Of Martyrs and Robots

Propaganda and group identity

Megan Hyska
A poster with figures whose bodies include the letters C and P. El Lissitzky, Basic Calculus, 1928.
El Lissitzky, Basic Calculus, 1928.

A concern with propaganda has animated American liberal discourse in a new way since the run-up to the 2016 election. On virtually any day since at least July 2016, a naive observer set in front of a bank of screens streaming center-left national news coverage might quickly glean the following: that “propaganda” had something to do with illegitimate attempts to influence an electorate, and that standing accused of this illegitimate conduct were both foreign actors (Russia) and domestic ones (broadcasting bodies like Fox and Sinclair, websites like Breitbart, the Drudge Report, Gateway Pundit, and Infowars). Perhaps less obvious is what this observer would conclude concerning why this messaging was supposed to be illegitimate.

In some cases, the messaging in question was made possible by breaking the law; illegitimacy here would seem easy to place. Russian interference and misinformation campaigns, for instance, involved the clearly illegal hacking of Democratic National Convention servers. Another of this campaign’s elements, the purchase of targeted ads on Facebook, inhabited murkier legal terrain, though the fallout has seen Facebook investigated by the Federal Trade Commission and served a criminal search warrant by the Mueller probe on the strength of suspected elections law violations. Generally though, our observer would surely conclude that an attempt to influence does not need to violate a law to count as propaganda.

When the messaging in question is false, and known by its purveyors to be so, this is of course also sufficient to ground its illegitimacy. But from here, the reasons for regarding some messaging as illegitimate become murky. Many are likely to feel about the illegitimacy distinctive of propaganda as Justice Potter Stewart did about pornography: we don’t have a definition ready at hand but we “know it when we see it.”

Those cases of propaganda that we take as paradigmatic tend to work via conspicuous mechanisms: explicit assertion; exaggeration of negative or positive traits in the subject of visual depiction; emotionally evocative slogans. Nazi posters caricature Jews and proclaim that “Der ist Schuld am Kriege!” (The war is his fault). Contemporary pro-natalist campaigns in Italy depict a white, plausibly ethnically Italian woman, one hand on abdomen, one holding an hourglass, and declare, “La bellezza non ha età, la fertilità sì” (Beauty has no age, fertility does). Reconstruction-era ads from the former Confederate states of the United States depict black freedmen indolently consuming the fruits of white laborers’ toil; as sexually aggressing against white women; or as succumbing to lynching at the hands of a Klansman in vaguely Greco-Roman dress.

A theory of propaganda necessarily proceeds from examples, both contemporary and historical, whose status as propaganda is agreed upon, and aims to say what these examples have in common. The theory of propaganda embodied by contemporary liberal discourse, working from examples like those listed above, has the following underlying form. First, and most often implicitly, it asserts a norm for the deliberation of political agents; standardly, that agents should be rational, should be independent minded, should be educated and curious, should perhaps take as a premise that others in their political community are entitled to a certain sort of respect. Second, it describes the way that propaganda undermines this state: say, by misinformation, by incitement to cowardice, or by application of strategic pressure to the loaded mousetrap of self-serving bias.

Where one wants anti-propaganda and anti-indoctrination tactics to work on masses, a campaign of appeals to the individual virtue of each constituent seems doomed to fail.

En route to considering whether something like this account is correct, we might start by considering one fact that any account of propaganda must reflect. This is that propaganda’s mission of belief change has only instrumental value to its wielder. Political belief is simply a more directly manipulable antecedent of political (in)action, which is what political influencers, be they parties, states, corporations, or interest groups, actually care about.

But of course, sometimes manipulating beliefs, via propaganda or otherwise, is not necessary for the purposes of controlling action. The S.S. leader Heinrich Himmler noted that education in the concentration camps “consists of discipline, never of any kind of instruction on an ideological basis.” While Himmler’s stated rationale for the relative dearth of propaganda in the camps invoked an article of his particular racial ideology–that those interned were constitutionally incapable of receiving doctrine–the strategy can easily be understood purely in terms of cold political efficiency; there is no need to alter the beliefs of those whose actions are already subject to total physical domination. In the United States, mass incarceration and felony disenfranchisement have worked to minimize the power of black Americans as a constituency in electoral politics since tactics like poll taxes and literacy tests were done away with. In jurisdictions particularly affected by these policies, racist law-and-order propaganda needn’t be nuanced to capture the vote of the currently or formerly incarcerated, because these Americans can’t pull the lever for anyone. Once again, altering beliefs is rendered unnecessary, this time because the criminal justice system has already so severely constrained a class of people’s political actions.

