07

Words Left Unspoken: The External Forces Shaping Online Discourse

Introduction

1The digitalization and increasing networkability of information has led to dramatic shifts in the way reality is perceived and communicated. Journalism, for example, has undergone rapid transformations in business models and practices. Governments around the world utilize the surveillance capabilities of the emerging information and communication technologies (ICTs). Meanwhile, upheavals in global labor markets have given rise to a new working class, tasked with moderating and manipulating web content (Chen). All of these are examples in which specific forces come to co-determine the contents of the digital environments we encounter on a daily basis. Whether it is news content catered to the needs of market-based algorithms, government agencies seeking to keep oppositional voices or potential terrorists in check, or content management based on specific legal frameworks and cultural norms: the words we encounter online—and those we do not—are co-determined by a host of factors which shape the boundaries of the digital information environment.

2These factors gain in importance to the degree that online interaction—the presumably ‘virtual’ world—comes to progressively constitute social realities. And in many ways the Internet is a discourse environment in the truest sense of the term.^[1] It is fundamentally based on information exchange and speech acts of various types. Yet while most acknowledge that external force shape online interaction, ‘the Internet,’ ‘social media,’ or other such signifiers are commonly employed as what social science may refer to as a causal variable: ICTs are thus primarily regarded through their capacity to effect, affect, cause or otherwise alter social constellations and individual behavior within the offline world.

3In contrast, I will substantiate the claim that the increasingly important realm of deliberation as it is provided via digital, networked media cannot be regarded as an ephemeral (cyber-) ‘space’ or blank-slate infrastructure. Instead, we would do well to conceptualize such discursive environments as a complex set of forces, which intertwine, converge, and overlap in any given form of digital interaction.^[2] Moreover, despite perceived processes of decline regarding other political and socio-economic spheres, the United States currently still subsumes various important nodes of power governing many of the dynamics addressed herein. This applies to both the notion of the US as a specific set of political and economic institutions, as well as a set of ideas shaping cultural values and practices.

4I will draw a distinction between a set of factors I have classified as systemic and agential. These will generally overlap and may become indistinguishable within a particular interaction. This is why there are various pitfalls associated with this classification, which is invariably a highly artificial one. Nevertheless, I hope it might prove useful in exemplifying the key building blocks of my argument and illuminate the important role played by the US in shaping (digital) realities.

Systemic Factors

5Systemic limitations to online interaction may include technological affordances and infrastructures, languages, as well as the potentially unintended consequences of parameters set by media specificity. These will be instances in which forms of communicative interaction are inhibited by factors, where no immediate form of agency plays a role in deliberately influencing a given speech situation. This is not to say that these are necessarily ‘neutral’ forces—they themselves might come to reflect power dynamics and distribution. Nevertheless, there is a conceptual line to be drawn between forces governed by immediate interests and those where this is not directly the case.

Limits of Infrastructures

6The Internet is habitually compared to a web, net, highway, or, infamously, a “series of tubes” (Kliff). To some degree such metaphors are sensible: The digital transfer and exchange of information on a global scale is made possible through a vast infrastructure of telecommunication cables, server hubs, and nodes connected to the network (Kliff). The useful distinction between “bits and atoms” (Livingston and Walter-Drop) differentiates between the material infrastructures—atoms—which are at the same time intricately linked to digital infrastructures and units of information—bits. The organization and scope of the latter will be intricately linked to the mode of transportation afforded to them by the networks through which they are accessed and shared. Likewise, the bandwidth afforded to a specific set of information might come to be favored by a particular type of infrastructure, thereby encouraging its exchange.

7The type of access users enjoy therefore co-determines forms of online engagement and interaction: dial-up modems will facilitate a very different form of communication, for example, than the use of mobile devices, which are perpetually online. We may also consider the evolution of social interaction in online environments from simple text sharing (laced with the occasional colon-bracket based emoticon), to the use of complex and socially sophisticated memes, first in JPEG and increasingly in GIF or short video clip formats.

8The ability of users to employ such increasingly sophisticated modes of communication, which may convey complex emotions or express cultural practices within a single or a rapid succession of images, will be directly dependent on the hardware and bandwidth available. The same goes for the rapid exchange of visual media via live-feeds or peer-to-peer messaging, the immediacy of which can come to be seen as an indicator for the authenticity of an experienced event. As recent scholarship on political participation in the digital age has found, networked communication might increasingly come to rest on the co-creation and rapid dissemination of cultural modes of production and reappropriation (Bennett and Segerberg). The ability to participate in such forms of interaction directly depends on modes of access to the appropriate infrastructures.

