Caton considers himself fortunate to have been a graduate student in the 1970s in the Department of Anthropology at the University of Chicago, where a new cultural paradigm was being forged—variously identified with Clifford Geertz (before he went on to establish the School of Social Science at the Institute for Advanced Study at Princeton University), David Schneider, Victor Turner, and Nancy Munn—that has been called “symbolic anthropology” (among other terms), when into this mix stepped Michael Silverstein, at the time a very young PhD in linguistics who had studied under Roman Jakobson at Harvard University. As he would later tell the story to Caton, “There I was listening to what these folks were saying about ‘culture as symbols and their meanings,’ and wondering what on earth they were talking about, coming as I did from linguistics.” Of course, he was referring not to the linguistics of Chomskyian “transformational grammar,” (hailed as a “revolution” at the time) but to the structural linguistics of Roman Jakobson (among others) that viewed language as a form of communication and that was influenced by the semiotic theory of Charles S. Peirce, whose sign-modalities of icon, index, and symbol Jakobson, of course, had already used in his analysis of what he called “shifters” in Russian verbal categories (Jakobson Reference Jakobson1957). Silverstein’s brilliant theoretical move was to introduce this semiotic perspective, particularly the notion of speech indexicality, into cultural analysis, which he proceeded to do in the seminal article “Shifters, Linguistic Categories, and Cultural Description” (Reference Silverstein, Basso and Selby1976).
The boldness of his move may not have been apparent to us at the time, however, for not only was he critiquing the concept of symbol in “symbolic anthropology,” he was more controversially asserting that what was called “culture” was to a large extent a discursive construction, not in the Foucaultian concept of discourse that was gaining ascendancy in the 1970s, but in the linguistic anthropological sense best associated with Edward Sapir and that today is often called linguistic pragmatics. To put it simply, sign-use in social interactional contexts creates cultural meaning, the stuff of anthropological analysis, and thus our study of “culture” is dependent on linguistic pragmatics. By now these ideas have become commonplace (if not universally accepted), and it may be difficult to appreciate just how novel Silverstein’s interventions were at the time, but they formed the foundation of what would in effect be a paradigm shift within linguistic anthropology, as well as a shift in the way the relationship between language and culture was understood. Over the next several decades Silverstein would forge many of the ideas that have now become standard within linguistic pragmatics—metapragmatics, linguistic ideology, entextualization, the poetics of interactive communication, and so forth—which he developed largely in his challenging introductory course in linguistic anthropology, his many articles over the years, his mentoring of dozens of PhD dissertations, as well as his collaboration with several of his students that resulted in a number of major publications.
This article’s intention is not to examine the theoretical and historical development of this paradigm or the extent of its influence within the field of anthropological linguistics or more broadly within anthropology. Rather, it is to look at one effort, originated by Michael Silverstein, to link up various aspects of anthropological linguistics to the analysis of a key sociopolitical problem, that of presidential speech. If one looks for a cultural area to whose analysis linguistic pragmatics might be said to have made a significant and distinctive contribution—to demonstrate the utility of its approach to scholars who are not necessarily interested in linguistic pragmatics per se but rather in the way speech processes or communication more broadly are central to their concerns—then one has to look no further than Silverstein’s seminal Talking Politics: The Substance of Style from Abe to “W” (Reference Silverstein2003) and the sequel he wrote with Michael Lempert, Creatures of Politics: Media, Message, and the American Presidency (2012). It is on these two texts that this article focuses, and in a sense it updates their survey by encompassing the Trump era and suggesting that perhaps presidential talk itself is undergoing something of a major cultural shift with Trump and the Twitter medium.
The Question of the Base
Throughout the 2016 election, there were constant discussions among pollsters about Trump’s “ceiling”—that is, the theoretical maximum amount of supporters he could possibly win over, given his extreme unpopularity and divisiveness. That ceiling turned out to be approximately 46 percent of voters, not enough to claim a majority win, but just enough to swing the electoral college in his favor.
However, this ceiling quickly collapsed postelection, with a sizable number of Republican voters expressing regret at electing a president who they had expected to pivot toward, if not a more moderate political stance, then at least a more disciplined, less chaotic style of governance. That pivot of course never arrived, and Trump’s approval ratings quickly slipped into the thirtieth percentile, never recovering to their preinaugural numbers.
