You Say Tomato
Exploring the impact of partisan language on group identity
Knowing I’m a writer, you’d probably guess that I’m interested in language. Those of you with exceptional memories (and, I suspect, a peculiar interest in semiotics) may also assume from my brief discussion of diaphoristics that I’m particularly interested in how differences arise depending on how speakers of a language use it.
Since languages are dynamic systems that adapt to the needs of their speakers, languages may simplify complex structures, or develop more efficient ways of expressing ideas, according to changes in a population’s communication patterns. The intimate connection between language and culture means that changes in the former reflect changes in the norms, values, and trends of the latter.
Borrowing between languages that come into contact with one another leads to the incorporation of foreign words, phrases, and grammatical structures. Meanwhile, some societies may standardize their languages with centralized institutions, like the Académie Française or the Real Academia Española, to select certain linguistic features as “official” or “correct” to promote uniformity and to facilitate communication.
Not to mention other intersections of language and statecraft that engendered Max Weinreich’s famous formulation, “A language is a dialect with an army and navy.”
I wonder, therefore, if the absence of any ‘American English Academy’ might explain the differences in how the opposing factions of its political class use the same language. Because, if it wasn’t obvious to everyone subjected to American politicians’ continual abuses of the colonial tongue, a 2016 working paper for the National Bureau of Economic Research lays it bare:
America’s two political parties speak different languages. Democrats talk about “estate taxes,” “undocumented workers,” and “tax breaks for the wealthy,” while Republicans refer to “death taxes,” “illegal aliens,” and “tax reform.” […] Partisan language diffuses into media coverage and other domains of public discourse. Experiments and surveys show that partisan framing can have large effects on public opinion, and that language is one of the most basic determinants of group identity.
As one of the shared attributes around which a population forms a group identity, language naturally has a political scope. For example, the Tory faction’s loyalty to the British crown during the American Revolutionary War meant that in the subsequent decades their name became a political epithet throughout the former colonies. Today, however, such rhetoric seems to have taken on a new dimension:
Is today’s partisan language a new phenomenon? In one sense, the answer is clearly no: one can easily find examples of partisan terms in America’s distant past. Yet the magnitude of the differences, the deliberate strategic choices that seem to underlie them, and the expanding role of consultants, focus groups, and polls suggest that what we see today might represent a consequential change. If the language of politics is more partisan today than in the past, it could be contributing to deeper polarization and cross-party animus, both in Congress and in the broader public.
Since the study analyzed the United States Congressional Record, the potential consequence of a more polarized public lies outside its remit. Nonetheless, that seems to have been among the authors’ top concerns for further research. As they write in their paper’s conclusion:
A consistent theme of much prior literature is that political polarization today—both in Congress and among voters—is not that different from what existed in the past. We find that language is a striking exception: Democrats and Republicans now speak different languages to a far greater degree than ever before. The fact that partisan language diffuses widely through media and public discourse implies that this could be true not only for congresspeople but for the American electorate more broadly.
Subsequent years saw further exploration of how different languages prevail among different ideological factions of the American electorate. In 2020, researchers at Carnegie Mellon University used computer translation tools to analyze comments on more than 200,000 videos from the YouTube channels of CNN, Fox News, MSNBC, and OANN to investigate ideological differences in the use of language among the general public.
Unfortunately, those researchers discovered that earlier fears of a public divided along lines of tribal partisanship were, indeed, well-founded. Byron Spice reports their findings:
[I]t seems that viewers of right-wing outlets think of “Burisma,” in the same way that their left-wing counterparts think of “Kushner.” A “protest” to one set of viewers is a “riot” to another. For one, it’s a "mask," to another, a “muzzle” […] Even more extreme, some right-wing news viewers use “BLM” in the same context as left-wing news viewers use “KKK” […] “Some of these so-called misaligned pairs seem pretty obvious,” said Mark S. Kamlet, University Professor of Economics and Public Policy. “But it’s surprising how different some of them are. It gives you a sense of the really tragic polarization that exists today.”
The researchers for the National Bureau of Economic Research mentioned earlier had merely posited that reporting framed in partisan language might seep into popular discourse. Four years later, it seems that researchers for Carnegie Mellon University had confirmed their predecessors’ suspicions.