9The connection between physical infrastructures and digital content also goes a long way to explaining why debates centered on technocratic sounding terms such as network-neutrality regulation often elicit such heated and pervasive political disputes. The same goes for the socio-economic and political participatory problems arising from ‘digital divides’ within populations on a national and global scale. The fundamental questions of which bits can be provided via physical infrastructures, quickly turns into complex issues concerning the nature of free speech, representation and social inclusion. Access to particular forms of ICT infrastructures may therefore increasingly come to determine not just who gets to speak (i.e. transfer information) but who gets to speak in what manner.

Confines of (the English) Language

10Any system of symbols is by definition finite and rooted within a particular regime of signification and meaning making. One of many ways to think about such systems would be to focus on the complex dynamics between language and the political realities it may create (Edelman 11). This connection implies that certain thoughts are not thought, potential observations never made, ranges of experience never shared, if they are inhibited by the limits of one language system. From a technological standpoint, any such type of signifying symbol may come to be transported from any given point A to point B within the globally networked, online environment. The spectrum of potential modes of communication thus seems virtually limitless—at least theoretically.

11When empirically examining the use of language on a global scale, however, clear patterns emerge. These seem indicative for factors inhibiting the modes of communication. What is more, they correlate with a number of external and material factors: As of November 30, 2015, the best estimates based on a number of reliable data sources count 25.9% of the world total of Internet users as English speakers and Chinese speaking Internet users trail at 20.9% (“Top Ten Internet Languages”). This means that about 75 % of all Internet users are not native English speakers.^[3] Yet despite this 3-1 ratio, the most recent analysis currently available concludes that, “English is used by 53.6% of all the websites whose content language we know” (“Usage Statistics”).^[4] Russian, the fastest growing web-content language, trails in the single digits (6.4%), while Chinese comes in at a mere 1.9% of all website content. Various explanations might account for this, ranging from early adoption arguments, the ‘English as a Global Language’ argument pointing towards the language’s unrivalled international importance,^[5] to more utilitarian explanations, which might see the benefits of accessing or contributing to an extended network of English-based information.

12The dominance of one particular language might very well be a temporary state of affairs. Unsurprisingly, compared to other languages, the ratio of English speakers connected to the Internet as well as the amount of web content has been steadily declining over the past years (Crystal, “Internet” 231).^[6]  Yet while Internet access is progressing rapidly among non-native speakers, the amount of web content is trailing behind at a slower pace. Various incentives appear to presently be in place, which cause the use of digital communication via the spectrum of available services to rely first and foremost on one of the dominant sets of languages.

13For many parts of the world this is English. Other parts of the world such as East Asia (Chinese Mandarin), Eastern Europe (Russian) or South and Central America (Spanish) might have other forms of more immediately hegemonic languages. Hence, the top ten languages spoken by Internet users (out of which places 3-10 are already confined to single digits) comprise almost 80% of all Internet users globally, leaving roughly 20% for all other languages combined (“Top Ten Internet Languages”). Furthermore, what connects these regions of the world linguistically tends to be the fact that English is often spoken as a second language—much more so than, say, Russian would be spoken in Latin American countries. Hence it is English that ultimately facilitates and shapes the exchange between various parts of the world dominated by various regional hegemonic systems.

14Like many of the technological factors giving shape to the possibility of communication, a common language is at once a useful mode of disseminating thoughts and observations, yet at the same time limits the scope and range of the potential content being created and exchanged. Though trend lines and the shifting usages might make this factor far from totalizing or rigid, it nonetheless remains an important factor currently shaping the content of many online interactions and thus fosters specific modes of perceiving and communicating reality.

Boundaries of Media Specificity

15Of late, scholars have become increasingly wary of the term “Internet,” as the term seems to encapsulate too many, starkly diverging modes of social interaction (Farrell). The various forms of digital networks facilitating information exchange become ever-more broken down into specific categorizations such as “traditional” websites, email, (micro-) blogs, vlogs, social media, etc. Even more specifically, these networks are divided into the various prolific platforms such as Twitter, Facebook or more personal messaging systems like Skype, Facetime, WhatsApp or Snapchat. The reasons for this might be the same as those making these platforms popular with their user base: each of these services defines its own user niche by allowing for specific modes of representation and communication.