At this point, what may be a more important metric to analyze is not Trump’s ceiling, but his floor. Throughout the Republican primaries and the general election, and continuing throughout the turmoil of his first few months in office, Trump has enjoyed a consistent base of supporters who remain resolute in their trust and enthusiasm for him. Lempert and Silverstein make note of the increasing importance of understanding a candidate’s base in early twenty-first-century American political campaigns when it comes to understanding whom the candidate is addressing in his or her speeches:
Observe in this connection [addressivity], for instance, the increasing use among political commentators of the concept of a candidate’s base, his or her enthusiastic supporters who resonate with the figure’s Message as built around Issue slogans and Issue shibboleths and other framing semiotic flotsam and jetsam with which a characteristic demographic or cluster of demographics can identify. Political figures seem to absorb energy—or at least maintain the momentum of their Message—from such a socio-demographic base, expansive or narrow as it may be in relation to the larger public, and woe to the politician without such a base in the overall system of Message politics (Reference Lempert and Silverstein2012, 35)
The passage invokes a key theoretical term in the analysis of presidential speeches, the Message.Footnote 1 There is more to say about the Message (for which, see below), but for now it is enough to note that it is also always addressed to someone, a fact that Lempert and Silverstein (Reference Lempert and Silverstein2012, esp. 109–12) refer to as its “addressivity.” This, we argue, can be a sliding scale of inclusiveness within the US electorate. It may have been sound political strategy in the past to try to shape a Message that could be addressed to as wide a constituency as possible (as in the “good old days” of coalition politics), but what is striking about Trump’s campaign, and his eventual win, was his consistent address of his base, often at the expense of a wider reach within the electorate. Two recent polls released by Morning Consult/POLITICO demonstrate this starkly defined phenomenon. In one poll, Trump’s overall support ratings can be seen dropping over the last month (Easley Reference Easley2017b). Another shows the number of Trump supporters who “strongly agree” with his job as president. This number hardly wavers, and in fact climbs slightly during the same time period as the aforementioned drop (Easley Reference Easley2017a). Much has been written about this steadfast group of supporters in Trump’s “base.” They are alternately defined as the white working class and a “basket of deplorables” and are regarded with varying degrees of empathy and disdain by the majority of Americans that disagree with them. And while there are many ways to explain why Trump’s message communicates itself so consistently to this group, it’s perhaps even more important to understand how it does so. This may be a one-off phenomenon, and certainly a candidate like Marine Le Pen failed at her presidential bid in France precisely because she could not move very far beyond her base’s appeal; but the argument of this article is that this appeal to the base is not only a relatively neglected study but is neglected at the cost of understanding both how people like Trump get elected and how they continue to address their base when in office. Trump, in other words, represents the narrow end of what we might call the Message’s addressivity and further may be a harbinger of US electoral politics to come.
Another of the points Lempert and Silverstein stress is that a candidate’s Message appeal is increasingly dependent upon various media, and they point to the increasing importance of the internet and social media such as Facebook and Twitter in that regard (Lempert and Silverstein Reference Lempert and Silverstein2012, 51–52, for mediatization; 24 for Twitter). Our article is an anthropological linguistic analysis of Twitter, one of the first of its kind. However, our reading may diverge from their claims of Twitter and other forms of social media as potentially democratizating political discourse and challenging those in power, a view that in fact corresponds to Twitter’s own trumpeting of its medium’s “liberating” potential (see below). Thus, it has been argued (Levingston Reference Levingston2017) that Twitter has allowed the Alt-Right, an ultra-right-wing movement mobilizing for a white supremacist political agenda, to constitute a “counterpublic” in the Twittersphere, hardly a liberating response, and what we see in the Trump Twitter phenomenon is something similar—in other words, an attempt to co-opt the Message in the public sphere by viciously suppressing all opposition to it. This will become clearer in our analysis of “trolling” on the internet and Twitter, in particular.
The question of how a Message narrates a biography about a candidate (which may, or more likely may not, correspond to how a biography may be written about the candidate by an independent researcher) is also taken up by Silverstein (Reference Silverstein2003) and more fully by Lempert and Silverstein (Reference Lempert and Silverstein2012).Footnote 2 Unfortunately, it is beyond the scope of this article to analyze in depth the Message construction of that biography and its influence in the campaign, except in regard to the character trait of “immediacy” or “genuineness” that Trump’s rhetoric, particularly as it played on Twitter, indexically constructed for his base. Nor can we tackle here what no scholar to date seems to have considered, which is that—despite glaring inconsistencies between that Message and the historical record—the Message-constructed biography did “stick” in the Trump campaign, despite numerous efforts by Democrats and the media to discredit it. One explanation we can offer takes us back to the base, the candidate’s devoted following who will believe in their candidate no matter what, especially if the revelations come from the mainstream media for which they harbor a deep and abiding distrust.