However, one shouldn’t take it as a given that only the media can account for the accelerating loss of a common language for the American public. Writing in 2022, Harry Cheadle offers a more nuanced explanation for our divergent dialects, or at least for the liberal faction’s:
This is Style Guide Liberalism: a fixation on terms and language that is well-intentioned but inevitably creates a murky layer of jargon between speaker and listener, writer and reader [...] Highly educated people in the bubble know their way around jargon. They’re used to adapting to new ways of speaking and writing; at every new job they’ve likely had to internalize a fresh lineup of buzzwords and acronyms, a set of tools that allows them to demonstrate their knowledgeability without having to know anything. When they got wind that everyone was putting pronouns in their email signatures and saying “Latinx,” they got with the program easily. They know language is powerful and important, and they know that buzzwords impart a kind of insider status to those who can wield them confidently.
(Side note: I really didn’t like “Latinx.” The only way it made sense to me is if you pronounce the X like the one in “Xola” but I only ever heard it pronounced like the one in “X-Men.”)
Cheadle argues, in other words, that the language of American liberals isn’t determined just by political ideology, but by economic class. That dialect’s vocabulary developed in response to the social rewards for its use as presented to what Barbara and John Ehrenreich called the professional-managerial class (PMC): “a formation specific to the monopoly stage of capitalism” consisting of “salaried mental workers who do not own the means of production and whose major function in the social division of labor may be described broadly as the reproduction of capitalist culture and capitalist class relations.”
Those social rewards, however, comprise more than just the validation of their peers. As Cheadle observes, enterprises and other institutions support and reinforce the professional-managerial dialect:
While most people greet these arguments over terminology with a blank stare, corporations love Style Guide Liberalism. They love it for the same reason wealthy white liberals love it: It costs them nothing to use the “right” words. [...] Sometimes this stuff is criticized as being the product of progressive overreach, a.k.a. the old “political correctness gone mad” trope. But Style Guide Liberalism is only loosely connected to progressive politics. Really, it’s an expression of the worst kind of cynicism—the notion that we don’t really need to reform society or power structures but merely slap new labels on things. It’s a dodge, a pathetic sop to the left from corporations and other powerful institutions who at bottom don’t give a shit about any of this but assume that invoking on-trend progressive words and phrases will make up for all the injustice and misery they cause.
Here, I suspect some of you will discern echoes of previous Radio Free Pizza dispatches locating the recent-years zeitgeist of ‘woke politics’ as an effort to subvert the revolutionary zeal for economic justice following the 2008 global financial crisis, the 2011 Occupy Wall Street movement, and Bernie Sanders’ 2016 campaign for the Democratic presidential nomination. Framed within the broader historical context covered in our last free dispatch, what I’ve called here the professional-managerial dialect—cultivated as a reactionary effort to stifle revolutionary discourse—amounts therefore to a kind of cultural austerity (adding to Clara Mattei’s fiscal, monetary, and industrial varieties) imposed through projects of social engineering undertaken by financial interests to maintain the capital order.
Of course, social movements and cultural shifts have always introduced new terms and expressions to languages as spontaneous evolutions to reflect the changes in its speakers’ lived experiences. Surely, many of the changes introduced have been met with resistance. (If I’m correct in recalling that today’s French descends from the medieval Parisian dialect, then I suspect this may have upset a few Lyonnais at some time or another.)
However, I believe the contemporary conditions of late-stage capitalism and a deadlocked American duopoly make the modern professional-managerial dialect a linguistic development without clear historical precedent.
Moreover, and in keeping with its context within globalized neoliberalism, this cultural austerity extends beyond the Anglosphere: hence the October 2022 letter from the conservative Peruvian political party Renovación Popular to their country’s Foreign Minister “rejecting any OAS [Organization of American States] document which may force Peru to ‘grant improper recognition to transsexual men and grant them the rights of women’” as Jacob Kessler reports.
Here, I suppose, I risk finding my own strange bedfellows among social conservatives of Latin America: despite my interest in protecting demographic minorities from violence and subjugation, I can’t deny in myself a growing skepticism of the discourse promoting trans rights. The coincidence between its increasing presence in public debate with agendas of social engineering led by private interests seems suspicious to me now. Whether or not those agendas were designed to smother mass movements favoring economic justice, the support they’ve receive according to the reward structure of ESG financing correlates too closely for comfort with the ongoing imposition of technocratic austerity that enriches private business at the expense of public services.
Suspicious now of anything off which I catch a whiff of technocracy—and agreeing too with the similarity that Richard Wolff draws (covered earlier on Radio Free Pizza) between the modern partnership of Big Tech and the U.S. intelligence community in comparison with the economic structure of authoritarian fascism in the 20th century—I therefore became curious about the extent to which the oligopoly of the tech industry might reinforce the professional-managerial dialect.
For that reason (or otherwise just for a lark) I asked ChatGPT, “What is a woman?”
