Simon Bridges recently stated that “people on the average wage shouldn’t be paying almost 33 per cent in the dollar”. I wonder whether this is a “double secret dog whistle”: a statement designed to provoke smug responses that engage with the literal claim about tax rates, but fail to engage with the intended message that people on the average wage should be paying less tax.

In September 2016, Salena Zito wrote that the press took then-Presidential candidate Trump’s statements “literally, but not seriously” while his supporters took him “seriously, but not literally”. Zito was writing in particular about the response to Trump’s claim that “Fifty-eight percent of black youth cannot get a job, cannot work.” With respect to the accuracy of this claim, Zito states:

 It’s a claim that drives fact-checkers to distraction. The Bureau of Labor Statistics puts the unemployment rate for blacks between the ages of 16 and 24 at 20.6 percent. Trump prefers to use its employment-population ratio, a figure that shows only 41.5 percent of blacks in that age bracket are working. But that means he includes full time high-school and college students among the jobless.

 The point Zito makes it that, while the press (and the same can be said for Trump’s political opponents more generally) fixated on testing the actual claims Trump made, Trump’s supporters were less concerned about the literal factual claims, but the message that they took from Trump’s statements: “I care about bringing back jobs for black youth”.

 In spending time wrangling over the possible ways Trump’s statement could be interpreted, in case there was a possible charitable interpretation that made his statement accurate, and setting out in detail precisely why he was wrong, Trump’s political opponents ended up doing two things.

 First, they failed to engage with the message that Trump’s supporters (or potential supporters) were taking from Trump’s speech – that he cared about bringing back jobs for black youth.

 Second, by responding through extensive fact-checking, grammar and spelling nitpicking, and stating or implying that Trump was an idiot for getting facts or details wrong (here’s a recent example), they lent support to the narrative that Trump’s opponents were part of a “liberal elite” who looked down on those who were less educated and well-spoken. Which is counter-productive if the aim was to decrease Trump’s support. Not everybody cares about grammer, especially those with more pressing concerns like unemployment and housing, a demographic that Trump was trying to court.

On campaign, Trump was also fond of saying that he would “build a wall” between the USA and Mexico, and implement a “muslim ban”. The seriously/literally thing applies here also. While Trump’s opponents would react with outrage to the idea of actually doing these things, and question the legalities and practicalities, the take-away message for Trump’s supporters was something quite different. Echoing Zito, Peter Thiel stated:

I think one thing that should be distinguished here is that the media is always taking Trump literally. It never takes him seriously, but it always takes him literally. ... I think a lot of voters who vote for Trump take Trump seriously but not literally, so when they hear things like the Muslim comment or the wall comment, their question is not, ‘Are you going to build a wall like the Great Wall of China?’ or, you know, ‘How exactly are you going to enforce these tests?’ What they hear is we’re going to have a saner, more sensible immigration policy.

 Closer to home, I’m reminded of the “seriously but not literally” discussion when it comes to leader of the opposition Simon Bridges’s recent statement that:

People on the average wage shouldn’t be paying almost 33 per cent in the dollar.

 Of course, people on the average wage aren’t paying almost 33 per cent income* tax in the dollar, as various commentators and political opponents were quick to point out, providing extensive explanations breaking down how our progressive income tax system works. Responses also could be taken as conveying the message that Bridges was a bit of an idiot for not appreciating this (“Good news for Simon Bridges: his big tax idea is already happening” and “Well that’s all good then, because they’re NOT!” respectively). 

Which reminded me of the whole literally/seriously thing. And I think these sorts of responses run the same risk of failing to engage with the message that Bridges’s supporters (or potential supporters) might take away from these statements – which must be something like “people on the average wage should be paying less tax”. Further, they risk coming across and smug, elitist and nitpicky.

Now you might look at his CV and think there are at least some reasons for thinking that Simon Bridges is, at least in some respects, a smart guy and not so much an idiot. So it might seem a bit odd for him to be that out of touch with what the average wage is, and how income tax works. And you might think there’s something else going on here.

You may have heard the term “dog whistle politics”. It’s a term for political speech that uses language that is taken to mean one thing to the general population, but simultaneously sends a different signal to a targeted smaller group. The term is usually used in allegations that a political actor is using coded language to court support from a sub-group that the general population might be unimpressed to see them seeking support from.

Some examples:

•  References to “family values” to seek support from Christians without referring to overtly Christian values, which might put off others;

• New Zealand First’s 2019 proposal that migrants respect “Kiwi values” was labelled as dog whistle politics” i.e. seeking to court support from the far right while using language that does not seem obviously unobjectionable;

• National’s 2019 proposals including fining parents of truants and reducing the numbers of beneficiaries were labelled as dog whistle politics: i.e. seeking to court racist/misogynist/anti-poor voters without saying so explicitly; 

• In 2017, Labour and the Greens were accused of whistling to racist voters by blaming immigration for unemployment and Auckland house price, and calling migrants “low-skilled”.

 The thing about dog whistle politics is that it only works If the general population believe that the meaning they take is the intended one, and the dog whistler wasn’t really trying to get support from those racists (or whoever the sub-group is). That means that political actors tend not to put their hands up and say “Yep, you got us, we were trying to be sneaky but you saw through our trick.” So, while dog whistle allegations get thrown around a lot, political actors tend to deny dog whistling, and it can be difficult to prove what someone really intended by their speech. 

