphetics, cf. Meletis Citation2020a, Chapter 1). Instead, it appears to be a rather intuitive and subjective affair. For example, in his process of creating a mimicking typeface, French type designer Jean-Antoine Alessandrini (Citation1983)

[…] has to soak up the graphic universe of the country or culture he wants to evoke. To do this, he collects a wide range of documents, from sacred texts to daily newspapers. This allows him to build in his mind a subjective representation of this exotic country or culture. He then produces pages and pages of drafts to figure out which style of drawing will best evoke the exotic country to his Western imagination and to what he expects will be the collective unconscious vision of his fellow countrymen. (Celhay, Boysselle, and Cohen Citation2015, 169)

This described method is interesting for several reasons. Firstly, because the basis for the finished typeface is a "subjective representation" of an "exotic country or culture" (whose perceived exoticism is further highlighted by the term Alessandrini coined: exotype), and secondly, for the perspective from which this is judged, i.e. collective Western imagination. There is nothing subtle about not only the othering involved here but also the (cultural) appropriation. A point that will be picked up below is that the goal underlying typographic mimicry is seldom to represent authentically a different culture but rather to reproduce and reinforce one’s own image of it from an outside perspective. This is where the line between a visual homage and the perpetuation of visualised stereotypes becomes thin and the Western view of (non-Western) cultures dominates and supersedes their self-image, underlining a lack of agency on their behalf (cf. Section 5).

In striking contrast to this intuitive method, Coulmas (Citation2014, 18–19) outlines how an objective structural analysis might proceed:

First, a set of distinctive features is to be extracted from the sign inventory of the writing system in question. The next step is to develop a grapho-grammar on the basis of this set that generates real and possible graphemes, but excludes impossible ones. Further, a list can be compiled of conspicuous features that distinguish a writing system from others of the same family, such as, for instance, characteristic ligatures like […] German <ß> originating from <sz>, unique letters and diacritics as well as punctuation marks as used in digital language recognition software.

What makes an analysis such as this challenging is that no uniform method of describing the visual features of scripts, i.e. no "grapho-grammar," has been proposed yet (cf. Meletis Citation2020a, 225), which, given the sheer visual diversity of the world’s myriad scripts, is scarcely surprising. Developing such a grapho-grammar is, of course, also beyond the scope of this paper and remains an important desideratum for future research.Footnote⁸ In terms of form, a fairly superficial distinction that can be drawn at first glance is between, on the one hand, typefaces in which letters in their entirety resemble characters of a foreign script (cf. the typeface Sunrize in , in which Latin shapes mimic characters from Japanese hiragana) and typefaces in which salient features such as loops and curls are being integrated into letters that otherwise still very much resemble the original letters (in this case Latin) on the other (cf. Pad Thai in ). In terms of a continuum, the former mimics the target script to a higher degree than the latter. It is also paramount to note that some typefaces rely not primarily on the emulation of shape but instead on mimicking the way in which a script is produced (including the materials used); for Chinese, the stereotypical belief is that it is written with a brush (see below). Making sense of this kind of typographic mimicry thus requires additional (or different) graphic knowledge than knowing only what the target script looks like. Finally, typefaces such as Hot Tamale are interesting as they do not emulate a specific script (writers of Spanish in Mexico also use Latin script, which is why no example of a different script is given for it in ) but instead visually evoke a culture based on repeated uses in contexts associated with said culture (e.g. in the signage and menus of taco restaurants, see also below).

In the specific instance of typographic mimicry that is at the centre of this paper, Latin typefaces are designed to emulate Chinese script (cf. Kashima Brush in ). From a descriptive point of view, the primary strategy in the design of these typefaces is to mimic the curved brush strokes characteristic of Chinese calligraphy (cf. Yang Citation2012). Even though this may seem straightforward, it is largely unsuccessful, as Shaw (Citation2009) notes: "[…] the strokes, forced onto the armature of Roman letters, are assembled in a manner that completely ignores a calligraphic emphasis on structural balance and harmony."Footnote⁹ Criticism not of the use(s) of typographic mimicry but of the quality of type design will also be addressed in the analysis of metapragmatic discourses below.

At the end of this section, it must be noted that a static and isolated analysis of visual features alone is of only limited use since for typographic mimicry to function as social practice, the emulated features must of course be identified by users as sociosemiotically significant, i.e. as being culturally indexical or, simplistically stated, "constituting foreignness." Thus, since the product-oriented analysis of a culture’s semiotic "toolkit" at most addresses the semiotic potential of individual typographic resources, it must imperatively be complemented by an interactional investigation of users’ semiotic "‘struggles’ and ‘negotiations’" (Spitzmüller Citation2015, 137). In other words, typographic design does not simply carry meaning by itself, it is used to make meaning (cf. Spitzmüller Citation2016a, 112). Against this background, the focus of the investigation must be shifted onto the uses of typographic mimicry.

Typefaces used for typographic mimicry enjoy great popularity: on dafont.com, a website that offers freely downloadable typefaces, for example, the category "Foreign look" subsumes the six options "Chinese, Japanese,"Footnote¹⁰ "Arabic," "Mexican," "Roman, Greek," "Russian," and "Miscellaneous." Notably, the subcategory at the centre of this study, "Chinese, Japanese," counts (at the time of writing)Footnote¹¹ by far the most typefaces.