In her seminal 1951 work The Origins of Totalitarianism, Hannah Arendt suggested another way of rendering propaganda unnecessary, albeit one turning on an unconventional notion of propaganda. “Wherever totalitarianism possesses absolute control,” she suggested, “it replaces propaganda with indoctrination.” If this pronouncement is striking, it is because we often take propaganda and indoctrination to be much the same thing. For Arendt, though, indoctrination is the process by which doctrine is handed down to those already “initiated in the movement,” while propaganda is for the “outside world.” Propaganda does, essentially, aim to persuade, even where its methods of persuasion consist in shady emotional appeals. Propaganda, in other words, works on those not already conditioned to form a belief merely on the basis of fealty to the party that advises it; but once an agent is susceptible to believing that, as it were, “we have always been at war with Eastasia,” they are past being propagandized.

If indoctrination operates within an organized group with a clear internal authority structure, while propaganda operates outside of it, how do propaganda and indoctrination interact? The answer lies in Arendt’s suggestion that “organization and propaganda (rather than terror and propaganda) are two sides of the same coin.”

To be sure, terror and propaganda are related: terror can secure compliance just as propaganda can. So propaganda and terror are related to one another as the screwdriver is to the socket wrench; they can both be used to tighten or loosen a screw, and thus are different means to the same end. But propaganda is related to organization as the screwdriver is to the screw itself; the propulsion of the latter is the end of the former. For Arendt, the aim of propaganda is to draw people into organized groups of a certain kind, the internal dynamics of which stabilize the alignment of group members’ interests with those of of the group’s recognized authority. What is distinctive about these groups is that members share an intense epistemic deference toward this shared authority; that is, they are susceptible to indoctrination. Propaganda is intimately related to indoctrination because it is what makes it possible.

“Organization” is, for the purposes of the propagandist, an essentially psychological matter; it is an individual’s identifying with the group that is relevant when it comes to his or her recognition of that group’s authorities as, well, authoritative. And it is this recognition which is essential to that group’s ability to indoctrinate its members. But it is worth noting that Arendt’s discussion of organization dwells heavily on how totalitarian movements work to organize individuals in more concrete ways. Supporters are organized into youth groups and women’s groups, paramilitaries and sympathetic professional associations; they are thus given regular opportunities to perform, and so reinforce, their group membership. For all the dark prognostications of creeping fascism that characterized particularly the early days of the Trump administration, it is notable that America’s forty-fifth president has not envisioned some extra-electoral exercise for his base other than attending rallies. Notwithstanding the widespread impression that Trump is deeply attracted to authoritarianism, there is, for instance, no Trumpist equivalent to the Hitler Jugend, the Soviet Konsomol, pro-Putin Nashi, Opera Nazionale Balilla of Fascist Italy, or Duterte Youth of the contemporary Philippines. Had Trump made a study of populist authoritarianism, rather than fixating merely upon such power’s pettiest incidentals, it is hard to imagine he would have made this omission.

In any case, the result of an individual’s being “organized” is a peculiarly intense, but also fragile, commitment to an ideology: Arendt makes the memorable observation that “Members of totalitarian movements, utterly fanatical as long as the movement exists, will not follow the example of religious fanatics and die the deaths of martyrs (even though they were only too willing to die the death of robots)… . Without the force of the movement, its members cease at once to believe in the dogma for which yesterday they were willing to sacrifice their lives.”

The picture here is of a movement which transfigures the mental lives of its members. But it would be an error to infer from Arendt’s emphasis on group dynamics that the individual is completely helpless in the face of the group, as she, certainly, did not think so. From her famous coverage of the Adolf Eichmann trials, through her final unfinished work, The Life of the Mind, she returns frequently to the theme that truly thinking can prevent an individual’s complicity in the worst of group conduct. But while the individual may retain this transcendent power, there’s no guarantee that pushing her to use it will bear fruit. And where one wants anti-propaganda and anti-indoctrination tactics to work on masses, a campaign of appeals to the individual virtue of each constituent seems doomed to fail.

The contemporary inclination to characterize some political environments as “post-truth” reflects an impression, not unrelated to Arendt’s, that the way political agents in these environments form beliefs has altered in some profound way. The perceived change is more fundamental than merely having lowered the bar for the quantity and quality of evidence required for belief. Instead, goes the “post-truth” gloss on this phenomenon, these agents have ceased to aim at truth at all. What Arendt’s description of these agents prepared to “die the deaths of robots” suggests is that this transfiguration is not achieved once and for all but must rather be held in place, at every moment, by the forces of group identification and movement momentum. This is a frightening picture of the power of group identity to distort individual reflection. But it also places its finger on the profound vulnerability of such distortion.