The dog whistle analogy is based on the idea of a “dog whistle” that produces a sound that dogs can detect but humans cannot. So if you blow a dog whistle then you might get an excited reaction from the dogs in the room whereas the humans are unsure what all the fuss is about. The humans, who can’t see or hear anything wrong, might then think that the dogs are barking up the wrong tree, and dismiss their frantic warnings. This post is about trying to articulate a variant on dog whistling in political speech that’s intended to create that scenario, which I call the “double secret dog whistle”. 

If you google the phrase “double secret dog whistle”, you’ll see it’s been used before, generally for hyperbolic effect. What I have in mind, though, is something different. In a regular dog whistle, the secret signal is targeted at a potentially sympathetic group. The whistler either intends that the sub-group appreciate that they’re being sent a secret signal that the general population won’t receive, or doesn’t mind if they think so. In a double secret dog whistle, the sub-group is not sympathetic to the whistler and the whistler doesn’t intend them to realise that they’re being manipulated into a particular reaction.

 It’s not exactly news at this point that this sort of thing is part of the right’s playbook. Around this time last year, Danyl McLauchlan suggested that National’s “all sizzle no sausage” ad was what I’d call a double secret dog whistle:

For progressives the ad is an offensive failure because it’s an egregious instance of mansplaining. It’s men explaining maths and politics and economics to a wide-eyed woman! National supporters feel that the ad succeeds because the target audience for the ad is not outraged progressives who would rather die than vote National; instead they’re communicating to demographics who won’t be offended by mansplaining but who might be persuaded by the critique of KiwiBuild.

Here’s my grand conspiracy theory. Progressives are actually the primary target for this ad and it is designed to offend them. Offense and controversy makes things newsworthy and earns you coverage in the mainstream media, thus potentially reaching a far greater number of viewers than National would get through making a non-controversial, non-mansplaining ad. The way you communicate the KiwiBuild critique to the wider public – who are never going to watch a political ad in their feed, even if you boost it – is by breaching progressive rules of etiquette and provoking a controversy. This is Trump’s great innovation in political marketing: you don’t need to pay for advertising you just repeatedly outrage progressives, especially those who work in the media, and they’ll give you all the free coverage you could hope for.

I don’t think a double secret dog whistle is just about getting more publicity and, in this case, communicating the critique of Kiwibuild to the wider public. A double secret dog whistle succeeds if the general public and the target group take something different from the message. If you pull that off that the general public think that the targeted group’s reaction is akin to dogs barking at nothing. Not only have you got more publicity you’ve made your political opponents look foolish in the process. If the general public agreesthat the ad is sexist, for example, then it’s counterproductive: it loses you votes, and the public might even be grateful to those noisy progressives for drawing their attention to it. 

One way to set up a double secret dog whistle is to put out content that you know is going to be criticised by progressives as racist or misogynist or something like that. Another is to put out content that, in its form, defies elite conventions of grammar, spelling, or even font use – like the Conservative Party’s tweet in all caps comic sans proclaiming that “MPs MUST COME TOGETHER AND GET BREXIT DONE”. To potential supporters, this is intended to convey the literal message, but the tweet was also intended to – and did – provoke a response that ridiculed the form but failed to engage with the substance.

Digital marketing firm Topham Guerin, founded by Young Nationals, worked on the Conservative campaign, and have also worked for the NZ National Party and the Liberal Party in Australia. So we should not be too surprised to see them attempting to work their magic here, leading Chris Trotter to wonder whether Labour’s “communications strategists can match the clear-headed and disarmingly amoral cynicism of Topham-Guerin and their imitators”.

As I said earlier, one of the problems with labelling an act of speech a dog whistle is that politicians don’t admit (at least, not publicly) that that’s what they’re doing. It’s easy to accuse but hard to prove. When you include double secret dog whistles, it’s tempting to see them all over the place. Is “National hates gangs” designed to be taken literally by most voters, but provoke accusations of racism from others? Should the official Trump re-election campaign Twitter’s post, superimposing Trump’s face onto Marvel villain Thanos in a pivotal moment of “Infinity War”, be taken literally as meaning that his supporters see him as “a genocidal warlord hell bent on destroying half of existence in the universe”, or is a less literal take a more reasonable one? Was Simon Bridges’s accent developed in a lab to invite mockery of his broad Kiwi pronunciation while simultaneously appealing to “everyday New Zealanders”, whoever they might be?

While I’m sure that double secret dog whistles are definitely a thing, I don’t think it’s easy to stick a claim that any particular communication is one. Indeed, spending a lot of time and effort arguing that a particular comment is a double secret dog whistle potentially falls into exactly the trap the whistler intends: engaging with the message that you took but not the one others will be hearing. 

So, am I just barking at the moon here, or can I say something useful about engaging with double secret dog whistles? First, I think that knowing that they exist as a strategy is useful. Second, engaging with the message intended for the more general audience is probably more important than engaging with the literal meaning, at least when it comes to communications that are intended for, or going to be available to, the general public. Third, it might be possible to engage with both at the same time. It might be possible to accept Bridges’s narrative that people on the average wage should have money in their pockets while providing a different solution (admittedly somewhat difficult when all we know about National’s plan is “tax cuts”), while also pointing out that Bridges’s description is a bit wonky. Or, to argue that instead of directly giving people more money, it’s better to reduce the costs of housing, medical care etc. Finally, although the most comprehensive way to defeat a double secret dog whistle would be to convince the general public that the communication in question is exactly that. But, this is a risky play because, if you fail, the double secret dog whistle has worked exactly as planned.

* Perhaps once you factor in other taxes, the overall tax is almost 33 cents in the dollar. But when we talk about tax on the dollar of income, we’re generally talking about income tax.