When Twitter first emerged in 2006, it wasn’t entirely clear what the value of the platform might be. What had originally been imagined as a system for facilitating group text messaging across one’s social circle evolved quickly (and organically) into a new form of mass communication. Only in an alternate timeline of history, in which the cost per character of typesetting was so prohibitively expensive that the nascent newspaper industry never evolved past a single page of headlines, can we imagine a parallel to Twitter as a media form. But after the 2008 election made evident the importance of social media for political campaigns, Twitter quickly became recognized as a vital new tool for both electioneering and governance.
The company estimates that there are about 330 million monthly users, largely in the United States but also around the world.Footnote 3 It also claims that the groups Twitter serves are quite diverse, and it is as a result of the platform being able to reach a potentially huge public combined with its accessibility that it has been touted by many to be a “democratic” platform. Vann Newkirk, writing for The Atlantic about the political usage of Twitter, rehashes a familiar “democratization” narrative, arguing that “discursive access to politicians transfers power away from politicians” (Newkirk II Reference Newkirk2016), exposing them to critique and discord that they might be insulated from in a more controlled forum designed for careful, disciplined messaging. It is decentralized, antiestablishment movements like the Tea Party, Alt-Right, and Black Lives Matter that have thrived on Twitter, able to organize millions of people across thousands of events and present a true challenge to public figures. But in the case of Donald Trump, Twitter seems to have been one of the key enablers for his meteoric rise to power, rather than the means to challenge or take away that power. True, many of his followers on Twitter have recently expressed disillusionment with his tweets and turned away from him, but that’s not quite the same thing. There does not as yet exist a mass-organized “trend” on Twitter to consistently attack and discredit him and to call for his resignation or impeachment. Does this have something to do with the medium itself and his ability to manipulate it adroitly?
Before diving into that, question we ought to first consider how Twitter works as a form of writing. There are certain structural limitations that guide how it can be used and, therefore, what kind of content can emerge from it.
First, posts are famously limited to 140 characters or fewer. This is a legacy of its original design as an extension of text messaging, and it is therefore bounded by the constraints of the SMS protocol. Once that was no longer a factor, however, the limit remained, the company’s CEO Jack Dorsey referring to it as a “beautiful constraint” (Ha Reference Ha2016). Indeed, constraints can be beautiful, as haiku poets or adherents to the Danish Dogme 95 film movement would agree. They force one to do more with less, stimulating a creative dialectic between austerity of means and richness of implication. At least this is how it works in theory.
Second, a Twitter “feed” displays posts in reverse chronological order, which descend down the page as a continuous ribbon of loosely connected thoughts extending as far as someone is willing to scroll (or perhaps as far as one’s browser can handle before collapsing under the weight). It follows that in most everyday usage of Twitter, this kind of arduous deep diving isn’t common. Twitter therefore seems to encourage a synchronic, largely ahistorical engagement, which depends entirely on the reader’s own preexisting knowledge, or a willingness and ability to seek out additional research, in order to contextualize a particular tweet. In their efforts to track the inconsistencies or contradictions in Trump’s Twitter statements—and therefore discredit him—mainstream critics have exhaustively scrolled down these feeds to show that on such and such a date he said “A” on a particular issue and then on this or that date said “Not-A” on the same issue, the contradiction supposedly indexing something about his unreliability in terms of either veracity or logical thinking or both. But Twitter users are by design discouraged from engaging in this way, the point being that a tweet responds to the instant of the event and to the immediate co-tweets to the event, after which the Twitterers and the reader both move on.
Third, a characteristic that helps define Twitter’s form is the structure of the retweet. Rather than simply replying to another user, this function actually embeds the content of the original user’s post within the body of the retweet along with their own commentary, encouraging a more collaborative form of disseminating information. A single feed might actually contain many voices, creating a dialogical narrative curated by one user. Twitter’s linguistic ideology is to encourage dialogue and free speech, as well as to enable users to form their own chat communities in which they can pursue topics of common interest. While this has happened, Twitter speech communities have formed, such as the Alt-Right, that have also had a chilling effect by harassing and intimidating other Twitter users (for examples, see below), resulting in the opposite of what Twitter had in mind. At the end of the message may be a “hashtag,” a word or phrase preceded by the pound sign (#), which presumably indexes a particular topic or something about the user himself or herself (this aspect of the form thus encourages the construction of a biography rather like the acronyms one might read in a personal column of a newspaper like “SWM, well-educated and into fine wines”). These words or phrases may be shared by others or may catch the attention of others who either strongly identity with the hashtag or disagree with the post’s contents, allowing for communities of messengers to carry on a dialogue with each other for as long as they want. This reverts back to the point above about the retweet function that allows embedding of tweets within tweets, making possible an extensive dialogue spread across the platform. However, these usually flame out quickly, reinforcing the point that Twitter is largely synchronic.