16While the forms of media available to users widen in scope, the possibilities for communication are far from endless within these admittedly abundant environments. Rather, each of these platforms or services brings with it an ever-more specific discursive logic in the way interaction will be structured. These specificities create different incentives for different kinds of communication and social interaction. Yet the parameters set by a specific medium do not just enable modes of communication; they also shape and delimit what is being said. As Ithiel de Sola Pool remarked as early as 1983 about novel communication technologies: “There is a tendency for any one user community to adhere to common standards” (146).

17This means, for example, that the individual user might find herself refraining from typing, writing, or ‘speaking’ in a particular manner on one social media platform and in full view of colleagues, family, and friends. Secondly, the ‘virtual’ subjects thus constructed, might differ immensely, depending on the medium used to construct them. For sophisticated forms of social media, used for dating or professional networking, where the digital avatar comes to closely resemble the physical person, this ensures complex interdependencies and feedback dynamics between the subject and its environment(s).

18This mediated blending of the private and the public might also come to hold immediate political implications: A web-based messenger service, for instance, which promises encrypted transmission and strongly limited accessibility, might enable (or provoke) very different speech acts than one whose privacy appears to be compromised. Likewise, anonymity and privacy experienced in the virtual world may lead individuals to experiment with different personas and lifestyles.

19Furthermore, such dynamics might not only refer to individualized acts of expression. Specific media environments may come to foster for a particular form of communication and content creation. In predefining a particular response, Facebook’s “like” button, for instance, spawns a very specific type of post. The recent addition of other emotional responses is very likely to have the effect of widening the spectrum of interaction between the platform’s global users. Similarly, the collaborative but hierarchically structured process of knowledge accumulation enabled via the website Wikipedia leads to a very specific form of discourse guided by the negotiation processes of deliberative truth-finding. Twitter has fostered novel modes of large-scale communicative action, whereby outrage and polarization can hone in on specific users or particularly contentious topics. And users of discussion thread based websites like Reddit, Imgur, or 4chan will quickly become aware of the specificity of discursive culture that tends to dominate these sites.

20Over time, the possibilities afforded by these technologies thus foster specific logics of how debates are structured, which comments elicit responses and which posts garner attention. Here we witness the emergence of specific norms and practices—what works and what does not—via specific forms of social media. An immediate connection exists between the interactions these sites explicitly allow for and encourage, and those they implicitly prohibit. In terms of the specific role played by the US here, these factors bear consideration in conjunction with state power, market consolidation and the specific cultural environment which gives shape to the majority of media currently employed worldwide. These are the forces that I turn to next.

Agential Factors

21What we may refer to as agential factors, are tied to specific sets of political and economic interests, institutions or legal frameworks, as well as individual forms of human agency involved in content moderation. Here, specific modes of communication are shaped by external forces linked to the perspectives, motivations and behavioral patterns of individuals or groups thereof. These can be located on a macro level (states, markets, international institutions, etc.), a mezzo level (governance structures, various social fields, corporations, etc.), and a micro level (individual acts of content moderation, curating, etc.). In the following passages, I present brief overviews and examples for some of the most salient domains in which agential forces may shape the contours and content of online interaction.

Interests of the State in Governing the Net

22Contrary to early predictions (Barlow), the emergence of the Internet has not spelled the end of state power. Writing as early as 1999, the law professor Lawrence Lessig has stressed the roles of regulatory regimes in shaping what he referred to as the laws of cyberspace: “The original architecture of the Internet made regulation extremely difficult. But that original architecture can change.” Lessig went on to predict that, “under the architecture that I believe will emerge, cyberspace will be the most regulable space humans have ever known” (32).

23In light of the Snowden revelations, it may almost seem quaint to question such assessments. Critics of US ability to monitor global information flows, such as Wikileaks founder Julian Assange, have come to conclude that, “the very concept of the Internet—a single, global, homogenous network that enmeshes the world—is the essence of a surveillance state” (“Julian Assange”). Indeed, this interpretation of the contemporary global communication infrastructures seems to be shared by the internal communication of the US security apparatus. Within the PowerPoint slides made public by Snowden, the States are depicted as the “world’s communication backbone,” due to the fact that “much of the world’s communications flow through the US” (“NSA Prism Program Slides”). The NSA’s STORMBREW program, which is aimed at accessing and monitoring key communication choke points around the globe, takes advantage of this fact. As the journalist Glenn Greenwald points out, this privileged form of access is, “a residual by-product of the central role the United States had played in developing the network” (107).