Latin script typefaces that mimic Chinese script or a subset of its visual features are known under several names, including Chop Suey fonts, Wonton fonts, Karate fonts, or Takeout fonts.Footnote¹² These labels already illustrate ostentatiously the stereotypical cultural associations typefaces of this kind (are supposed to) evoke. As design historian Paul Shaw (Citation2009) explains, these Chinese-imitating typefaces emerged in the United States between the mid and late nineteenth century. Central to establishing a strong association between them and Chinese culture (in particular what is perceived by Western culture as genuine "Chinese cuisine") was their frequent use in San Francisco’s Chinatown, which had been reconstructed following an earthquake in the early twentieth century. What is remarkable, as Shaw notes, is that even back then, this kind of typographic mimicry was primarily practiced by Chinese American restaurant owners themselves. This, he argues, provided the typefaces in question – much like the dish chop suey (which lends these typefaces one of their many names, cf. above) – with a degree of authenticity despite the fact that both actually originated in the US (cf. Kim and Kim Citation1993, 34). According to Shaw (Citation2009), one of the first mimicking typefaces was the obviously titled Chinese, initially released in 1883 by the Cleveland Type Foundry;Footnote¹³ from the 1950s on, it became known as Mandarin (cf. ). It is characterised by the above-mentioned (clumsy) emulation of brush strokes which were/are considered typical of calligraphic Chinese characters. As this example implies, early specimens of typographic mimicry were not necessarily more sophisticated than modern ones;Footnote¹⁴ however, a large-scale historical study of the origins and first uses of typographic mimicry would be necessary for more fine-grained assessments.

As for the use of typefaces mimicking Chinese script today, they are still found mainly in signage of Chinese (and to some extent other Asian) restaurants in the West, especially Europe and North America (cf. Spitzmüller Citation2007, 412).Footnote¹⁵ As will be shown below, the common belief among the participants in online discourses is that these typefaces are used mainly by business owners who wish to signal themselves and their culture to people who are not members of it, cf. Kim and Kim (Citation1993, 31): "Exterior signs […] may reflect the proprietor’s self-conception, as well as his or her conception of customer expectation and association regarding the type of business represented." In the eyes of people who are not members of Chinese culture, this also legitimises the use of typographic mimicry by everyone (see below). From (mostly restaurant) signage, the use of typographic mimicry – and this is not restricted to the type emulating Chinese script – has extended to the food industry in general (cf. Sebba Citation2015, 219), e.g. in the form of the packaging and marketing of food (cf. Celhay, Boysselle, and Cohen Citation2015), an example of which will be analysed in more detail below.

In the context of investigating the use of typographic mimicry as social practice, several groups of actors as well as their knowledge and ideologies are of relevance: in addition to lay producers and recipients, who represent the majority, they also subsume those who have a professional background in design, including the very type designers who create typographic mimicry typefaces in the first place. As Murphy (Citation2017, 69) elaborates, "typeface […] can become a critical source of meaning and a site for cultural intervention for various users of text, including readers, writers, the professionals who design text artifacts, and even people consuming text ambiently in the world around them." Of course, these categories of actors de facto often overlap: people who use typographic mimicry have usually perceived and noticed it in the past (meaning producers were/are also recipients) and, in special cases, for example when companies place marketing orders at design firms, designers of advertisements have created a typeface specifically for the company/the ad (meaning they are not "merely" producers who use pre-existing typefaces but the designers of these very typefaces in the first place). Regardless of the group users belong to, they require certain typographic knowledge,Footnote¹⁶ which is defined as "the sum of all assumptions about the use of typographic elements that are considered ‘given’ in a certain collective (at a certain historical point in time)" (Spitzmüller Citation2009, 471, my translation). Importantly, this knowledge is by no means evenly distributed. Rather, it is socially stratified as "specific graphic forms have different meanings for different interactants" (Spitzmüller Citation2016b, 107, my translation). Notably, the awareness of this individual variability of typographic knowledge does not appear to be widespread, which is why "members of a given social community take [them] for granted and assume [them] to be shared within this community" (Spitzmüller Citation2015, 132, emphasis in original). In other words, users often superficially believe others share their exact knowledge.

The majority of producers and recipients of design products incorporating typographic mimicry are lay users – understood here as those users who have no typographic training and have not studied the subject in depth. They, of course, still require specific typographic knowledge in order for typographic mimicry to have a sociosemiotic effect. This knowledge comes in various forms: (1) on the one hand, they may be familiar with the actual appearance of and scribal practices associated with the target script (e.g. Chinese) and, on this basis, recognise the emulated visual features (e.g. the mimicked brush strokes) and thus establish the intended association with the indexed culture. It is vital to note that for this to work, it is not necessary for or even expected of recipients to be able to read the target script (cf. Guerini Citation2019, 84–85), which aptly underlines that the target audience of typographic mimicry is not members of the indexed culture but precisely people who cannot read the script in question.Footnote¹⁷ (2) On the other hand, recipients can make meaning of typographic mimicry purely on the basis of their previous exposition to similar instances. Notably, in these past receptions, the cultural indexicality was not constituted by a recognised visual similarity to the actual target script but rather by the mimicry’s repeated embedding in specific cultural contexts (e.g. being used on the signage of Chinese restaurants).Footnote¹⁸ These two different types of typographic knowledge reflect a central discrepancy: between what is genuinely a part of a given culture and recognised as such (= the target script) and what is assumed to be associated with said culture (= in this case typographic mimicry that has been repeatedly encountered in culture-specific contexts). Regarding the question of how cultural or "ethnic" connotations, a subtype of typographic indexicality, can emerge in the first place even when a visual link to the target script is not necessarily established, Shaw (Citation2009) explains:

These fonts’ ethnic connotations have developed gradually, through recurrent appearances on book covers and posters, by people who connected the typefaces with their own cultural biases and perceptions, slowly reinforcing the fonts’ ethnic associations in viewers’ minds.