Fourth, Twitter does not require users (unlike Facebook) to identify themselves by their real or legal names, or to be forced to verify their identity, though they are required to use some sort of Twitter name, or “handle,” often made-up or fake, which can provide anonymity and cover. This anonymity, or so it is presumed, allows users to express views that are of concern to themselves and others and perhaps need to be made explicit in the public sphere but that may at the same time attract criticism or even censure because of their sensitive or controversial nature. Teachers, for example, have often had to face such expressions in classrooms and somehow conduct a respectful and civil discussion around them (it is not always a matter of enabling a minority or subaltern voice to be heard in the classroom but sometimes also a privileged, hegemonic one from the perspective of race or class). How much easier to have a web-based class platform on which students can express those views to their instructors and their fellow students in a way that does not necessarily identify them or lead to their embarrassment and intimidation or fear of compromising their grades or the support of their instructors. This is a vexing question, that is, of how much one should rely on internet media as the forums in which to have these difficult discussions, as opposed to face-to-face classroom discussions, but we bring up the example as a way to explain Twitter’s anonymous format with the intention of promoting speech on thorny topics. But the format can also have the unintended outcome of allowing those who want to verbally abuse others to get away with such harassment with few consequences. The gray area is what legally or theoretically constitutes abuse. It is true that someone who “egregiously” oversteps the bounds of what is considered “tolerable” abuse can be suspended from or even removed from Twitter, and such cases have happened (in mid-July 2016 Twitter “permanently suspended” or in effect removed Milo Yiannopoulos, a British media personality associated with the Alt-Right movement, after his repeated hateful name-calling of actress Leslie Jones). From the very beginning of the platform, the Twitter Company has struggled with how to stop harassment, though with marginal success (see Warzel Reference Warzel2016). Lately, it has started to clamp down on the Alt-Right, which has said it would create its own rival media platform, Gab, in retaliation (Ellis Reference Ellis2016; Nash Reference Nash2016). But it would be myopic to attribute such a license for verbal abuse solely or even primarily to the cloak of anonymity Twitter provides. Whether people willfully engage in what is considered “intolerable abuse” (please explain the difference between “abuse” and “tolerable abuse”) is in the end a moral and political question, not a formal or structural one. We argue that there has been a sea change in the linguistic ideology of “free speech” that has not been adequately examined ethnographically (though arguably it is urgent for linguistic anthropologists to do so). Apart from the question of whether or not the term free speech has been used correctly in its legal/technical sense (a question that is certainly pertinent in the realm of law) by groups who spew all sorts of hateful messages about others in its name, the ethnographic question of what they mean by “political correctness,” which they impugn, or by “free speech,” which they invoke in order to abuse others, has hardly begun to be examined.
Finally, a speech practice on Twitter called “trolling” requires comment, one that may in part explain the seeming oxymoron of “tolerable abuse.” For the analysis of trolling that follows we are endebted to the Harvard senior thesis in anthropology by Ivan Levingston (2017), a perceptive ethnography of the use of Twitter by the Alt-Right movement, in which it is argued that trolling is perhaps their signature speech practice. As Levingston insightfully points out, this practice—called by another meta-pragmatic label as “griefing”—is found in other web-based interactive communities such as Second Life, recently studied, for example, by Boellstorff (Reference Boellstorf2015). Boellstorff attributes this practice to the “disinhibition” residents in virtual worlds feel due to the ability to keep their actual identity secret while coming together under the assumption of shared interests or like-minded views. Among examples of griefing, Boelsttorff mentions a range of speech practices from provocation (a seemingly innocent “playing devil’s advocate”) to harassment by abusive verbal means or disturbing images.Footnote 4 But let us insist once again on the argument made earlier, which is that it is not enough to explain behaviors called “abusive” on the grounds of either technology (“it was the medium that allowed me to do it”) or psychological drives (“it was the lifting of the censors that made my id cry out”), but to a sea change occurring across internet media, campus “free speech” forums, and political campaign debates such as the ones we witnessed in 2016 having to do with ideologies of what constitutes “free speech,” a cultural shift we have not even begun to study ethnographically. However, to get back to Twitter, it can be argued that provocation with the intent of abusing someone, for whatever reason, is meta-pragmatically deemed not only “acceptable” but downright “praiseworthy.” To be sure, and as Levingston points out, “It can be hard to tell when trolls are genuinely expressing their beliefs or just trying to provoke a reaction—and that is partly by design, as it provides cover and plausible deniability for the [Alt-Right] movement’s actions” (2017, 55). A comparison with another tradition might be apposite. We are reminded of the agonistic forums of challenge and retort that can be heard in oral poetic traditions such as are found among the tribes of Yemen (Caton Reference Caton1990), except that in that tradition the “game” of provocation is distinguished from the “insult” of vilification because the latter is not condoned and can even lead to dire consequences if practiced, such as open warfare. To be sure, there is also an ambiguity in the Yemeni tribal tradition of how to interpret a particular poetic gambit as either a “challenge” or a “curse,” but the difference is that no cover of anonymity exists: if a tribal poet vilifies another, everyone knows who the perpetrator is. On Twitter, there are no such consequences of overstepping the moral bounds of the rebuke/challenge, except when the company steps in to evict the “intolerable abuser” from the community of users (rather like the larger moral community in oral traditions).
