In the first 2020 U.S. Presidential debate, Donald Trump said the following:
“These cities don’t want to talk, like you [Joe Biden] about law and order … if he [Biden] got to run this country … our suburbs would be gone.”
On its face, Trump’s juxtaposition of Biden’s not talking about “law and order” with the end of the suburbs seems to be a non-sequitur. To understand what is happening, we must appreciate the context of these remarks: the debate took place during an extended period of protest and civil unrest throughout the United States in response to police killings of Black Americans. With that context in mind, Trump’s implicit message comes into focus: if these (largely BIPOC-led) protests are not quashed (i.e. no “law and order”), they will eventually spill over from urban cores to the suburbs, where they will threaten the lives and property of White Americans (the presaged “end of the suburbs”).
By raising fears of violence against middle class White Americans, Trump conveys something to the effect of: I will protect White Americans from minority violence, while Joe Biden will not.
Yet, crucially, Trump never says any of this explicitly, instead relying on coded speech to convey this message. I will be exploring three questions: Why do people speak in code? How does coded speech work? What should be done about coded speech?
Why speak in code? Suppose you want to communicate that it is raining to your friend. You might say so explicitly,
(1) It’s raining.
Alternatively, you might do so indirectly, perhaps by saying,
(2) I regret not wearing my waterproof shoes today.
The first strategy has the benefit of directness – you’re not liable to be misunderstood. The second strategy lacks this, since your friend might not pick up on your communicative intention. But the second strategy has the benefit of deniability. Suppose it turns out not to be raining. Then, if you had said (1) you would have been under pressure to retract this claim, and your friend might reasonably complain that you lied. But if you had said (2) instead, you could plausibly deny that you meant that it was raining – for instance, maybe you regretted not wearing your waterproof shoes because you had to walk through a swamp.
Why is gaining some degree of plausible deniability over what we mean an advantage of speaking in code? To answer this, I will draw on the work of Tali Mendelberg (in particular, her book, The Race Card). According to Mendelberg, throughout much of contemporary U.S. society, there are social norms against racism, among them norms against the promotion of racist attitudes. Evidence of the existence of such norms comes from the fact that people have been fired from their jobs for saying the N-word, and some have lost friends over support for Donald Trump.
Suppose you’re a politician who wants to appeal to an electorate made up of individuals with racist attitudes as well as individuals who endorse norms against the promotion of such attitudes (note that there is an overlap between these groups). To attain broad appeal, you should try to signal your allegiance to the first group without being punished by those in the second group. Using a code word provides a strategy: you can express the racist attitude implicitly, leaving yourself plausible deniability for having violated the norm, thereby making it harder to hold you accountable to it.
This seems to be what was happening when Paul Ryan said,
We have got this tailspin of culture, in our inner cities in particular, of men not working and just generations of men not even thinking about working or learning the value and the culture of work, and so there is a real culture problem here that has to be dealt with.
Here, Ryan uses “inner city” as a code for “Black” to promote the racist attitude that poor Black Americans are poor because they are lazy. Yet, by not saying this explicitly, he can disavow having violated any norm of racial equality. In fact, this is exactly what happened. Congresswoman Barbara Lee called out Ryan for his statement, saying,
My colleague Congressman Ryan’s comments about “inner city” poverty are a thinly veiled racial attack and cannot be tolerated. Let’s be clear, when Mr. Ryan says “inner city,” when he says, “culture,” these are simply code words for what he really means: “black.”
In response, Ryan said,
This isn’t a race-based comment; it’s a breakdown of families, it’s rural poverty in rural areas, and talking about where poverty exists – there are no jobs and we have a breakdown of the family. This has nothing to do with race.
Whether we are persuaded by Ryan’s response is beside the point – there is space for him to make such a move without contradicting himself, and this allows his supporters to tell themselves and others that his comment didn’t violate any norm of racial equality (and thus that, while of course they would never support a racist, Paul Ryan isn’t a racist, thankfully).
How does coded speech work? We often draw inferences from what people say explicitly. So, for instance, if I tell you that John went out in the rain without an umbrella, you’ll probably infer that he got soaking wet, even if I didn’t say so explicitly. Savvy speakers can exploit knowledge of what their audience will infer from what they say explicitly to convey implicit information. Schematically, we have:
Speaker says P explicitly
Background belief: if P then Q
Audience infers Q
Here’s a simple example:
Explicit: “John went outside without an umbrella.”
Background: “Anyone who goes out in the rain without an umbrella will get wet.”
Inference: “John got wet.”
This structure sheds light on why implicitly conveyed information is deniable. When a speaker merely invites (or allows) her audience to infer something from what she says, she does not (explicitly) commit herself to that information, and thus, she can coherently deny it by clarifying her commitment to the background bridge principle used to generate it.