Determining how or whether to apply Arendt’s insights concerning totalitarian propaganda to the contemporary United States is complicated. On one hand, Arendt followed in a tradition, already established at the time of her writing, of acknowledging the debt that the propaganda of the Third Reich, say, owed to American advertising; to this extent, it is evident that Arendt saw a continuity between totalitarian propaganda and methods of persuasion used outside totalitarian contexts. On the other hand, though, Arendt regarded totalitarianism as truly different in kind from even other forms of authoritarianism. While it takes no particular penchant for the dramatic to see authoritarianism in the historical contour of American democracy, and in every projection of the current president’s personality, there is no evidence that the chilly bureaucratic rigor of totalitarianism, narrowly conceived of, is of particular interest to Trump, or even to the true ideologues of his camp. Insofar as Arendt’s description of propaganda is specific to totalitarian states and movements, then, we should be cautious about applying it elsewhere. In particular, we might suppose that those initiated in totalitarian movements are susceptible to the mute acceptance of dogma in a way initiates of non-totalitarian movements are not, simply by virtue of the difference in the movements’ cultures and logics. We might then expect the propaganda/indoctrination distinction to be less sharp with respect to these non-totalitarian movements, as even initiates can be expected to require something a bit more akin to genuine persuasion.

But of course we have ample evidence that group identities have extraordinary power over political decision making in non-totalitarian settings. Social and political psychology of the past century has attested to the way that, for low-information voters particularly, partisan identification often drives policy preference, rather than vice versa. That party identification exercises control over policy opinions the way that it does suggests that the essence of Arendt’s propaganda/indoctrination distinction–that the presence or absence of group-identity cues makes a difference in how political messaging is taken up by individuals–has currency outside of totalitarian settings. Distilled in this way, Arendt’s distinction in fact resembles one that another prominent theorist of propaganda, Jacques Ellul, drew between pre-propaganda (analogous, confusingly, to Arendt’s notion of propaganda) and propaganda (analogous to her notion of indoctrination), which wasn’t intended to be reserved for totalitarian settings. Accounts of propaganda benefit from distinguishing between the messaging which aims to form new group identifications and that which aims to exploit existing ones, whether or not they restrict their usage of “propaganda” to cover only the former while appending the label “indoctrination” to the latter.

That group-level identity is so formative and indelible a feature of political belief formation returns us to the consideration of the first element of the schematic account of propaganda sketched above: the description of a normative state for political agents. It is fine and good to suggest that rational independence is among the ideal features of a political agent and that propaganda undermines that independence. However a question arises as to what this account must make of the real world, in which true independent-mindedness concerning key political issues is exceedingly scarce. Must it explain this fact by concluding that propaganda is incredibly ubiquitous? Only if we suppose the account to say that all instances of group identity–related bias are the result of propaganda. If, alternatively, it is conceded that propaganda is only one cause of such bias, if we indeed conclude that these biases come into existence organically whenever groups do, then the undermining of independent-mindedness ceases to seem distinctive of propaganda.

Here’s another way of putting this point: political liberalism elevates the role of individual rationality in a way that can at least seem to be in tension with regarding group identity–and the biases that come with it–as conditions which political justice and legitimacy will have to be realized alongside of if they are to be realized at all. This perspective will tend to give rise to a notion of propagandicity tasked with bridging the entire divide between a political psychology benighted by deliberate indoctrination and one free of any irrational compulsions whatsoever. This muddies the account; propaganda comes to be a byword for all the many temptations that lure us from the path of rationality, rather than being recognized as a tool with a particular, circumscribed purpose, the exercise of which is discernible even in an otherwise flawed world.

Arendt’s distinction helps us add nuance to the posited relationship between propaganda and group-identity biases. Propaganda doesn’t create group identities, or the biases that come with them, from scratch. If propaganda, per Arendt, is that which draws the individual into groups united by a shared pattern of deference to a shared authority, we needn’t suppose that the individual was previously bereft of group memberships. Instead, we can imagine individuals’ existing group identities, alongside other reasons for susceptibility to bias, exploited in order to draw them into new identifications. In this way, a campaign of propaganda will often be a project of at least minor political realignment: where one aims to draw in individuals on the basis of their antecedent identities, but aims to do so for a mass of individuals, one ipso facto seeks to alter the group-identity landscape. This may involve packing some existing groups together to create new coalitions, and cracking others to dissolve the political potential they had when unified. The organization which is the characteristic end of propaganda we can then see as reorganization.

Does the foregoing entail the indictment of all political realignments as propagandistic? Whatever the answer to this question, the option remains to deny that this would be problematic; that all propaganda is bad is an entailment of the liberal perspective we have been considering, but it is an assumption that someone discarding other liberal tenets might be happy to discard as well. One response along these lines says that whether campaigns of propaganda and indoctrination are bad depends on whether the identity group that the propaganda organizes individuals into reflects a natural coalition of interests (such as class interests). Of course, others will reject the idea that there is any such thing as an enduring natural coalition of interests, or else that the presence of one would redeem the use of propaganda.