Trump the Twitterer
In the section that follows, we will return to the five characteristics of the Twitter form, and articulate how each worked to benefit Trump in his rise to power and gain him the trust among his core constituency or base, which has not waivered since. (Although the Alt-Right is a key constituency of Trump’s base, the latter is not neatly coterminous with it, for there are Trump supporters who would not affiliate themselves with the Alt-Right, and indeed Trump has publicly distanced himself from the Alt-Right). One could, and many do, make the claim that Trump masterfully exploited Twitter to spread his message and gain power. But the reverse is also true; Twitter was always primed and ready to create his kind of persona; it was only waiting for Donald Trump to arrive. A medium like Twitter is a Janus-faced figure or a double-edged sword, at once promoting the likes of a demagogue as easily as it might a democrat, Caesar as easily as Cicero.
No doubt because of the relatively recent ascendance of Donald Trump to national political power, the examination of his rhetoric and his use of Twitter are still in their infancy. George Lakoff (Reference Lakoff2017) has analyzed his tweets, arguing that what is masterful about them is that they always seems ahead of everyone else’s in framing the message (in the informational sense), and while that may be true to a certain extent, it leaves out of account the question as to what degree this is due to the medium itself that, as we pointed out above, is almost always focused on the last and latest tweet and is itself a reaction to the “crisis” of the moment. It is really timing due to the way the medium functions more than framing that seems key to Trump’s ability to co-opt the message. The American Ethnologist recently published a forum on Brexit and Trump in which Carole McGranahan (Reference McGranahan2017) wrote a perceptive piece on Trump and what she calls the “political sociality of moral outrage,” noting that although it has always been said of politicians that they “lie,” the number of Trump’s falsehoods seem to usher in an era of unprecedented political mendacity, and she asks how anthropology might understand this phenomenon historically, culturally, and politically. In many ways she is calling for a broader contextual examination of Trump-speak that we have been arguing is necessary. At the same time, however, this kind of analysis must be combined with the features of Twitter’s forms we have described above in order to get a more complete analysis. On the one hand, one of the meta-pragmatic comments Trump’s constituents often make about his speaking is that he is “uninhibited” in the expression of his views (which reminds one of Boelstorff’s comments about Second Life users) and that, as a result, his language is more “genuine” than that of his opponents because he “speaks his mind,” even if what he says turns out to be factually incorrect. To some this may seem paradoxical, but to his supporters it testifies to his trustworthiness more than volumes of factually correct statements uttered by a candidate like Hillary Clinton, whose intentions are entirely suspect to them and indeed can even bring the veracity of those facts into question. Trump’s supporters recast him as a refreshingly candid and naïve speaker who is above the despicable canniness of mainstream politicians, who may speak the truth when it comes to the factual record but lie through their teeth when it comes to their intentions. Of course, what is implied is that Trump’s Message should lead the way in forging a new kind of political discourse in American politics. Whether that shift can take hold beyond Trump’s supporters remains to be seen. On the other hand, when it comes to Trump’s statements on Twitter, it is noteworthy how the medium’s format aids this kind of telling of falsehoods. The medium is geared toward speaking on whatever is construed as the issue or crisis du jour, and “speaking off the top of one’s head” on that issue is a sign of spontaneity and—in relation to what appears to be an emergent linguistic ideology within American politics—an index of a candidate’s honesty and forthrightness, even if that spontaneous comment turns out to be factually incorrect. When “mainstream media” then do the fact-checking and point out either the incorrectness of Trump’s remarks or their inconsistency with his previous tweets, his Twitter supporters have already moved on to his next tweet on the next issue of the moment because the medium favors this form of interaction. It makes fact-checking entirely beside the point to a core of Trump supporters.
