By BRIAN LEITER and SAMUEL LEITER
Review of How Propaganda Works, by Jason Stanley
Princeton University Press, 2015
Jason Stanley, who teaches philosophy at Yale University, is well-known among specialists for important technical work in philosophy of language and epistemology. In this new volume, he brings some of these technical tools to bear on issues of moral and political significance related to propaganda, democracy, and inequality.
We must caution the prospective reader at the start, however, that the book is elegantly, but misleadingly, titled. Stanley is not actually concerned with “propaganda” in the senses familiar to most readers. Thus, a standard dictionary says propaganda are “ideas or statements that are often false or exaggerated and that are spread in order to help a cause, a political leader, a government, etc.” Stanley himself acknowledges what he calls “the classical sense” of propaganda—“manipulation of the rational will to close off debate” (48)—and his own account largely incorporates the idea that propaganda influences people primarily through emotional or non-rational means. But Stanley also rejects many of the familiar ideas about propaganda, including: that it is necessarily false or biased; that it is necessarily a bad thing; and that it is a message intentionally spread to promote a political cause.
Consider one of Stanley’s most striking examples of “propaganda,” taken from the 2012 primary campaign, when the journalist Juan Williams, during a public debate, challenged the conservative candidate Newt Gingrich as follows (156):
You recently said, ‘black Americans should demand jobs, not food stamps.” You also said, “poor kids lack a strong work ethic,” and proposed having them work as janitors in their schools. Can’t you see that this is viewed, at a minimum, as insulting to all Americans, but particularly to black Americans?
Gingrich, of course, denied this, offering “a bromide on the value of hard work” which was met with applause by the audience (156). Stanley writes: “Williams followed up by pointing out to Gingrich that expressions such as ‘lacking work ethic’ were associated with negative racial stereotypes…and it was disingenuous for Gingrich to deny them. The audience loudly booed Williams’s response” (156). In other words, “lacking a work ethic” is code for “lazy blacks,” just like “food stamps” or “welfare queens.” In 2012, you don’t need to say, “Negroes are shiftless freeloaders” to tap into the racist ideology of a portion of the audience. You just need to use the right code words.
Stanley says of this example that “[t]his is how propaganda works” (157)—let us call it S-Propaganda—but, in fact, it isn’t how much (maybe most) propaganda works, a point to which we return, since many (maybe most) cases of propaganda do not involve code words and do involve clear falsehoods promulgated to achieve political goals. S-Propaganda is distinctive--perhaps idiosyncratic as propaganda goes--but the particular phenomenon it picks out is of interest and real, as the Gingrich anecdote nicely illustrates. Readers will be rewarded with real insight (and a clear explanation of some technical philosophy) if they approach the book as it might have been more accurately titled, namely, How “Code Words” Work to Undermine Liberal Democracies.
So how do they work? Chapter 4 is the highlight of the book deploying linguistics, formal semantics, and pragmatics (the study of how context affects meaning) to explicate the pernicious mechanism of rhetoric like Gingrich’s. “Participants in a conversation begin with certain information in common, or presumed to be in common, and it is that body of information that the speech they perform are designed to influence” (131, quoting the philosopher Robert Stalnaker). This “common ground” can be altered as a conversation progresses, though it can be altered in one of two ways, either by explicit attempts to change the “at-issue content” of the “common ground” through assertion or by implicitly changing the “not-at issue content” of the common ground by simply presupposing such a change (134-135).
Consider a conversation about whether a particular problem was solved and, if so, who solved it. The assertion, “It was John who solved the problem” asserts that it should be “common ground” of the conversation that John solved it, and that is now put forward for discussion. But the assertion also “presupposes the proposition that someone solved the problem” (135) which, if left unchallenged, now makes it part of the “common ground” of the ensuing conversation that the problem really was solved. Understand that the claim here is an empirical one about how pragmatic presuppositions function in ordinary discourse. And the worry is that “presupposition can be used to smuggle in content that one would not necessarily accept if it was presented as the content asserted” (138).