The application responded:
A woman is an adult human female. In terms of biological sex, women typically have two X chromosomes (XX), but it’s essential to recognize that not all individuals who identify as women may fit this biological definition. Gender identity is a separate concept from biological sex, and some people may identify as women regardless of their biological sex.
Gender identity refers to a person’s internal sense of their own gender, whether it aligns with the sex they were assigned at birth or not. Some individuals may be assigned male at birth but identify and live as women, and they are also considered women. Gender identity is a complex and diverse spectrum, and it’s important to respect and acknowledge each person’s self-identified gender.
Women, like men, come from various cultural, social, and economic backgrounds, and they contribute significantly to all aspects of society, including politics, science, arts, business, and more. Understanding and respecting gender diversity is crucial in promoting equality and inclusivity for everyone.
Compared to the above, I prefer Lacan’s answer (“Woman does not exist”) for the accompanying proposition: that something like an all-encompassing archetype of a gender role exists in the unconscious for men—“The Man,” I guess we’ll say—without the same for women. While I certainly think it’s open to criticism, I nonetheless appreciate the proposition’s attempt (as I understand it) to account for the phenomenon of gender identity itself, rather than simply saying it exists and emphasizing its importance.
But regardless, of my own preferences (and despite my aforementioned suspicions), I think it’s hilarious that the public debate over whether women’s rights and trans rights are complementary or contradictory should cause political figures to trip over their own words. At the 2022 confirmation hearings for Ketanji Brown Jackson as a nominee to the U.S. Supreme Court, Sen. Marsha Blackburn (R-TN) asked the nominee, “Can you define the word ‘woman’?”
“I can’t [...] Not in this context. I’m not a biologist,” Jackson answered.
The exchange touched off a minor storm in the mainstream press, and of course opinions were polarized. Kylie Cheung dismissed it as a bad-faith inquiry:
Conservative politicians have chosen to rally around dehumanizing trans people and obsessing over trans children’s genitals as a winning political strategy to mobilize an increasingly and unapologetically hateful base. [...] Narrowing the definition of “womanhood” to a conservative idea about what women should look like and behave like has become a galvanizing issue for Republican voters, who are terrified, in general, of social progress.
Elle Purnell disagreed, noting that the nominee used the words “woman” or “women” fourteen times in the first two days of her confirmation hearings, and thirty-four times in her legal opinions. “Repeatedly using a word she says she ‘can’t’ define is not a good look for a potential Supreme Court justice,” she writes. “Blackburn should ask Jackson: What did you mean by ‘woman’ all of those times you said and wrote it?”
As Greg Wilson reported, Purnell isn’t alone in her concerns:
[S]ome critics, including Sen. Ted Cruz, said buying into the idea that gender is a social construct could have serious implications for Jackson’s role as a Supreme Court justice. Not having a working definition of gender could make it difficult, if not impossible, to rule on cases revolving around women’s rights, they said.
While I suspect that the aforementioned Cheung would find Sen. Cruz as perfidious as Sen. Blackburn, he appears to have become an unexpected bedfellow to some supporters of federal abortion rights. As Matt Lavietes wrote four months later:
The toppling of Roe v. Wade has catalyzed a heated debate among some abortion rights supporters over whether gender-neutral language—like “pregnant people” instead of “pregnant women” and “chestfeeding” instead of “breastfeeding”—should be used in advocating for abortion rights [...] A number of leading abortion rights and civil rights groups [...] and some Democratic lawmakers [...] have shifted their messaging in this way to be more LGBTQ-inclusive [...] Even the National Education Association, the country’s largest public teachers union, recently proposed changing the word “mother” to “birthing parent” in contracts.
To my eyes, it’s noteworthy that this professional-managerial dialect should contribute the vocabulary for labor contracts under the neoliberal ideology of late-stage capitalism. However, the aforementioned researchers at Carnegie Mellon University seem not to have identified any partner for “woman” to examine the two within their schema of misaligned pairs: an intriguing absence, in fact, since “[s]ome of the pairs map to the same entity […] while the rest map to completely different entities and beliefs”—but perhaps alternatives hadn’t yet reached household use by the time of their research, either on the liberal mainstream platforms studied or among those platforms’ audiences.
While researchers for the National Bureau of Economic Research and for Carnegie Mellon University have persuasively identified increasing divergences in the languages used by legislators and by television commentators, the aforementioned partnership between U.S. intelligence and Big Tech leads me to suspect that the mainstream media no longer remains the preeminent vector for the diffusion of ideological dialects.
Perhaps on adjacent suspicions, the research scientist David Rozado investigated whether OpenAI’s content moderation systems favored one demographic over another when classifying hateful content.