Let us now examine Trump as a Twitter user in relation to the five features discussed above. First, there is the character limitation, Twitter’s “beautiful constraint.” It’s not a bold statement to suggest that while it may indeed encourage users to find creative ways to convey their ideas, there’s always going to be a simplicity to those ideas that perhaps doesn’t lend itself to a nuanced discourse on contemporary global politics. And yet, even within this reductive mode of message-making, there are different strategies of simplification that are perhaps more effective in creating an aura of authenticity and trust for distinct audiences. As an exercise in contrast, take this series of tweets posted Barack Obama’s @POTUS page on January 1, 2017:
As we look ahead to the future, I wanted to take a moment to look back on the remarkable progress that you made possible these past 8 years.
We traded foreign oil for clean energy, we doubled fuel efficiency standards, & we acted on a global scale to save the one planet we’ve got.
We brought home more of our troops & strengthened U.S. leadership—leading with diplomacy & partnering with nations to meet global problems.
From realizing marriage equality to removing barriers to opportunity, we’ve made history in our work to reaffirm that all are created equal.
Each tweet is composed as a grammatically complete sentence, carefully linking several clauses to fit precisely within the 140-character limit, something that would have been impossible to do on the fly. Several are accompanied by infographics displaying quantitative data to support the statements. The content addresses as diverse a range of topics as possible, appealing to every interest group (it is not a single base Obama is addressing). Even the language style feels like it was run through a focus group. Using the combination of “we” + verb (past tense), Obama’s speech is both declarative and inclusive (“we doubled,” “we acted,” “we brought home,” “we’ve made history”). What this indexes is not an off-the-cuff remark but a premeditated statement, meticulously constructed by committee before being posted online by an unpaid intern. This appeals to what is still the majority linguistic ideology where American politics is concerned: sentences that are factually true, crafted according to standard written English, and echo a prized tradition of political rhetoric (e.g., the last sentiment about reaffirming that “all are created equal,” echoing the Declaration of Independence).
Just one month later, Trump was occupying the same office, as well as the same @POTUS Twitter handle. On February 4, he posted this reaction after a federal judge suspended his executive order banning entry to the United States of citizens from Muslim majority nations:
When a country is no longer able to say who can, and who cannot , come in & out, especially for reasons of safety &.security—big trouble!
The message also squeaks in just under the limit, at 139 characters. But the route Trump takes to get there couldn’t be more different and carries with it an entirely separate array of indices.
First, the syntax is not crafted according to the rules of standard written English. The conditional clause at the beginning “when a country … ” sets up a second half, but the run-on sentence gets lost along the way, ending with an erroneously placed hyphen and fragment. There are other clear grammatical errors, including an extra space before a comma and a period instead of a space. Not only does this demonstrate a certain sloppiness of typing, it also represents a total lack of interest in making efficient use of the character limit. Removing all the extraneous commas and spaces would free up a total of five characters, providing enough space to add another word, perhaps even to link up the final fragment. But these criticisms are beside the point from the perspective of Trump’s main supporters. The syntax reflects Trump’s preferred Twitter style—short, punchy statements, often accented by an exclamation point—and indeed reflects more closely the style of other Twitter users. The general takeaway is that this tweet was indisputably composed and sent by the hands of Donald Trump himself. He didn’t feel the need to ask for anyone’s counsel beforehand, otherwise they certainly would have spot-checked his grammar. He also didn’t spend any time self-editing, which means he also didn’t take time to self-censor. We’re left with a feeling that this is exactly what Trump was thinking at this particular moment, urgently submitted to the world without filter (or without inhibition). All of these characteristics confirm his “authenticity” to his supporters. Even though Obama was widely recognized as a charismatic figure, he always remained somewhat aloof, with tightly controlled press access and a carefully crafted public image and message. Trump, by contrast, is laid utterly bare for the viewer. There’s no sense of mystery to what he’s thinking or feeling. His Twitter feed feels like a window directly into his head, displaying his thoughts and feelings changing in real time. It recalls the nascent days of contemporary reality television, before slick editing and narrative techniques became a central component of the genre form, which is no surprise, given Trump’s experience with that format. As Lempert and Silverstein remind us (Reference Lempert and Silverstein2012, 15–16), contemporary political discourse takes on many of the genre features of reality TV, and Trump is perhaps the exemplar par excellence of this phenomenon.