Now back to Gingrich who says “black Americans should demand jobs, not food stamps.” The asserted at-issue content is about what Blacks should demand, but the presupposition is that Blacks are people who need money, but prefer welfare to work. Because the presupposition was not part of the assertion—after all, Gingrich didn’t say, “You know Blacks are lazy, impoverished freeloaders, but they really should ask for work instead of food stamps”—it is easy for him to deny the “not-at issue content,” as Gingrich did when confronted by Williams (156).
Although Stanley’s analysis of the linguistic pragmatics of “code words” like Gingrich’s is clever and illuminating, much propaganda, including lots of racist propaganda, does not operate that way. During the current campaign for the Republican nomination for President in the U.S., Donald Trump, the celebrity heir to a real estate fortune, has made “illegal immigration” a major issue, asserting that the Mexican Government is “forcing their most unwanted people” into the U.S., “criminals, drug dealers, rapists.” This sounds like a paradigmatic case of propaganda, yet its harms do not derive from any pragmatic presuppositions: Trump asserts explicitly, and falsely, that illegal Mexican immigrants are largely “criminals, drug dealers, rapists,” even though illegal immigrants from Mexico are disproportionately hard-working and law-abiding, for the obvious reasons (e.g., illegal residents are vulnerable to deportation as soon as they come to the attention of law enforcement). Yet Trump’s false assertions appear to have been effective in mobilizing political support among a portion of the white population.
Trump’s kind of propaganda is hardly anomalous, and it fits the dictionary definition we began with better than S-Propaganda. But S-Propaganda has further wrinkles that make it even more distinctive. Propaganda, for Stanley “is a kind of speech that fundamentally involves political, economic, aesthetic, or rational ideals, mobilized for a political purpose” (52). Propaganda is not a bad thing when it exemplifies what he calls “Supporting Propaganda”--“[a] contribution to public discourse that is presented as the embodiment of certain ideals, yet is of a kind that tends to increase the realization of those very ideals by either emotional or other nonrational means”(53)—on behalf of worthy political ideals.
Propaganda in its more familiar, pejorative sense is either that which supports bad political ideals or that “undermines” worthy ones--the latter being (unsurprisingly) “Undermining Propaganda,” which “is presented as an embodiment of certain ideals, yet is of a kind that tends to erode those very ideals” (53). Undermining S-Propaganda operates by “undermin[ing] a political ideal by using it to communicate a message that is inconsistent with it” and that in so doing “exploit[s] already existing flawed ideological belief,” the latter performing the function of “mask[ing] the contradictions of undermining propaganda” (57). The crucial mechanism of bad propaganda, then, is that it invokes an ideal it actually undermines, where this “contradiction” is masked by “flawed ideological beliefs” which “erect difficult obstacles to recognizing tendencies of goals to misalign with certain ideals” (57). To take Stanley’s best example (60), think of those paid to create the misleading impression that there is “uncertainty” about the role of human contributions to global warming who cloak themselves in the ideal of scientific objectivity and the need to consider “all points of view.” We will call this the “Contradiction Mechanism,” a mechanism that depends for its success on a background “flawed” ideology.
For Stanley, the “distinctive” feature of ideological beliefs is that they are resistant to rational revision in light of evidence. He says, implausibly, that this “is the philosophical puzzle at the heart of…David Hume’s philosophy” (178). Hume did not, of course, write about a problem of “ideology,” but that is minor. The real difficulty is that while Hume clearly believed that our beliefs were not rationally justified, he famously observed that skeptical arguments to that effect “admit of no answer but produce no conviction.” He did not think this puzzling, since he did not think human beliefs were fundamentally the product of reason. One might also worry that on Stanley’s account, phobias and the beliefs that produce compulsive behavior are also “distinctive” in being immune to rational revision and thus also “ideologies.” Perhaps it would have been better to claim that an important, but not distinctive, feature of “ideology” in Stanley’s sense (call it S-Ideology) is that it is highly resistant to rational revision in light of evidence.
Ideologies, says Stanley, consist of “beliefs that unreflectively guide our path through the social world” (184), and so in this descriptive sense, “everyone in the world has an ideology” (184). Because S-Ideologies are typically connected to our social identities and allied social practices, they are largely immune to rational revision (185): if we give them up, we have to give up part of our sense of who we are and our place in the social world. Flawed S-Ideologies, says Stanley, arise under conditions of material inequality (199), but their flaw is an epistemic one: they impede knowledge of important aspects of the social world that. Those with wealth, for example, tend to think they deserve it (224-225); as Stanley amusingly notes, even lottery winners display “a rise in their belief in the justice of the current wealth distribution of society” (227). But this is a “flawed” ideology that obscures from view the role of other non-desert-based factors that produce particular wealth distributions.