In any case, the picture above of how propaganda and indoctrination work doesn’t require that all attempts to build or dissolve coalitions count as propaganda. Whether or not attempts at political realignment count as propaganda will depend on whether the group into which individuals are organized is capable of indoctrination. Now plenty of group identities exert influence on the beliefs and preferences of those who have them; this is what rendered the essence of the propaganda/indoctrination distinction applicable outside of totalitarian contexts. However, group identities vary with respect to a) their intensity and b) the degree to which they are exploited by group authorities.

It is concerning the intensity of group identification that we have reason to consider the phenomenon of political polarization. Polarization, very generally, is of course political disagreement that is deep, intractable, and systemic. But this general notion can be precisified in terms of the degree of divergence between agents’ beliefs, or else in terms of the degree of antipathy that exists among groups of agents with different political identities. The latter notion of polarization is sometimes referred to as “affective polarization.”

There is some debate among political scientists about whether non-elite Americans’ policy views have in fact come to cluster at increasingly estranged points on the political spectrum over the past several decades, as the truism of America’s increased polarization would suggest. What is clear is that the country’s political parties have, since the 1970s, become sorted in a way they were not before–that is, liberals now go Democratic and conservatives Republican, whereas the parties were previously ideologically mixed. A more rigorously sorted party system does not entail a more ideologically extreme or divided populace, and some have insisted that whatever radicalization might have taken place among the country’s politicians and media outlets, typical Americans are no more ideologically polarized than they have been historically.

The party sorting process, however, has had nontrivial effects on the strength of voters’ party identities, at least as gauged by how deeply they dislike the other party; it is uncontroversial that affective polarization in the United States has increased. Democrats and Republicans have lower opinions of opposing party members’ intelligence and generosity than they did, and are more likely to say that they would be displeased by their child marrying across party lines than they were a generation ago.

Polarization, in strengthening group identities, establishes a reflex of deference to one’s own group’s authorities. Where the authority recognized within that group (a party leader or elite movement figure) comes to rely on this deference on the part of its members, the group becomes one characterized by indoctrination.

In short, then, polarization transfigures the political messaging environment, shifting parties’ and movements’ internal messaging toward indoctrination. And because a group’s internal dynamics involving indoctrination was a condition of identifying this group’s campaigns of realignment as propaganda, polarization likewise shifts groups’ overtures to nonmembers toward the propagandistic.

If polarization makes propaganda and indoctrination worse, then alleviating polarization might make them better. But the suggestion that de-escalating American polarization is a means of defusing programs of propaganda and indoctrination will seem merely to connect one intractable problem to another. However, the picture of group identity dynamics that has emerged in this essay does suggest how we should think of solving the problem of polarization. Mass polarization, like indoctrination, is not susceptible to being reversed by urging members of polarized groups to simply resist the effects of their identifications and “return” to rationality. But, as Arendt suggests, the epistemic effects of a group identity, even where the group in question uses indoctrination to cultivate a fanatical membership, are tied to the group’s persistence. And just as political realignment can work to form such groups, it can also work to dissolve them. The initiation of a political realignment which cracks indoctrinated groups through “big organizing” is the best bet to disrupt the worst of American polarization, and the campaigns of propaganda and indoctrination it enables.

We might close by rejoining the baffled neophyte, described at this essay’s outset, who was sentenced to a day of watching the news. As a device for resetting our consideration of propaganda, we asked what this viewer would conclude about what was distinctive of propaganda. A perspective that I suggested is common, indeed implicit, in much contemporary liberal discussion is that propaganda is political messaging which works to undermine an otherwise present rational virtue. But our observer, watching coverage of twitter-bot-assisted astroturfing, or of must-run monologues disseminated to a broadcasting conglomerate’s many local news acquisitions, might well reach a different conclusion. That the distinctive function of propaganda is to consolidate individuals within identity-based groups and to condition a homogeneity of belief within these groups is suggested by all of the most readily identifiable examples of our moment. Even as she turns from this conclusion to the consideration of those hopeful political vistas still left open, our observer might be forgiven for wincing a moment in regret.

Megan Hyska is assistant professor in philosophy at Northwestern University, where she conducts research on language and politics.
Originally published:
October 1, 2018


Rachel Cusk

The novelist on the “feminine non-state of non-being”
Merve Emre


Renaissance Women

A new book celebrates—and sells short—Shakespeare’s sisters
Catherine Nicholson

Fady Joudah

The poet on how the war in Gaza changed his work
Aria Aber

You Might Also Like

Computational Propaganda

Public relations in a high-tech age
Renée DiResta

Incarcerated Language

Reginald Dwayne Betts


New perspectives, enduring writing. Join a conversation 200 years in the making. Subscribe to our print journal and receive four beautiful issues per year.