While Obama came into office promising a new era of transparency in government, there remained a feeling among many Americans, especially Trump supporters, that he did exactly the opposite. That we now have an administration seemingly incapable of maintaining a consistent, unified message or keeping its chief executive from sharing every thought that comes to his head fulfills precisely that promise of transparency to his supporters, even though it might drive others of us around the bend (as it is probably meant to do). A BBC correspondent interviewed a number of Trump voters over the course of the election. Said a Kansas City supporter, “The other politicians are controlled by their handlers. He’s not,” while another in Florida added, “He doesn’t hold back. You get what he really believes in, even if everything that he says isn’t what is the right thing exactly” (Brown et al. Reference Brown and Ayoola2016).
Second, let us consider the characteristic of Twitter’s immediacy and ephemerality of the feed that helps generate perhaps the most confounding aspect of Trump’s ascendance, or at least for those that oppose him: his apparent ability to say anything he wants with very few long-term consequences. (This is what we referred to previously as the nonstick factor of his Message.) Despite their own reputation for having a low attention span, the mainstream news media might obsess over a single comment for days, deconstructing its meaning, grilling surrogates for further information, and inviting commentators to debate opposing positions equally misinformed by the same lack of information. Trump delights in calling out this cycle as “fake news,” and to a certain degree he has a point. For himself and his supporters, there is an understanding that these statements aren’t meant to be definitive. Everything is a work in progress. His own advisors have continually tried to explain to the media that Trump’s tweets are not policy (Nussbaum Reference Nussbaum2017). George Lakoff describes this characteristic of the Trump tweet as a “trial balloon” (Reference Lakoff2017), designed to pitch a position or perspective to the public without needing to commit or be held accountable to it.
By the time the 24-hour networks have picked up on a particular tweet, it’s already disappeared down the page, and this pocket of the country has moved on. Sites such as the Trump Twitter Archive have been created to maintain a searchable database so that past comments can be easily accessed, but it seems Trump supporters aren’t interested in discovering his past positions and therefore have little use for this kind of service. Besides his tendency to repeat easily falsified information, perhaps the single biggest criticism levied at Trump from all sides of the political spectrum is that he is “unpresidential.” But even this feeling is subverted by Trump’s use of Twitter. Said one supporter, “Pastors sometimes need to be politically correct, and Donald Trump is not politically correct, and I love that about him” (Brown et al. Reference Brown and Ayoola2016). One Twitter user went further, suggesting that the president “fire anybody within ur circle that advises u to stop tweeting. If we wanted “presidential,” we wld have voted 4 Jebb #MAGA” (Garcia Reference Garcia2017).
Third, we turn to the retweet. From the start, Twitter was designed to be a communicative platform, which is why so many writers have enthused about its potential for “democratization” of media. Many of Trump’s supporters feel that his use of Twitter puts them in conversation with him. In an interview with CNN, one supporter from Pennsylvania said, “I feel it’s a great way to reach out to your constituents and create a give-and-take, because people obviously respond to his tweets, retweet the tweets”, while another said his tweeting was “like a modern-day constituent letter” (Presto et al. Reference Presto and Gingras2017). Trump’s particular usage of the medium helps solidify this perspective, often retweeting random constituents from across the country when they support his message. In one case, Trump retweeted a post from a 16-year-old boy in California attacking CNN for suggesting that the president had no evidence to back up his claims of massive voter fraud in the 2016 election.
But Trump’s usage (or perhaps misusage) of Twitter creates his own subversive form of dialogism. More often than not, he actually disregards the retweet or reply function, instead responding to an unknowable third; be it an unnamed individual, the “mainstream media”, or even more abstractly, a particular talking point of indeterminate origin. There is a covert intertextuality embedded in these tweets, a conversation always already taking place just outside its edges (a form of entextualization; see Silverstein and Urban Reference Silverstein and Urban1996). If you’re a Trump supporter, this indirectness can sometimes be interpreted, the third deduced through shared knowledge. This was evident in Trump’s comments on terrorism in Europe, in which he alluded to something that had happened “last night in Sweden.” As there had been no such event, commentators grappled to make sense of his meaning. In fact, what Trump was actually referring to was a Fox News report that had aired the previous night and had focused on a handful of crimes committed by immigrants in Sweden. His lack of attribution may have resulted from his cavalier Twitter etiquette, but perhaps equally likely is an assumption that his readers (particularly his base) are all watching the same news he is and therefore are already privy to a certain set of talking points. Here is another example. Trump posted the following tweet on February 9, less than a week after the “big trouble” tweet, following a Ninth Circuit Court of Appeals ruling against his travel ban:
SEE YOU IN COURT, THE SECURITY OF OUR NATION IS AT STAKE!