As mentioned above, in the case of Chinese typographic mimicry, it is not book covers or posters but mainly restaurant signs that serve as the basis of associations. In others words, as it is repeatedly encountered in signs of Chinese restaurants, people begin to associate Chinese typographic mimicry with this particular context, habitualising it as a social practice. Crucially, it is not only associated with a specific context but also "with certain groups of people who […] are thought to be typical users" of it (Spitzmüller Citation2016b, 101, my translation). In this case, the predominant assumption among recipients is that it is mainly members of "the" Chinese cultureFootnote¹⁹ who are users of Chinese typographic mimicry – thus, as a practice, it is socially enregistered (cf. Agha Citation2007). The question of who engages in typographic mimicry is also related to an interesting strand of discourse that will be examined below: whether it is appropriate for typographic mimicry to be practiced by people who are not part of the (cultural, ethnic, social, …) group to whom the emulating typefaces are meant to indexically refer. With this question, the focus shifts from the study of typographic knowledge to the analysis of typographic ideologies, attitudes and "assumptions about ‘value,’ ‘sense,’ ‘meaning,’ ‘function,’ and typical circumstances in which type and forms of textual design are used" (Spitzmüller Citationforthcoming, my translation). What is of particular value for metapragmatic discourse analyses is producers and recipients explicitly expressing their graphic ideologies. Nowadays, they can easily do that online, and as a result, the internet stocks countless reflections of discourses that researchers can discover and analyse systematically. This leads us to the methodology adopted in the present study.

The main methodological framework underlying the present analysis is Herring’s (Citation2004, 367) general approach of computer-mediated discourse analysis (CMDA). Accordingly, the basic steps are (1) articulating a research question, (2) selecting data samples, (3) operationalising key concept(s), and (4) selecting and applying the fitting methods, after which (5) the results are summarised/synthesised, and (6) the research question is answered. Here, these steps are undertaken on the basis of three methodological strands relevant to discourse analysis, the first of which is the qualitative heuristic approach (cf. Kleining Citation2004; Fraas and Pentzold Citation2008) that strives to optimise the chance of discovering relevant results by adhering to four basic rules (Kleining and Witt Citation2000, Section 2, emphasis in original):

Rule 1: The research person should be open to new concepts and change his/her preconceptions if the data are not in agreement with them. […]

Rule 2: The topic of research is preliminary and may change during the research process. […]

Rule 3: Data should be collected under the paradigm of maximum structural variation of perspectives. […]

Rule 4: The analysis is directed toward discovery of similarities.

Against the background of these rules, it is important to disclose that my general research interest in the materiality of writing motivated the study; however, going into it, I did not have any knowledge about how the specific and little-studied phenomenon of typographic mimicry was perceived and represented discursively and thus had no preconceptions about what the analysis would bring to light. This also ensured that the topic of research had the freedom to develop and change during the process and that diverse data and differing opinions would be considered even though criticism of the use of typographic mimicry (in the form of an article, cf. Yang Citation2012 and Section 5 below) functioned as the starting point of the study. It was indeed this initial discovery of how the topic appeared to be the subject of lively debates on the internet which led to the formulation of a (purposefully) relatively broad research question: How is typographic mimicry negotiated in metapragmatic online discourses? Limiting the scope to the online realm is in accordance with the view that "the affordances of online communication are obvious requirements for fontroversies to thrive" and that "[i]ncreased public attention to fonts in recent years beyond the world of graphic design is unquestionably related to the rise of social media and other communication technologies" (Murphy Citation2017, 80).

There were no rigid criteria for data selection. Instead, the comment sections under two articles (Coville Citation2013; Helfand Citation2007) were chosen for the vivid discussions they displayed; additionally, several articles dissecting the topic from various perspectives were taken into account. As will be shown below, the fact that the respective websites on which these articles were published cater to different target audiences (lay users vs. experts) made possible a rough comparison that shows how varying degrees of typographic knowledge may influence the main arguments in discourses. One more aspect regarding the data must be mentioned: While it cannot be guaranteed that the analysed online comments will remain accessible (cf. Fraas and Pentzold Citation2008, 297; Meier Citation2014), they are cited here with the date of their creation as well as the names of the users who posted them; the most recent date at which all of them were accessed is 11 May 2021. Furthermore, in some cases, it can be assumed that users published comments under their actual names; citing these in order to ensure full traceability can be regarded ethical as the comments were (most likely) knowingly and voluntarily posted in public comment sections that can be viewed without registration or joining a group (cf. Tanskanen Citation2007, 89).

The other two methodological approaches that shape the operationalisation of core concepts as well as the actual analysis are the societal treatment approach (cf. Garrett Citation2005) and the DIMEAN model (from German Diskurslinguistische Mehrebenen-Analyse "discourse-linguistic multilayered analysis", cf. Spitzmüller and Warnke Citation2011 for an overview). The societal treatment approach, on the one hand, is a discourse-analytical method that "follows the attempt by communicative ideology research […] to concentrate on articulated values and beliefs (as opposed to the attempts […] to reveal covert values and beliefs […])" (Spitzmüller Citation2012, 258, emphasis in original), i.e. emphasises what is actually communicated in online discourses. DIMEAN, on the other hand, is a model that understands itself as a collection of "linguistic methods that allow us to approach different layers" of discourse (Spitzmüller and Warnke Citation2011, 81). It consists of the intratextual layer (texts), the agent layer (actors), and the transtextual layer (knowledge); each of them can be subdivided in further sublayers and stages of analysis. The present study focuses on the latter two, which allow highlighting as core phenomena the social stratification of (typo)graphic knowledge, (cultural) indexicality, (self-)positioning, othering, and group membership, as well as the perceived (in)appropriateness of typographic design choices. These have been introduced in the previous sections and will recur throughout the analysis of typographic ideologies that we now turn to.

