Trump and the truth: Insights from pragmatic linguistic

Trump and dog whistling

As the elections are upon us, it is necessary to ask ourselves who we want to be as a society and how we want our leaders to run the governments that are supposedly meant to represent us.

This is necessary, not only to make future choices, but also to understand the impact of the past four years on our social fabric.

Trump has made a flurry of statements over the four years that he has been president (and before) that were touted as sexist or racist. Some of his statements were not difficult to interpret, using language that is always used with a sexist and racist mindset. But somehow, this language is always seen as an exception, or unrepresentative of his actual views.

Other of his statements have been the object of some controversy. We call such statements dog whistling, because only a portion of the population is meant to hear its intent without a doubt.

Being able to assign the proper intention to language helps us infer what the mindset of the speaker is, and it is an action we all do at some level when we speak. Some speakers have influence over us, and their intentions hold sway over our actions. They represent what we think our group stands for, and thus how we should practice our group ideology. Unpacking why some of Trump’s statements have been understood one way, while argued to mean another helps us both predict the consequences of his discourse, and arms us against the obvious defense of the intractable intentionality behind words.

Racism, sexism, and the broader context of society

Our society is by and large built on sexist, racist, elitist and ableist ideologies. We cannot escape this reality, and our discourses are touched by semantics that draw their power from these roots. Our insults carry weight because we are aware of the valence around certain categories in social groups. For instance, the n word carries weight because for a long time (and still today), white groups oppressed black people through social force. Similarly, being called a “whore” only carries weight in the context of gendered violence and hierarchy. For instance, being called a jock is not as much an insult as being called a nerd because the high-school microcosm highly values one category over the other.

But we don’t have to use insults to harken to these semantic realities. To understand this phenomenon, let’s look at speech acts more closely.

Speech acts

Austin (1975) offers us three types of speech acts. To him, speaking entails a speaker, which, through language enacts an intended (or unintended) change in the world. He thus defines the locutionary act as the act of saying something, the illocutionary act , which is the intended change by the speaker, and the perlocutionary act, which is the actual effect on the world by the fact that the speaker spoke.

Thus we can see that words themselves carry more meaning than simply what the individual words, outside their spoken context could mean. When taking into consideration a speech act, one must ask themselves first about these three parts of semantics.

The issue with this classification is that one could say that all speech acts enact a change in the world, and that intent is thus unimportant.
To this, Searle (1980) answers that we can break down the intent behind speech acts in different types into 5 categories. He discriminates the illocutionary act between the assertive act, the directive act, the commissive act, the expressive act and the declarative act. The assertive act engages the speaker to the truth of their proposition. I.e, such a statement can be proven wrong and asserts a statement as true.

The directive act intends to get the audience to do something, and this often takes the form of orders or advice.

The commissive act entails a future action by the speaker, like a promise.

Expressive acts express a personal position by the speaker, like an opinion, or even expressing gratefulness and congratulations.

And finally declarative acts which intend to change reality itself, such as baptisms or marriage.

We can already see which areas are blurred by Trump’s dog whistling. In terms of assertive acts, it has been very difficult to hold Trump to a specific position. Any time he has made such statements, even when proven wrong with evidence, Trump is never held to the truth in his propositions. This is a dangerous principle to erode. Being able to infer that our interlocutor speaks the truth enables us to cooperate easily. It helps us infer the next proper action bu using the group as extended cognition.

This erosion of the assertive illocutionary act also erodes our capacity to hold him to other types of illocutionary acts, and hold him responsible for his intents. Since nothing can be proven true, and lack of truth seems to hold no social weight on him, it is always possible for Trump and his associates to annul the semantics of the statements he has made, or its inferred intent, but we will get back to this point

Similarly, when it comes to intent, positions of authority blur the lines between illocutionary acts. As many people will be influenced on an epistemic level by positions of authority, many others will also consider that authority should guide our actions. That is, after all, the whole point of authority structures. Authority without practical impact is not authority at all. In this case, directive acts and declarative acts blur into one, as statements instantly become performative, by virtue of holding a position of authority. In this manner, expressive acts and assertive acts become different layers of influence, that trickle down into action because the individual holds a position of authority. Hence if the president of the united states says: “I am thirsty”, we can infer that this is simply an assertion of reality. He very well could be thirsty. It is also an expressive act, it expressed an emotion.