The message is strong, resolute, and completely ambiguous. It doesn’t provide any framing to inform its audience of what he’s talking about. It’s assumed that the reader is already following the news and presumably is as outraged as he clearly is. But the most unclear aspect of this tweet, and most illustrative of Trump’s style of dialogism, is the word “you.” There’s no sense of who “you” might actually refer to, if anyone. He might be speaking to opponents of the ban, such as the ACLU, which filed a series of lawsuits on behalf of plaintiffs affected by rejected visas and detention at airports. If that was the case, it seems a bit after the fact, given that the ACLU had in fact already had its day in court. Referring to the Ninth Circuit is even more ludicrous, given that they are the court. Therefore without an actual “court,” and a “you” to see in it, Trump’s tweet is better understood as a performative act rather than a constative, to use Austin’s original dichotomy (Austin [1962] Reference Austin1975). “See you in court” becomes somewhat analogous to “I challenge you” (referring again to the Yemeni tradition). It also refers to Trump’s tendency in his real estate days to threaten to take people to court with whom he disagreed even if in the end he didn’t follow through. It thus alludes to his combative nature. It functions purely to perform defiance and resistance. In this case, who or what is being resisted is unimportant. The only thing that matters is that “the security of our nation is at stake.”
Ambiguity is perhaps Trump’s greatest ally on Twitter, and this kind of generalized, invisible enemy allows him to feed into many of his supporters’ key affect: a prevailing sense of persecution created by a culture of political correctness and increasingly progressive social values that seem to be both forgetting and silencing them. Indeed, one of the more striking campaign slogans resuscitated by the Trump campaign was the Nixon-era “great Silent Majority.” Nixon was referring to a majority that he thought would support his Vietnam War policies against the vociferous protests of the antiwar movement; just so, the heretofore “Silent Majority” (silent except on Twitter) would support Trump’s right-wing politics. This constant dialogue with an invisible aggressor creates a powerful affect of strength through solidarity for Trump’s supporters. Another supporter commented, “Fighting, whether it is justified or not, or whether you are right or wrong, is completely irrelevant, people just want someone that is not afraid to stand up for their cause,” adding, “I know it’s probably crazy to say, but I think the American people felt like it was being bullied by corporate media’s view of them. Trump is the first candidate to figuratively punch the bully in the face, and people love that” (Zito Reference Zito2016).
As for Twitter feature four, the use of a handle or Twitter name to allow for anonymity, this clearly does not apply to Trump, who is a “brand,” as Lempert and Silverstein (Reference Lempert and Silverstein2012) develop the idea, first in his business (the Trump brand or label) and now in his distinctive politics. Even if anonymity were possible, he would shun it. What he craves is constant attention, and so it is imperative that everyone identify his tweets with his personage. That then raises the question of trolling, Twitter feature five, and whether Trump can be said to engage in the practice. Indeed he does, having trolled his long-term critic Rosie O’Donnell, which, in turn, elicited a storm of tweets against her from his supporters, though she is only one of his many victims. The tactic of trolling has also been used against President Trump, to be sure, but the scale of the attacks can hardly compete with the barrage by Trump and his Twitter followers.
Conclusion
Michael Silverstein provided anthropological linguistics with a rigorous sign theory in relation to use-in-context and an explanation of how cultural meaning is constituted in communicative practice that has proven immensely important for scores of anthropologists and linguists. Perhaps his most accessible (though obviously not his sole) attempt at demonstrating the utility of his anthropological linguistics has been in the realm of presidential speech. This article has attempted to analyze Twitter as an anthropological linguistic practice and Donald Trump’s use of Twitter (both on the campaign trail and in his presidency) as presidential (or should we say “unpresidential”) talk. Though Twitter has not gone unnoticed or unanalyzed, ours is perhaps the first linguistic-anthropological analysis of the medium. Ours is also the first sustained analysis of Trump’s Twitter practice. We argue that at least three things need to be considered when trying to explain his rise to power: his Message, the Mediatization of that message through Twitter, and the Address to his base (all theoretical concepts developed by Silverstein and Lempert). We have focused on Trump’s use of Twitter with his base, however, arguing that this may signal a shift in presidential rhetorical strategy from an address of a wide constituency (built on coalitions) to a core constituency (built on a base). However, we have also argued that without a deeper understanding of how linguistic ideology has shifted in America from what Silverstein (Reference Silverstein2003) called the Madisonian ideal—a question that is only now beginning to be explored—the efficacy of Trump’s tweets will still remain largely inexplicable. And so, we have perhaps raised as many questions in this article as we have provided answers for, but this is inevitably the case where one inhabits a topic that is emerging as key to contemporary presidential talk, a topic to which Michael Silverstein contributed with immense insight and even greater foresight.