To begin with the specific example that motivated the present study as it prompted a strong reaction and thereby laid bare conflicting typographic ideologies, consider . It shows a screenshot of a newsletter promoting new instant meals from US company FreshDirect (from 2012). What is of interest is the typographic mimicry in the visual design of the product titles "stir fry kits" and "dumplings."

As mentioned, this example is significant because it ignited a discussion. Journalist Jeff Yang even devoted an article to it that he polemically titled "Is Your Font Racist?" (Yang Citation2012), which also lends the present paper (part of) its title. In it, he details how a friend of his complained to the company about the "stereotypical typography" in the design of the newsletter as she was disappointed by the "laziness of using an outdated and unimaginative font." The company’s response is illuminating under a metapragmatic lens as it shows how previous instances of a practice are instrumentalised to legitimise its continued use:

The FreshDirect creative team did not in any way intend to offend anyone with the use of the typography […], which was meant to echo a classic typeface found in many menus and takeout boxes […]. (quoted from Yang Citation2012)

Thus, according to the company, what the use of the typefaces meant to "echo" was not Chinese culture directly but the established practice of using this kind of typographic mimicry to advertise (Chinese) food. This corresponds with the second type of typographic knowledge and basis for making meaning of typographic mimicry mentioned above. Crucially, in such cases, the producer’s intention is arguably still to index (albeit indirectly) a given culture. Thus, shifting the blame to someone else for having used this practice before – and directly, as in referring to the culture rather than relying on previous instances of typographic mimicry – does not absolve a producer from the responsibility of engaging in it.

What must also be noted is that in FreshDirect’s response, the company claimed that its design team had indeed reflected and discussed at length the use of these typefaces and had even consulted an article on the topic published in the Design Observer (Helfand Citation2007; see the discussion below). In its reply, the company quotes the following passage from said article:

And yes, it’s all about appropriateness: fine to use Fake Hebrew for a deli; not so fine on, say, a yellow bracelet. Likewise, nobody questions a sign for a burrito restaurant designed in Hot Tamale,Footnote²⁰ but what about when it’s used for a border crossing sign in Texas? (Helfand Citation2007)

"We agreed with this evaluation of ‘appropriate use’" (quoted from Yang Citation2012), the company states in reference to this presented view. Interestingly, it omits more critical passages of the cited article (see below). This example reveals several issues relevant to a metapragmatic analysis of typographic mimicry: (1) criticism of the typographic quality of the typefaces and the design products in which they appear (and by extension also criticism of the type and graphic designersFootnote²¹), (2) the negotiation of whether the use of such typefaces is appropriate either for all actors, only those actors who belong to the culture (in the broadest sense) that the typefaces are associated with, or no one, as well as (3) the question of whether these typefaces are offensive to the supposedly represented cultures and the corresponding discussion of (whether there even exist) contexts in which using these typefaces may be (in)appropriate.Footnote²² Often, these questions cannot be separated from each other, and in her blog entry, aptly titled "StereoTYPES," Fernández (Citation2015) argues that they are indeed intricately related. Regarding the quality of typefaces, she writes: "Ethnic fonts are generally less structured, often handwritten, have purposeful flaws, look old, and can often be based on traditional writing." For her, the lack of a vocal criticism for using such typefaces is incomprehensible because it is not primarily their emulation of visual features but their inferior typographic quality that represents a form of "othering" (us/Western vs. them/non-Western) and, more importantly, a devaluation:

It matters because it adds to the "currency of signs" that valorizes western aesthetics. It places modern, progressive, industrial, and democratic values to standard typefaces like Helvetica […], while ethnic typefaces allude to the exotic, backwards, wild, and maybe even slightly savage.

Social positioning through typographic design is always positioning in relation to other (perceived) social positions (cf. Spitzmüller Citation2015, 128), and in the case of typographic mimicry, it often appears to be doubly coded: not only as "native" vs. "foreign" through the typefaces’ visual features but also as "good" vs. "bad" due to the (lack of) quality of their design.Footnote²³ This leads to the perception of typographic mimicry not as a mere emulation of (stereotypical) cultural features but as a bad, even caricaturing imitation. This is precisely what Yang’s (Citation2012) eponymous question of whether typefaces can be racist refers to (cf. also Quito Citation2021). Some people who actively engage in the metapragmatic online discourses affirm this sentiment, which is also why typefaces emulating Chinese script are sometimes referred to as yellowfaceFootnote²⁴ fonts (cf. Fang Citation2012) – (especially) when used by the non-Chinese. In a related vein, typographic mimicry has sometimes also been described as the visual equivalent of accented speech (cf. Kim and Kim Citation1993, 33) that also often carries negative connotations. Salen (Citation2001, 153) suggests that it could even be interpreted as a visual euphemism in that it is "used to avoid another word thought to be too direct, blunt, harsh or offensive." Put differently, the connotative level of meaning constituted by typographic mimicry may be intended to convey what producers do not "dare" to say through actual denotative linguistic content.

Not everyone shares these critical and culturally sensitive typographic ideologies. This becomes particularly evident in the exemplary and small-scale analysis of comments posted under two pertinent articles, the first of which is a playful blog entry and the second of which is the above-mentioned article in the Design Observer. As will become evident, reactions to these articles offer a rich source of data for a first metapragmatic analysis of discourses on typographic mimicry. Furthermore, given that the outlets in which the selected articles were published cater to readerships with differing knowledge backgrounds, readers’ reactions to them will be used for a rough comparison of discourses of lay users on the one hand and discourses of experts on the other. In the lay article, the jokingly titled blog entry "5 Genuinely Offensive Font Choices that Must be Stopped" (Coville Citation2013), published on the humorous website Cracked.com, the author argues that typographic mimicry is offensive, writing (hyperbolically):

It is hard to comprehend the brain pattern of the people who choose this font, but it must go something like: "How on earth is my audience meant to know that my sign that reads ‘Chinese Restaurant’ refers to a Chinese restaurant if I don’t write it in wacky calligraphy-y, bamboo-y letters?"