And since the president’s emotions are waited hand and foot by his staff, it becomes both an order and a performative act, as his staff understands that he wants water, and will by virtue of his position of authority and white house institutional script, go get the water for him. By stating his thirst, the president has summoned water, and knowingly so.

But then, one might ask: how did the staff really know this was a request for water? Is it a coded demand? Does the president speak in a language we all understand but cannot speak ourselves?

Well, sort of.

Grice (1979) helps us shed some light on this matter.

Grice introduces conversational implicatures to explain how our social semantic field helps us assign meaning to sentences. Specifically, some sentences state only some part of reality, and have a sub-statement latent in their semantic content. When both speakers know (and are expected to) about the underlying statement, it is thus always inferred as present. Thus, in order to not have this statement present in the meaning of the sentence, it must be actively negated, lest we all know it is meant.

Thus, if we all know Dave is not very smart, we say that Dave is a brave boy, but we all understand the implied: Dave is not very smart. This is what Stalknaker (1978) refers to as common knowledge.

If you wanted to negate this implied meaning, you would have to actively say: Dave is a brave boy and a very smart one at that.

But if I come into the conversation, and I do not know Dave: I might have some doubt, because, while the sentence has a shape I recognize as usually used in conversational implicature, I’m not privy to the common knowledge that everybody else understood. I cannot prove that this is what was meant about Dave.

So what does this mean for trump?

Trump seems to heavily rely on this conversational implicature when making statements. He expresses the direct meaning of a sentence and he does not negate the implied meaning. Hence, all those who know the implied meaning understand that it is not negated, and therefore as good as said out loud.

Everyone understands this underlying meaning, implied without shame. But the problem arises when other parties deny this meaning was ever intended and was not understood by them at all.

Specifically, when the African-American community of the United States understands the underlying racism under some claims, because they were not negated, but an entire swath of the population denies they understood this meaning at all, it calls into question this shared knowledge.

It thus becomes difficult to prove, because intent is always inferred, and cannot be proven for a fact, mostly when, as we have seen before, the speaker cannot be held to truth in his statements.

This capacity to negate this common knowledge is explained by Ducrot (1983). Ducrot explains that implied meaning can be annulled, because the speaker was not engaged in this meaning, whereas the posed meaning engages the speaker, and can thus be targeted by negation. A stated content is not supposed to be annulable, whereas the implied statement was never truly engaged in by the speaker. We all engaged in this meaning for them by virtue of having the common knowledge.

This common knowledge is even more unavoidable as we live in a hyper connected society, where meaning travels across weakly tied groups. We have access to semantics, and categories from other groups. We know what they could mean, even if we ourselves do not subscribe to those categories, or use them on a common basis. Identification with a group is also the social statement that we adopt the semantics and categories of that group. Thus, if someone, identified with a group, does not negate the underlying semantic statement from that group’s common knowledge, it is assumed this knowledge is coopted.

Thus, when Trump, who identifies by association, and rejection of the democrats, with heavily right wing groups, and was appointed as the republican party nominee, engages in speech acts, he essentially engages in performative speech acts. Those speech acts are understood by all to entail some underlying meaning that drives action by various groups. The more radical the group, the more potentially violent these actions are, as their semantic common knowledge is driven by more violent rhetoric.

So, in the end, should it matter what a president meant?

Possibly not. Their identification is enough to give us probabilistic semantic content, and their position of authority makes these semantic contents into orders for groups who also identify the same way.

Our judgement of presidential speech acts should not only be linguistically pragmatic, they should be consequentialist. We should consider the consequences of potential speech acts, regardless of the annulable nature of the implied meaning.

Our role in this conundrum is thus not simply to infer the correct meaning. Our role is to ensure that individuals with the capacity to make such speech acts are not prone to brushing aside their engagement with truth, and are not prone to reckless endangerment by at best lack of consideration for the consequences of their statements, and at worst willful ordering of violence.

Anscombre, J. C., & Ducrot, O. (1983). L’argumentation dans la langue. Editions Mardaga.

Austin, J. L. (1975). How to do things with words (Vol. 88). Oxford university press.

Grice, H. P. (1979). Logique et conversation. Communications, 30(1), 57–72.

Searle, J. R., Kiefer, F., & Bierwisch, M. (Eds.). (1980). Speech act theory and pragmatics (Vol. 10). Dordrecht: D. Reidel.

Stalnaker, R. C. (1978). Assertion. In Pragmatics (pp. 315–332). Brill.