In the case of the code word “inner city” the speaker may exploit background stereotypes about the racial makeup of poor urban cores as follows:
Explicit: “There is a culture problem of laziness in the inner city.”
Background: “Inner cities are mostly home to African Americans.”
Inference: “There is a culture problem of laziness among African
If members of Ryan’s audience do draw this inference from what was said explicitly, then Ryan’s explicit statement violates norms against the promotion of racist attitudes. Insofar as we think this outcome is likely, we may be justified in holding Ryan accountable to the norm. But if we think this outcome is unlikely, then we would be justified in not holding Ryan accountable here.
Importantly, there may be disagreement among Ryan’s audience about exactly this point. And when people recognise such disagreement, they will be less inclined to hold Ryan accountable to the norm, even if they believe that he did in fact violate it.
The reason for this is that punishing norm violations carries a cost to the violation-punisher – she risks herself being punished by others (even those who also endorse the norm) who perceive her to be inappropriately enforcing the norm. Thus, if she knows there is disagreement about whether Ryan’s speech did violate the norm, she might not punish him on the grounds that doing so puts her at risk of punishment. As a result, implicit violations of the norm go unpunished.
What should be done about coded speech? Coded speech provides a way for individuals to subvert social norms – violating them without being held accountable for doing so. In light of this, how might we respond?
One strategy is to call out the violation, thus making it less implicit (as Congresswoman Lee did to Paul Ryan’s “inner city” comments). This strategy has the benefit of inhibiting particular instances of coded speech – for example, calling out uses of “inner city” as code for “Black” has raised awareness of the code, thus having the overall effect of making it harder for people to get away with using it as a code word.
However, even if we succeed in bringing awareness to the use of certain expressions as code, others can be easily substituted in their place. The contextual inferences exploited by coded speech are easy to come by, and linguistic innovations can take root quickly (see, for instance, the rapid growth and propagation of alt-right terminology throughout Reddit). It’s thus unclear whether the first strategy will actually achieve the end it aims for, namely, stopping implicit norm violations.
Another strategy is to adjust our norms, so that they include what were previously implicit violations as explicit ones (thus leaving the perpetrator no room for plausible deniability). This strategy runs into a different problem. Remember that a major reason people aim to implicitly violate these social norms is to appeal to individuals who resent them. More stringently enforcing such norms runs the risk of increasing opposition towards them, not just among the norm-resentful, but also among those who endorse the original norm but worry more about punishing non-violations than about not punishing actual violations.
To this, one might say, So what? Social norms do not exist to appease the sensibilities of those who don’t endorse them! This is right, of course. But the problem is that broadening the scope of resentment towards a social norm threatens to dissolve it. Social norms are sustained by the dispositions of people to enforce them. If there is no significant endorsement of the norm in the population, or only among members of the population who lack the relevant authority or power to enforce it, the norm will cease to exist. More stringently enforcing a social norm may increase resentment to it, thus leading to its dissolution.
Are we then in a pickle? The most natural responses to coded speech either won’t secure the result we want, or may contribute to the dissolution of the social norms we aim to promote.
To see the way forward, we need to distinguish two social goods that a social norm against racism secures. By raising the social costs of promoting racist attitudes, the norm contributes to making such behaviour less prevalent in the society – that’s a social good. However, as we’ve seen, coded speech mitigates the norm’s ability to bring about this social good.
A second kind of social good secured by such norms is public assurance of the dignity of members of marginalised racial groups. By “dignity” I mean Jeremy Waldron’s notion of “a person’s social standing, the fundamentals of basic reputation that entitle them to be treated as equals in the ordinary operations of society” (discussed in his book The Harm in Hate Speech). Norms punishing the promotion of racist attitudes provide some degree of assurance to members of groups most affected by such behaviour that their dignity is valued and important too. And this kind of assurance is crucial for the enfranchisement of those individuals, signalling that their voices will be heard and accounted for.
Such assurance of dignity is secured by a social norm even if that norm can be easily circumvented by coded speech. We can see this by the fact that, to avoid punishment, a person aiming to express racist attitudes must do so implicitly rather than explicitly. However, such norms are threatened by repeated overt and unpunished violations, such as those perpetrated by Donald Trump leading up to and during his Presidency. Unpunished overt norm violations are evidence that there is not sufficient endorsement of the norm to sustain it.
Thus, caution is my suggestion. Methods of subverting social norms (such as coded speech) will arise inevitably. We should be vigilant in calling out violations of norms against the promotion of racist attitudes, thus reinforcing commitment to such norms, which, as we’ve seen, are expressions of assurance of dignity to marginalised groups. However, we should not broaden our social norms to encompass implicit violations, because of the risk of exacerbating resentment of the norms and thus threatening their existence. We have to chart a narrow path between sustaining social norms (which requires buy in from a wide range of individuals) and enforcing them judiciously in order to ensure that they continue to provide assurance of dignity to those individuals most in need of it.