Many readers who reacted do not share the author’s opinion. In some cases, they are even hostile towards her, arguing that she is getting upset about a topic that does not represent a problem from the majority’s point of view. Notably, some do criticise the typographic quality of the typefaces (see above) but do not conclude from this that their use is racist. One user writes: "The use of typefaces which resemble the Chinese script in a Latin alphabet may be unintelligent, cheap, unprofessional, uncomfortable and all that but it’s not racist,"Footnote²⁵ while another notes: "It’s cheap and unimaginative. Nothing more."Footnote²⁶ Rather than denigrating typographic mimicry in general, different uses are indeed subtly distinguished: "[a] strategic and sometimes even ironic use of the font […] and the derogatory way in which it is more often utilized […] are radically different" (Lily Citation2012). Thus, the intentions ascribed to producers are central. A point that was already mentioned above and that is frequently brought up in the comments to refute that typographic mimicry is racist or generally ill-intentioned is that its primary users supposedly belong to the referenced culture(s). This (re)appropriation has long been acknowledged in the literature on typographic mimicry. As Kim and Kim (Citation1993, 34) note:

While initially springing from the Western designer’s view of the "other," these typographic forms have been appropriated by the increasing number of immigrant-owned businesses and used as self-representation in the commercial realm to render the various groups easily recognizable or to appeal to the American appetite for the exotic.

This has evidently also been picked up by lay users discussing the matter, as one user writes: "Pretty much every Chinese restaurant I’ve ever eaten at has been run by – are you sitting down? – Chinese people. So chances are they chose the font style for their signs. If they don’t think it’s racist, why should anyone else?"Footnote²⁷ Another user does consider typographic mimicry to be problematic but elaborates: "[…] the weird thing is, it’s the Asian restaurant owners who insist on the racist fonts. (I work in a marketing company and deal with a lot of them)."Footnote²⁸ In this vein, yet another user pleads: "So don’t blame the designer, blame the Chinese restaurant owner who wants his menu to look like all the rest out there and can’t handle something different."

Let us now turn to the arguments circulating in expert discourses. To trace them, the comments under the above-mentioned article used by FreshDirect as justification for its use of typographic mimicry will be considered. The article in question, titled "Why is This Font Different From all Other Fonts?" and written by Jessica Helfand, was published in 2007 in the Design Observer, a website devoted to design topics, run by designers, and arguably also aimed (mostly) at designers. In the article, Helfand comments on typefaces that emulate Hebrew and addresses how they – as well as typographic mimicry in general – may be problematic.Footnote³⁰ She closes the article by asking: "[…] what’s the difference between a celebrity making an unforgivable racist remark and a typographer making a font that clumsily perpetuates a cultural stereotype?" It is this question that many of the commenters reacted to.

What is striking when looking at the 47 comments made (to date) is that in many of them, commenters explicitly self-identify as designers, i.e. as experts on the topic. This is evident in formulations like "we as designers"Footnote³¹ or "[…] real designers won’t use it. We wouldn’t be caught dead using any of the junk fonts like it"Footnote³² (my emphasis), likely used to underline the legitimacy of their opinion. The fact that the commenters consider themselves experts leads to several interesting threads that run through many of the reactions; they include a justifying and at times even attacked tone as well as an attempt at distancing oneself from "other," i.e. lay users. Yet, much like in the "lay discourses," the overwhelming majority of commenters believes typographic mimicry is not an inherently "bad" practice:

The opinions expressed in these comments resemble those found in the lay discourse, most importantly that typographic mimicry should not be generally condemned. Because of the commenters’ (self-proclaimed) expert background, their focus is not on the use of mimicking typefaces by end users but rather on issues of type design. Accordingly, it is not only the typefaces themselves that are defended but also the process of designing them in the first place and, by extension, their designers as relevant actors. Much like there is supposedly no malice involved in certain uses of these typefaces, the process of designing them is also commonly not believed to be ill-intentioned. Interestingly, while the typefaces’ quality is mentioned negatively in some comments, it is not addressed as often in the (excerpt of) expert discourse as in the sample of lay discourse.

A noteworthy question that is raised by one of the commenters is whether "the mere idea of faux lettering itself [is] a typographic crime" or if "[there are] any fine examples of this."Footnote³⁷ This person goes on to describe that in their design study programme, they were assigned the task of designing a typeface inspired by a different culture, and concludes that it truly represents a challenge not to resort to clichés in the process.