24The potential chilling effects that such surveillance power may unfold can perhaps best be exemplified via the now infamous sentiment expressed by longtime Google CEO Eric Schmidt stating, “if you have something that you don’t want anyone to know, maybe you shouldn’t be doing it in the first place” (qtd. in Tate). As part of global FBI sting operations against cybercrime and online piracy or as efforts in the ‘global war on terror,’ such dynamics have been expanded to potential subjects and targets for law enforcement outside of clearly demarcated territories.

25Yet we would do well not to confuse an unprecedented concentration of power with omnipotence. While it may be true that the US government currently holds an unparalleled position from which it is able to mobilize massive resources to observe and potentially dole out punishment, the impact on discursive power this might pose are nonetheless limited. Instead of the ability to actually regulate and actively shape speech environments, the ability to exercise power may rest primarily on the capacities of military might and political influence.

26While the US may have had a head start, it did not take other states and their institutions long to catch up. Governments and legislatures have come to view and treat the Internet as an important “field of governance,” as Laura DeNardis refers to it in her article “Emerging Field of Internet Governance” as well as in her monograph Global War for Internet Governance. How are the IP addresses globally distributed and URLs agreed upon? What are the key decision points, which ultimately give shape to the current and future architecture of the Web? Such technocratic debates quickly morph into politically charged questions, which come to be hotly contested in international settings. The increasing ability and interest of states to govern online infrastructures will likely lead to very different types of Internet experiences for various locales. Depending on which state their ISP might be based in and which IP addresses are being used, the content, sites, and discourses individual users might have access to will gradually come to vary. This applies to Wikipedia entries, academic texts, dissident writing, movies, music, images, etc.

27As DeNardis further points out, the question of Internet governance should therefore not be reduced to single state interests or a definite set of regulatory institutions, “because relevant actors are not only governments” (“Emerging Field” 1). Instead, in any key decision making moment, there may be a complex interplay of international institutions, powerful state governments and market forces at work. In the US, the domestic discussions around these developments have therefore not been limited to the political implications of the surveillance state. In regard to government regulation of industries or net infrastructures, heavily polarized debates have ensued between various stakeholders arguing for or against various modes of government regulation and copyright laws.

28Legislative proposals aimed at fighting the global spread and distribution of copyrighted material online, for example, have already revealed deep rifts between the emergent tech sector, ISPs, and the ‘traditional’ branches of the entertainment industry. In early 2012, the introduction of two bills to Congress, best known by their acronyms SOPA and PIPA,^[7] led to massive coordinated efforts by the likes of Google and Wikipedia: A one-day blackout of popular web services raised public awareness against a perceived effort to alter the regulatory foundations of the Net. Hence, interest in this topic was not confined to the US political sphere and civil society, but made headlines and impacted Internet users worldwide (Faris et al.).

29Such blending of counter-veiling forces makes for interesting dynamics of political and legal disputes, involving questions of state sovereignty, free enterprise, and the rights of individual citizens. In the EU, for example, legislation has been passed affording their citizens the “right to be forgotten,” by forcing Google to remove particular links from their search results. While France has argued that these rules should also apply to the domain name Google.com, Google has stated that this would be at odds with the First Amendment of the US Constitution—the supreme law of the land in which the company bases its operation (Naughton).

30When such disputes concerning Net sovereignty emerge, it is highly likely that they will involve complex combinations of state specific and international institutions, coupled with complex market factors tied to specific legal frameworks. Nevertheless, due to the Internet’s current architecture it also remains highly likely that when such struggles do occur, US state or market interests will be involved.

Interests of the Market and the Web Economy

31The main proprietors of what we mean by the digital infrastructures (the Web) running on top of the physical infrastructures (the Net) are large, privately owned corporate conglomerates, the most important of which are currently based in the US (Bellamy-Foster and McChesney).^[8] Why is one particular sector located in the US in such a powerful position? Analogous to the US’s powerful security apparatus, one answer to this is technological progress and the dynamics inherent to powerful servers and computational capacities (Lanier 113).