That system, he writes, “classifies a variety of negative comments about some demographic groups as not hateful while flagging the exact same comments about other demographic groups as being indeed hateful” with “differential treatment of demographic groups based on gender [being] one of the starkest results of the experiments. Negative comments about women are much more likely to be labeled as hateful than the same comments being made about men.”
That differential treatment manifested in other ways, Rozado finds: “Another of the strongest effects in the experiments had to do with ideological orientation and political affiliation” with the system delivering moderation decisions “more permissive of hateful comments being made about conservatives than […] about liberals.”
The demographics considered range not just across genders and political alignments, but ethnicity and religious orientation. Differentials in treatment about match what I’d expect from a liberal bias. Quite intriguingly, however, about comments determined hateful in other contexts he writes, “[t]he same [negative[ comments are more likely to be allowed […] when they refer to high, middle and low socio-economic status individuals,” among other demographics unrelated to wealth—allowing, in other words, hateful comments about economic class, regardless of which class in particular.
While Rozado offers no further analysis on its permissive attitude for hateful content based on economic class, “the fact that OpenAI content moderation is more sensitive to hateful comments being made about women than about men […] suggests that the pattern is systemic rather than occasional.”
Nonetheless, its differential treatment based on political ideology seem to concern Rozado the most: “AI systems that are more lenient about one mainstream political group than another,” he writes, “feel particularly dystopian.”
I share the sentiment, and I’d add to it that the apparent degree to which a technocratic oligopoly can conceivably engineer and reinforce the use of an ideological dialect throughout a given culture feels just as dystopian. One can easily imagine that the same bias found in OpenAI’s content moderation system would appear in the company’s flagship product, ChatGPT. For more on that topic, Rozado provides further research.
Reflecting on the dynamic relationship between language, culture, and politics, I believe the ongoing debate over the standard definition of “woman” offers us a fascinating contrast to the typical process by which languages adapt to the needs of their speakers. Certainly, social movements and cultural shifts have introduced new terms and expressions to those languages that necessarily empowered the masses to express that of which one previously had literally no words to speak. In fact, the feminist movement of the 1960s and ’70s provides a striking example, popularizing terms like “glass ceiling” and “sexual harassment” to increase awareness of gender inequality and contributing linguistic changes that proved instrumental into shaping the public discourse around gender, thereby paving the way for further progress in women’s rights.
Today, however, the public debate surrounding women’s rights and trans rights doesn’t exemplify so much the complexity of language, but how its standardization can promote ideological divisions within a society supposedly united with a common tongue. Intertwined as it is with globalized neoliberalism, what I’ve called the professional-managerial dialect seems only to produce the fault lines for further misunderstandings throughout the population. These fault lines spread not just inside the capital order’s imperial core, but throughout its sphere of influence, as seen with Renovación Popular’s opposition to potential efforts from the OAS to confer the rights of women to biological males of the trans community.
Meanwhile, in the United States, Democrats and Republicans use different and frequently incongruent terminology to address the same issues. With each party’s allied media outlets doing the same, it seems to me that the impact of partisan language on group identity has encouraged misunderstandings within public debate and poses increasing barriers to comprehension.
While today it may be only a tiresome formality to switch into a professional-managerial dialect, the increasing changes such imposes on everyday American English (which have been developing since the global financial crisis and the subsequent subversion of revolutionary discourse) threaten one day not just to prevent polite debate, but to render unintelligible communications across ideological divides.
Observing how the social engineering that financial interests promoted throughout the early 21st century has coincided with (if not wholly produced) a public discourse in which the use of language itself has been divided along ideological lines, one also catches a glimpse how the nuances of our language shape our society’s dynamics and reflect our collective identity.
Rather than simply aligning ourselves against our perceived ideological opponents—than, for example, adopting a phrase like “birthing parent” supposedly because calling that parent “a mother” associates that person too closely with a gender identity, but in fact only doing so because conservatives reject the term—I believe it’s both more enlightening and more practical for us to investigate what class interests might be at play in cultivating such a linguistic turn. After some consideration of those, we may perhaps observe the human need for communication begin engendering the language necessary for us to bridge our ideological divisions.
Maintaining the current neoliberal framework, it seems to me, amounts to the ongoing decline of economic democracy. Instead of submitting ourselves to a public discourse that’s encouraged only in the interest of obscuring how the capital order consigns the masses of common folk to precarious lives spent relying on disappearing social services, it’s better for us to risk following in a few of Sen. Cruz’s footsteps and discovering strange bedfellows of our own.