To introduce the two relevant negative voices among the expert reactions, it is fitting to first quote another comment that is notable for its neutral tone and for systematically explicating what the commenter assumedly believes to be the typographic knowledge and awareness type designers require:

How do the design decisions that designers make (our intentions) as they become manifest in specific environments (restaurant menu vs. job application) affect the receptions by diverse groups of people (positive, neutral, negative) to ideas of self and other, within a context where everyone is not given the same access to the agency to define themselves.Footnote³⁸ (emphasis in original)

The most relevant concept in this explanatory comment is (the lack of) agency. It is – albeit indirectly – also present in the negative comments, the first of which reads: "Asian Americans don’t need a faux-Asian font to tell them that there’s Chinese takeout in those boxes. That’s been done for the white reader."Footnote³⁹ The second one is more explicit in its disapproval of typographic mimicry. That its author identifies as a member of a group affected by typographic mimicry and addresses the others, i.e. those who are "in charge" of it, reflects the perceived lack of agency: "Non-Asian designers, believe me: you’re not doing our community a favour by using this typeface or other stereotypical imagery like gongs, coolie hats, dragons, cherry blossoms, geishas, kung fu fighters, people bowing, chopstick, bamboo trees […] which still are painful, not funny, and not yet ironic for us."Footnote⁴⁰ This leads back to an aspect that is absent from the expert discourse but very much central in lay discourse: the perceived legitimacy afforded to typographic mimicry because it is used by those people to whom it indexically refers. However, in opposition to the persuasion that "everyone can use it because the referenced groups use it themselves," there is also a view that scrutinises this reappropriation of typographic mimicry as well as the self-referencing it nowadays frequently serves:

[…] if members of the Asian community are in part contributing to the perpetuation of certain stereotypical Asian cultural signifiers, which are then used and abused by non-Asian enterprises to exoticize the Far East, how do we break the cycle? […] Do we put an intra-communal moratorium on chopstick fonts […]? (Lily Citation2012)

Obviously, "lack of agency" refers mainly to the process of the initial design of the typefaces in which members of the referenced culture themselves have little or no say. Against this background, their subsequent use of typographic mimicry may be interpreted as acceptance (perhaps out of resignation), with the reappropriation serving the purpose of fulfilling Western expectations of one’s culture that have been materialised in the form of mimicking typefaces (and mostly for commercial purposes).

Given the preceding analysis of the use of typographic mimicry as well as the explicit attitudes shared by lay users and experts, one can reconstruct the communicative (and commercial) goals mainly pursued with typographic mimicry; they are summarised by Shaw (Citation2009):

Ethnic type – not just chop suey but all of the varieties – survives for the simple reason that stereotypes, though crude, serve a commercial purpose. They are shortcuts, visual mnemonic devices. There’s no room for cultural nuance of academic accuracy in a store’s fascia.

The effectiveness of typographic mimicry is encapsulated in another comment posted under Coville (Citation2013), in which a concrete example is provided: "When I’m driving down a street with dozens of signs per block, I don’t have time to read them. When I see that font, I KNOW ‘Chinese restaurant,’ and that might be all I need to make my decision."

This study’s main goal was to show how typographic mimicry is used and evaluated as a social practice. Today, producers of texts and multimodal visual products must choose from the seemingly endless possibilities afforded by elements of type and graphic design. If they pick a typeface that either directly emulates, for example, Chinese script or resembles other existing typefaces that emulate it, they (usually) aim to evoke an association with Chinese culture or other products that are associated with it, assuming that this will be recognised by recipients. As the exemplary metapragmatic discourse analysis has shown, the typographic knowledge required to identify this association is quite widespread. Related typographic ideologies are not seldom strongly pronounced, and not only as a product but also as a practice is typographic mimicry criticised by some users while it is justified by others. Interestingly, the criticism of typographic mimicry as a practice was even found to be less vocal than the condemnation of said criticism. Who is allowed to practice it and in which – if any – contexts it is considered appropriate are matters of dynamic and ongoing negotiations that are typographic reflections of more global power relations.

This paper is the first of its kind to investigate aspects of the use of typographic mimicry by consulting data from online discourses. It thereby adds a case study not only to graphic design research but most importantly to the sociosemiotic investigation of the effects and functions served by diverse typographic phenomena (cf. especially Jürgen Spitzmüller’s many contributions cited throughout this paper). Next possible steps would be comparing the discursive negotiation of typographic mimicry with how these other phenomena are perceived and talked about as well as integrating common threads into a unified semiotic framework for analysing typography. Furthermore, a broadly defined and rather obvious practical takeaway for designers and users of typography is to try to be aware of the myriad intricate associations and thus potential consequences (both positive and negative) that are bound to their choices.

With the words "Next time ask yourself, is the font I’m choosing based on my western white privilege? If yes, use Helvetica, it’s neutral" (Maani Citation2019), a designer cautions against the (ab)use of typographic mimicry. In doing so, however, he potentially sparks a new debate, as the "neutrality" ascribed to Helvetica – or any typeface, for that matter – can also be disputed.Footnote⁴² Indeed, in the depths of the internet, one can find (ironic) comments such as: "Ummm Helvetica is a racist cultural stereotype of Swiss people. I’m offended."Footnote⁴³ What the discourses sketched in this paper clearly underline is that "‘neutral’ typefaces cannot exist" (Willberg and Forssman Citation2010, 72, my translation) because they are "subject to and complicit in a range of cultural projects along various affective, ideological, and even political dimensions" (Murphy Citation2017, 65). In other words, the use of typography is always social practice and thus an integral part of the complex processes of making meaning.

No potential conflict of interest was reported by the author(s).

1 This paper is partially based on my German publication "Typographische Mimikry," to be published in: Typographie: Disziplinäre Zugänge – Fachliche Konzeptionierungen – Forschungsfragen und Projekte, edited by Ursula Rautenberg, and Anja Voeste. Stuttgart: Hiersemann.

2 Note that in typographic terminology, "script (typeface)" designates typefaces that emulate (mostly cursive) handwriting. Here, it is used in its grapholinguistic reading, in which script refers to a coherent set of characters/basic shapes used to write given languages (see also below). Furthermore, although the designation "Roman script" would be more accurate, I use Latin script in the context of this paper as "roman" is also used in typographic terminology, where it denotes the "normal" style of a typeface in which the characters’ vertical lines are not slanted (as is the case in italic type).