Ideologies block knowledge of the social world through various mechanisms. One particularly striking way is when an ideology fails to provide citizens with the concepts they would need to understand the aspects of the social world they encounter. Stanley, borrowing an example from the English philosopher Miranda Fricker, points to how victims of sexual harassment at an American university several decades ago were not able to articulate the wrong they suffered because they did not have the concept of “sexual harassment” (202). More pernicious, of course, are cases of intentional deprivation of relevant concepts. The Chinese government in 2013 required university professors not to discuss with students such concepts as “universal values,” “crony capitalism,” or judicial “independence. As Stanley observes: “This is a clear attempt to ensure that students lack crucial political concepts, precisely the ones possession of which would enable them to critique Chinese government policy” (203). Without the concepts to articulate experience, an ideology can deprive the subject of that experience knowledge of it.
In the final chapters of his book, Stanley argues that in “conditions of significant inequality…those who benefit, the elites, will have certain perceived epistemic advantages over the negatively privileged group” (267) and they “will employ their epistemic and practical advantage to claim expertise over issues of value, which are in fact beyond the domain of expertise” (268). He here deploys recent technical work in epistemology that largely seems otiose with respect to the real point, the one that Marx made a century and a half ago (and Stanley alludes to early on: cf. 184): those with wealth also own or influence the major media by which information is communicated, and, unsurprisingly, they communicate information in a way that suits their economic interests. But Stanley is surely right to observe that “spontaneous belief formation is in general an involuntary process. We acquire beliefs, spontaneously from testimony of authority figures, from the lack of reliable sources that contradict them” (236). When billionaire Rupert Murdoch controls the news, he makes sure the only “authoritative” testimony we hear is from those friendly to Rupert Murdoch’s privileges.
Authors may, of course, stipulate what their terms mean, but some theoretical benefit should be in the offing, apart from the fact that their preferred sense permits deployment of their preferred philosophical tools. This worry looms over large portions of Stanley’s book. Consider: non-pejorative, supportive S-Propaganda sounds a lot like what everyone else might have called “rhetoric” on behalf of morally worthwhile objectives. Stanley, for example, appeals to Du Bois’s use of “civic rhetoric directed at a white audience” to “undermine an understanding” of democratic ideals that is limited only to whites (116). What is the theoretical advantage in understanding this as an instance of propaganda at all? Even more surprisingly, Stanley calls the 1965 Selma March “a paradigm case of democratically acceptable propaganda” (113) because it involved “manipulation of the media to draw attention and empathy to the predicament of an otherwise invisible group” (113). But the so-called “manipulation” at issue involved a peaceful march that was met with violent resistance, revealing (accurately it would seem) the brutality of Southern segregationism. Certainly the Selma March was fraught with meaning, but why is it helpful to think of it as “propaganda”? We are not sure.
Also puzzling is why Stanley limits his attention to propaganda that involves the Contradiction Mechanism. Lots of familiar instances of propaganda in democratic societies seem much closer to the “ordinary” definitions with which we began than with S-Propaganda—recall Trump’s racist remarks about illegal immigrants from Mexico. Or take a different example: much of the propaganda that mobilized support in America for the 2003 invasion of Iraq—an example Stanley mentions at various points—involved simple falsehoods or misleading statements (e.g., Iraq had weapons of mass destruction and was a threat to the security of the U.S.), not the Contradiction Mechanism.