32Furthermore, specific regulatory frameworks also favor the US tech-sector: Through the system of copyright and patents the laws governing the ownership of digital infrastructures, online platforms and closed digital environments are increasingly set up to build on one another. Latecomers (i.e. late developers) are therefore at an inherent disadvantage. Once a standard is universally accepted the choice they would face is to develop within this environment or remain a digital outcast.^[9]

33While it remains debatable whether or not this applied to the early open-access structures of the WWW, the current era has seen the rise of global monopolistic communication structures provided by the likes of Microsoft, Google /Alphabet, Apple, Amazon, or Facebook, whose names have practically become synonymous with what we mean when we refer to the Internet. Hence, the time following the Web’s early boom-and-bust era of the late 90s and early 2000s has been referred to as, “the period in which digital media developed has been marked by consolidation and concentration of ownership, not openness” (Klinenberg and Benzecry 10). Others have phrased this more drastically: “[W]e are entering a world of digital feudalism, where a handful of colossal corporate mega-giants rule private empires” (Bellamy-Foster and McChesney).

34Considering the ever-increasing influence of Silicon Valley in regard to policies adopted as legislation and enforced by the US government (Mindock), such trends will also mean that scholars seeking to explain the focus of US foreign policy will need to take the global interests of the US tech-sector into account. The perhaps most ambitious example for such interests are recent activities by Facebook: Under the telling title of “Internet.org” the company announced in 2013 that it plans to provide “free basic” Internet services to remote areas and impoverished states in “Asia, Africa and Latin America,” where access to the ostensibly global information environment would otherwise not (yet) be feasible. The project’s website is host to a series of stock photography images showcasing smiling (but presumably impoverished or otherwise disenfranchised) people dressed in diverse sets of traditional garb. The images are coupled with blurbs of promotional text, candidly explaining its objective: “Free Basics by Facebook […] educates new and previously unconnected people on the benefits and value of the internet—and once they understand and enjoy the internet, we help our partners keep these new internet users engaged” (“Operators”). Yet the information access thus provided is very much contingent on a specific notion of what “the Internet” is and how corporate interests decide to shape it. As the New York Times reported in 2015, critics of Facebook’s project therefore likened it to a “walled garden rather than simply offering a small amount of free access to the whole Internet” (Goel).

Ideology, Cultural Norms, and Individual Forms of Content Moderation

35Why is it important to consider that Facebook is a US based company when discussing the limits of online interaction? What are the implications of Silicon Valley’s global economic reach and its practices of market disruption? No technology as increasingly complex as the applications and social media platforms currently running on top of the Net could be entirely free of some form of ideological features. In fact, even the early culture of openness and collaboration could be one whose legacy could be traced back to a specific blend of cultural values (Wu 201). Vague forms of ideational influence may become especially apparent, when all of the leading, globally operating corporations essentially hail from the same cultural and socio-economic environment. Hence ‘Silicon Valley’ not only alludes to a geographic location or an economic entity, but also a specific set of ideological beliefs.

36As early as the mid-1990s, two prescient observers described the nascent “alliance of writers, hackers, capitalists and artists from the West Coast” as having successfully defined “a heterogeneous orthodoxy for the coming information age” (Barbrook and Cameron 44). What they refer to as “the Californian Ideology,” came to represent an amalgamation of “market economics and the freedoms of hippie artisanship […] made possible through a nearly universal belief in technological determinism” (50). With a global proliferation of these technologies and products, the distinct cultural values they embody increasingly come to the fore, as they elicit pushback effects in parts of the world where they are perceived as embodiments of undue foreign influence.

37There is, for example, a very specific blend of norms at work when it comes down to decisions of what to include in app stores, which images to sensor, and which acts of speech to ban from popular online platforms. According to Evgeny Morozov, this is particularly problematic since “Silicon Valley doesn’t just reflect social norms — it actively shapes them in ways that are, for the most part, imperceptible” (Morozov, “You Can’t Say That”). Just how “imperceptible” the enforcement of such rules truly is, might be relative to one’s own perspective: The deeper individuals are rooted within a cultural framework marked by specific ideological values, the less inclined they might be to recognize these as such. From the perspective of western Europeans, Facebook’s harsh rules concerning exposed female breasts might appear prudish. At the same time, the relatively lax rules concerning graphic depictions of violence or ideologically insensitive insignia has elicited charges of hypocrisy (cf. Toor).