3 In other words, what is usually investigated in this field is the materiality and multimodality of discourses and not how these phenomena are negotiated in discourses.

4 Notably, "typographic mimicry" could be considered too broad a term to denote the phenomenon as "typography" includes more than just typeface design. While the present paper indeed focuses exclusively on typeface mimicry, other features such as writing direction that can be and are sometimes mimicked (e.g. printing a text in top-to-bottom lines that run from right to left to suggest "Asian" writing) warrant the use of "typographic mimicry."

5 Note the crucial difference between the concepts of script and writing system (cf. Coulmas Citation1996; Meletis Citation2020a). A script is defined here as a set of graphic basic shapes that is used for the writing system(s) of one or multiple language(s). The Latin and Cyrillic scripts, for example, are used for many writing systems, respectively (e.g. English, German, Swedish, Italian, etc. use Latin script while Russian, Belorussian, Ukrainian, etc. are written in Cyrillic script). A writing system, then, is the combination of a script and a specific language (cf. Weingarten Citation2011). The English writing system pairs Latin script with the English language, making it different from the German writing system, which pairs the same script (with some additions and modifications) with German. Given this distinction, typographic mimicry, as a primarily visuo-semiotic rather than linguistic phenomenon, concerns scripts and not language-specific writing systems. Sutherland’s (Citation2015) use of the term "writing system mimicry" is not in line with this.

6 Pseudoscript, which Coulmas (Citation2014, 17) lists as an alternative designation, is also used in a related but different sense, cf. Nagel (Citation2011, 229), who studies ornamentation in the early history of Italian art, some of which "consists of script, usually eastern in flavour, sometimes close to Greek or Hebrew, often close to Arabic, but in fact in no known language: for the sake of convenience I call them ‘pseudoscripts’ […]. These apparently arbitrary strokes, slashes, and squiggles correspond to individual letterforms; their sequences sometimes repeat the same form at considered intervals, giving the array the appearance of having a linguistic structure." Thus, in this alternative reading, pseudoscripts are fictitious illegible scripts that merely visually resemble writing.

7 Note that this illustration is Eurocentric. Although it is sometimes implied (cf., for example, the definition in Alessandrini Citation1979), Latin script is not always the source script in typographic mimicry. Indeed, many different combinations can be found (i.e. Chinese script emulating Arabic script, Japanese kana mimicking Thai script, etc.). Some examples are collected at https://www.flickr.com/groups/cross-script-letterforms/ (accessed 10 May 2021).

8 Not everyone agrees that striving for such a grapho-grammar is feasible or even possible. Spitzmüller (Citation2012, 258), for example, argues that systematising typographic features in this way is impossible as "their interpretation is itself dependent on the context that is set up by means of all communicative modes. Therefore, it does not make sense to set up a context-abstract ‘grammar’ of visual elements or to look for distinctive semantic characteristics of specific graphic features. Due to the dynamic nature of graphic elements, such attempts are bound to fail."

9 Cf., more generally, Kim and Kim (Citation1993, 32, emphasis in original): "Despite the visual integrity and ingenuity of many of these alphabets, Western letterers’ mimicry of calligraphic strokes used by other writing traditions inevitably fails to refer correctly to the ductus, or order and direction of strokes, of the different traditions."

10 Arguably, lumping these two into one category is already a questionable action, underlining that little to no distinction is made between them from the Western perspective that serves as the dominant lens through which these representations of other cultures are (simplistically) viewed (cf. also Shaw Citation2009). This category, incidentally, also includes typefaces that mimic Hangul, the script used in the Korean writing system, further underlining that scripts used for various Asian writing systems are commonly not categorised in a more fine-grained manner. Interestingly, this broad categorisation is also reflected in the reception study by Celhay, Boysselle, and Cohen (Citation2015, 171), who even note uncritically that it is "understandable that accurately determining whether an exotype is making a reference to the Chinese or the Japanese culture is difficult for a sample of Western respondents."

11 January–February 2021.

12 While these designations are all based on the term font (which is also used in many of the comments cited in this paper as well as in the provocative quote that makes up part of its title), here, the term typeface is preferred. For the difference between them, cf. Murphy (Citation2017, 68): "The term ‘font’ is more widely used than typeface in colloquial discourse, largely due to its endemic presence in consumer word-processing software, but there is a historical difference between the two terms: in traditional typesetting the word ‘font’ refers to a complete set of letters and other characters in one typeface, one style (bold, italic, etc.), one weight (the thickness of its lines), and one size." Thus, for example, while Times New Roman is a typeface, 12 pt bold Times New Roman is one specific font of that typeface.

14 And neither were some of their uses: for example, Patil and Owens (Citation2019, 13–16) trace that typefaces mimicking Japanese (which, due to the lack of fine-grained differentiation, resemble typefaces mimicking Chinese, cf. Section 2) were used during WWII in US propaganda material with the goal of spreading racial prejudice. Another example they present is the use of typefaces mimicking Hebrew that could be found in Nazi propaganda. Such historical uses justify (and definitely explain) the discussion of whether contemporary uses of typographic mimicry have an underlying racist motivation.

15 Its predominant use in signage makes typographic mimicry a relevant subject of linguistic landscaping research. Accordingly, it features prominently in case studies such as Kim and Kim’s (Citation1993) analysis of the typographic landscape of Los Angeles but also at a more theoretical level in general discussions of the functions of typography in space and time, i.e. as part of semiotic landscapes, cf. the Social Semiotics special issue "Typographic Landscaping: Creativity, Ideology, Movement" (Järlehed and Jaworski Citation2015).