We worry that S-Propaganda is not only underinclusive with respect to how propaganda operates in democratic societies, but also surprisingly overinclusive. For example, Stanley claims the U.S. Supreme Court opinion in Citizens United v. Federal Election Commission is a core example of “undermining” propaganda (61) because,
The Court presented the decision as if extending free speech and other rights to corporations…as an embodiment of the principles of democracy. Yet the unlimited corporate donations that Citizens United gave rise to are themselves an existential threat to democracy, promising to hand over the mechanism of government to corporations that do the bulk of funding for political campaigns. (61)
Corporations do not, in fact, do most of the funding of political campaigns, though wealthy people do. The Supreme Court’s opinion in Citizens United has seemed wrongheaded to lots of commentators, but is it helpful to think of it as “propaganda.”? Surely it could be used as propaganda by politicians advocating for no limits on campaign contributions, but the opinion itself, which hardly anyone reads, does not seem to be “propaganda,” except by stipulation. The opinion may be wrong, badly reasoned, naïve, and so on, but it seems to us eccentric and confusing to assimilate it to a model of “propaganda.”
What about Stanley’s claim that S-Ideologies are connected to our “social identities”? The empirical support for this claim is mixed, so how seriously should we take it? Sometimes we are correct to be skeptical of putative counter-evidence and not because we are in the grips of an ideology tied to our identity. We take it Stanley does not want to deny this (cf. 198-199). Yet Stanley also claims that his own “belief that the theory of evolution is a correct description of reality is connected tightly to my self-conception as…a rational, cosmopolitan intellectual who trusts certain sources of evidence over others” (76). Really? We believe that the theory of evolution is a correct description of reality because we have good reason to think that biologists, who believe this, are reliable epistemic authorities (i.e., sources of knowledge). That we believe that might be taken as evidence that we are “rational, cosmopolitan intellectual[s],” but that we may think of ourselves as such people is not an explanation of why we believe it.
S-Ideology shares with the more familiar Marxian sense of ideology that it is often difficult to get people to abandon it, but it differs in being so capacious as to encompass any beliefs resistant to revision, regardless of their content. (The Marxian tradition, by contrast, was always concerned with beliefs that involved mistakes about the interests or well-being of those who held them.) But S-Ideology differs from Marx’s conception (if not, perhaps, Horkheimer’s) in deeming ideological, in a pejorative sense, views that “exclude…narrative claims about personal experience as reasons for action or belief. The ideology of technicism is one that restricts genuine reasons in the public sphere to those whose contents contain only scientific or quantitative concepts” (208-9). One might worry that reports of “personal experience” are often notoriously unreliable, often tainted, as even Marx understood 150 years ago, by ideological delusions. (And one might worry that propagandists who deny climate change are also fond of dismissing the ideology of “technicism”!) Marx, famously, did not think genuine science was ideological in a pejorative sense, and it seems a virtue in a theory of ideology that it should not subsume theories whose epistemic status is robust, not dubious. One can nonetheless agree with Stanley’s emphasis on ways in which genuine bias can infect even empirical claims based on perception (cf. 211-214), but that point does not require us to treat doubts about “personal narrative” as ideologically suspect. Stanley is driven to do so, we suspect, because of his general worry that propaganda in democracies is often “employed to make the masses feel unqualified to weigh in on the central decisions about their lives” (40). But even from a Marxian point of view, the masses are often “unqualified” because they are in the grips of other ideological delusions. That worry does not receive adequate attention on Stanley’s account.
Readers who understand that this is a book about S-Propaganda and S-Ideology will find the analysis often compelling and, in the opinion of the non-specialist co-reviewer, generally intelligible to the educated reader. (Both reviewers agree, however, that the argumentative thread of the book is sometimes disjointed.) This is no small achievement given the difficulty of some of the technical philosophy Stanley wants to deploy. At the same time, it is clear that S-Propaganda and S-Ideology are interesting but somewhat idiosyncratic concepts. If there is a flaw in Stanley’s approach, we suspect it is an instance of the old saw that if you have a hammer, everything looks like a nail. Stanley is an expert wielder of his technical philosophical hammer against the nails it works on, but a lot of what is most vexing about propaganda and ideology in capitalist democracies demands different tools and different treatments.
Posted 12 October 2015
BRIAN LEITER is Karl N. Llewellyn Professor of Jurisprudence at the University of Chicago, and author, most recently, of Why Tolerate Religion? (Princeton, 2013) and Nietzsche on Morality, 2nd edition (Routledge, 2015). SAMUEL LEITER is an undergraduate at the University of Chicago currently studying the history and politics of East Asia, especially Japan.