38The lines between politics, ideology, and culture are, of course, blurred and complex. What do we make of Indonesia’s recent ban on same-sex emojis? These have only fairly recently become part of the standard repertoire on Apple and Android operating systems in use globally. Considering the time it took (and is still taking) ostensibly progressive Western societies to begin implementing basic civil rights for their LGBTQ communities, some form of pushback by conservative governments was, alas, probably to be expected. Nevertheless, such developments present a challenge to companies who tend to view their services as a non-ideological means of communication.

39The immediate role played by cultural norms in actively shaping online discourse may become most obvious on the micro-level of content moderation. When questions arise concerning the limits of acceptable content, it is typically individual human beings who determine the limits of acceptable content encountered online. This is why, behind the scenes, the collaborative knowledge repository that is Wikipedia actually has quite hierarchically structured organizational patterns.^[10] And Reddit’s fairly small team of top-rank moderators regularly clash with the wishes and perspectives expressed by large portions of the user base over what will be allowed to surface on their supposedly crowd-sourced content pages (Dredge).^[11]

40In his 2012 article on “The Deciders,” law professor Jeffrey Rosen investigates such structures of accountability at some of the largest social media networks. When a user complaint is registered here, it will typically fall to any one of a group of content moderators to decide how to deal with the matter. However, when this first line of moderation fails, it is passed on to higher-ups working at the companies’ headquarters. The buck ultimately stops at a specific, single person, depending on the social media outfit. It was thus a single high-ranking Google employee who decided on content that was admissible to any of the various “national Googles that are operated around the world, such as Google.fr, Google.de” (Rosen 1536).

41Meanwhile, most social media sites run large, offshore operations to moderate their sites’ content. As their user bases have exponentially grown, so has the need for cheap labor to fulfill these tasks. In his piece, aptly titled “the laborers who keep dick pics and beheadings out of your Facebook feed,” the tech-journalist Adrien Chen tells the story of the psychological toll this takes on individual workers in the US as well as the Philippines. Besides the lower wages, what makes places like Manila interesting to US tech companies is the fact that Filipino and Filipina moderators are typically in tune with the cultural sensitivities of Western audiences—more so than low-wage workers in other parts of the world. Chen cites the media studies scholar Sarah Roberts, who explains how US companies are not always keen to acknowledge the existence of such operations—nor, for that matter, their necessity: “It goes to our misunderstandings about the Internet and our view of technology as being somehow magically not human” (Chen).

Conclusion

42When a journalist decides to frame a news item in a particular way to optimize her story for search algorithms; when new technologies enable particular forms of information exchange, or when an international governance body decides on novel sets of shared standards and protocols—all of these are potential forces which may come to shape both the online environment and the discourses constituting it. Depending on a host of intervening variables, the situations journalists or social media users find themselves in may vary dramatically, just as the stakes involved in shutting down a particularly contentious online forum or in introducing novel forms of ICTs into politically volatile situations.

43Any given number and combination of these factors will be at work in one particular instance or speech situation. Yet the conceptual distinction between the various factors limiting and shaping digital discursive environments allows us to conceptualize the rich spectrum of forces at play within any communicative exchange. It lets us tease out where political factors, power dynamics and material interests may (indirectly) come to influence diverse forms of online interaction—but also where their limits lie. Nonetheless, one way of bundling such diverse sets of factors conceptually is to link them to a specific political regime or cultural sphere of influence. Herein, this has been the US, which currently appears to be the most potent focal point for such forms of analysis.

44At the same time, the argument presented here cuts both ways: With all of the potential factors pointed towards above, one must be careful not to over-emphasize their singular importance. In other words: point to any argument centered on a unitary explanation of one of these factors, and you will be pointing at a straw man in waiting. In contrast, my approach here does not seek to single out any one of these factors or even weigh the broader divide between specific systemic factors and forms of agency against one another. Rather, the goal is to treat them as interlocked and closely related variables, which define the limits, boundaries and filter mechanisms shaping online interaction. It is these formations which subsequently come to decide between those words that may dominate within an online environment and those that consistently remain unspoken.