16 Note that typographic knowledge is a special form of graphic knowledge, which also subsumes knowledge about other types of writing (e.g. handwriting). The same applies to typographic ideologies.

17 In fact, typographic mimicry, due to the evoked visual similarity, often makes reading even more difficult for readers who are literate in the target script. The biggest challenge is posed by actual characters from the target script which are used for visually similar characters in the source script although they have a different graphematic value. A typical example is <Λ> used instead of <A> in Latin typefaces emulating Greek script. Here, readers literate in Greek recognise the "real," i.e. Greek graphematic value of the grapheme <Λ> (a correspondence with the phoneme /l/), which is why they potentially have problems reading it as <A> (with the correspondence /a/). This borrowing of actual characters instead of a mere visual emulation of features of their appearance is referred to as graphematic crossing and often occurs in addition to typographic mimicry (for different types of graphic crossing, cf. Spitzmüller Citation2007).

18 This appears to be explicit typographic knowledge at least among experts, i.e. type designers, as implied by a user’s comment in response to Helfand’s (Citation2007) article (cf. below): "[…] can you trace the cultural associations of Hot Tamale [a typeface evoking Mexican culture, cf. , DM] back to a particular source? […] maybe it looks ‘Mexican’ simply because it has been used so many times for burrito menus. Maybe the Mexican look is merely a reflection of its predominant usage" (Rob Henning, 26 June 2007). For an analysis of how typefaces can develop such (initially unintended) connotations through the complex interaction of a variety of factors, see Giampietro (Citation2004), a study tracing how the typefaces Neuland and Lithos have "come to signify Africans and African-Americans, regardless of how a designer uses them, and regardless of the purpose for which their creators originally intended them."

19 Note that in this paper, the notion of "culture" is used rather broadly. Incidentally, this use reflects what producers and recipients of typographic mimicry often believe and what is thus a pillar of this practice: that there is "a" Chinese (Arabic, Japanese, …) culture they can evoke by using certain typefaces. This, of course, sweeps actual cultural heterogeneity under the carpet but at the same time aptly highlights the boundaries of (typographic) knowledge that typographic mimicry may not cross in order to stay sociosemiotically meaningful.

20 For a sample of Hot Tamale, cf. .

21 The institutions that train these designers are also criticised, cf. Fernández (Citation2015): "Maybe we should also hold accountable the institutions that teach design without teaching the social implications of design."

22 Similar questions – albeit sociopolitically not as pressing – are found in metapragmatic online discourses surrounding the (use of) typeface Comic Sans (cf. Meletis Citation2020b). In them, what users perceive as a clumsy and childish appearance of the typeface serves as one of the reasons that many people argue its use in more formal and serious contexts (such as CVs) is inappropriate. The fact that it is still so often used in such contexts leads to generally unfavourable attitudes towards the typeface.

23 This paper focuses on mimicking typefaces that can be downloaded for free on websites such as dafont.com (cf. ). This may invite the impression that the discussion of the poor quality of mimicking typefaces may be associated with the fact that freely downloadable typefaces are generally considered poorly designed/cheap. However, the ideologies traced here are by no means restricted to freely downloadable typefaces but extend also to commercial typefaces. The analysis of discourses suggests indeed that the evaluation of these typefaces as being of poor quality is related to the purpose they are meant (and believed) to fulfil rather than to the question of whether they are available for free.

24 The term yellowface is used primarily to denote white actors playing Asian characters (cf. J. Lee Citation2019). A linguistic parallel to yellowface typefaces can be found in so-called "Mock Asian," the imitation of stereotypical Asian speech (especially in English, cf. Chun Citation2008).

25 Written by user Msilvertant (12 July 2014). This as well as the following comments can be found under Coville’s (Citation2013) article.

26 Wereboar (2 October 2013).

27 Steve Dutch (25 October 2014).

28 Darko29 (26 September 2013). Interestingly, in the lay discourse, such as in this example, several users "out" themselves as (or at least pretend to be) designers in their comments, presumably to underline that they have expert typographic knowledge and thus emphasise the legitimacy of their opinions. The arguments, in any case, are similar to the ones prevalent in expert discourses (see below).

29 Graphicstyle7 (26 September 2013).

30 One aspect she mentions is the typefaces’ names, providing as an example the name Circumcision for a typeface mimicking Hebrew (cf. https://www.fonts.com/de/font/t-26/circumcision/regular, accessed 17 February 2021). This is picked up by a commenter who notes that "the names should be the target of scrutiny, not the letterforms" (Josh, 27 June 2007). This as well as the following comments can be found under Helfand’s (Citation2007) article.

31 Written by user Whaleroot (27 June 2007).

32 Jessica Gladstone (26 June 2007).

33 Rob Henning (26 June 2007).

34 james puckett (26 June 2007).

36 Per Baasch Jørgensen (14 April 2010). Notably, the author of this comment is the designer of two (commercial) typefaces named Bagel and Falafel that mimic Hebrew script. His (defensive) contribution to the discussion shows that type designers themselves also take part in the discourses on typographic mimicry, especially in contexts that cater to experts (such as the Design Observer).

37 visakh (27 June 2007).

38 Dori Tunstall (30 June 2007).

39 Rebecca (20 June 2010).

41 Sinaku (26 September 2013).

42 Notably, typefaces which are commonly perceived as "neutral" are seldom the focus of research, which centres instead on special and "non-neutral" forms of design, a situation that Spitzmüller (Citation2016a, 115) refers to as expressive typography bias. This bias is echoed in what is being predominantly negotiated in discourses, underlining that "norms go unnoticed while marked practice is interpreted" (W. Lee and Su Citation2019, sec. 2, para. 1; Kroskrity Citation2004).