Julie Sedivy is a cognitive scientist with an indomitable interest in language. She is a former faculty member of Brown University, and is currently an adjunct professor at the University of Calgary. She now devotes much of her time to writing about language for a general audience. She is the lead author of Sold on Language: How Advertisers Talk To You and What This Says About You. She blogs regularly for Language Log, Psychology Today and Discover, and tweets as @soldonlanguage.

One of the undergraduate courses that I most enjoy teaching is called Language and Advertising. Each year, I start off by presenting a particular scenario to students:

Imagine that the beef industry has joined forces with a company that makes a cola drink, and has worked out a deal to have a specific chemical that induces cravings for beef dissolved in the cola. The chemical is perfectly safe, and is listed among the ingredients; anyone is free to research its beef-craving properties.

Is this an ethical form of persuasion?

Most students say no. When pressed for an explanation, they typically say things like this: The companies are taking advantage of the fact that most consumers will be unaware of the effects of the chemical. Using techniques that are outside of the awareness of consumers is sneaky and dishonest. It undermines their freedom of choice, because consumers don’t know what is driving that choice.

I then ask them:

Do you think it’s ethical for a company to use language in its advertising as a means of persuasion?

As you might imagine, this question gets some strange looks. But the reality is that much of what happens in our minds as a result of language is just as hidden from our conscious awareness as the effects of fictional beef-craving chemicals. But while most people take it for granted that a great deal of their brain chemistry does its thing outside of their conscious knowledge or volitional control, they don’t have the same beliefs about language. This leads to a strange illusion about language: not only is much of what our minds do with it hidden from us, but the fact that so much is hidden is hidden from us.

Perhaps this is why a great deal of how language is discussed—whether it’s about linguistic pedagogy, language policy, language and persuasion, literary aesthetics, or language and culture—occurs without consideration of what is actually scientifically known about language. Authority figures, including members of the Académie Française, or literary luminaries such as George Orwell, have often made pronouncements about language which are readily accepted as intuitively obvious, but which make contemporary language scientists wince.

To me, the great fascination in studying language has been the discovery of its iceberg-like qualities, how some of its most intriguing properties lie well below the surface, accessible only through meticulous observation or ingenious experimentation. Scientific work on language has revealed that humans possess a truly staggering linguistic intelligence, a body of knowledge which is mostly made up of things we don’t know we know.

For example: Most English speakers, if asked, would report that the last sound in the word cleared and the first sound in the word dice are the same. But in fact, they are pronounced slightly differently. Your conscious mind may not be aware of this, but actually, your less conscious mind can put that information to good use as a perceptual clue about where words begin and end. This, it turns out, came in especially handy when you were an infant; before you actually knew many words, human speech sounded like an uninterrupted string of sounds rather than a series of words, much as a foreign language that you don’t know might sound to you now. But, as discovered by researchers Sven Mattys and Peter Jusczyk, infants younger than nine months can leverage the subtle differences in d sounds to make smart hypotheses about how a continuous stream of speech might be carved up into words. So even without knowing the words, they can guess at the word boundary in the phrase cleared ice versus throw dice.

In my own lab-based research, I’ve looked mainly at how people manage to cope with the  ambiguity inherent in language, how they make rapid-fire decisions in interpreting a chunk of language which could be consistent with several interpretations. What we’ve seen from this work is that humans are able to very quickly integrate a number of different streams of information, ranging from statistical expectations about various linguistic structures to inferences about a particular speaker’s probable communicative intent or capabilities. And again, most of these decisions are happening completely outside of people’s awareness. To tap into them, you can’t simply ask people what they were thinking while processing a sentence. Often, we have to look at more reflexive behaviors as a clue, such as tracking their eye movements while they listen to language.

Once you start to dig around seriously in the guts of how language really works, though, it can change how you look at everyday language around you. While I was still a graduate student, my research meetings with my advisor, Greg Carlson, would often devolve into observations about how TV advertisements we’d seen had exploited this or that aspect of linguistic knowledge or language processing. These lingering conversations were the seeds for the book we eventually wrote together, titled Sold on Language: How Advertisers Talk To You and What This Says About You.

In the end, my reason for encouraging my students to compare language with beef-craving chemicals is not so much to raise alarms about nefarious mind-controlling practices by advertisers. Rather, it’s to point out that most of the information-processing that we humans do—including language—hums along at a fairly automatic level outside of our deliberate control. This means that in order to understand how we respond to language, how it might influence our thoughts and our behaviors, we have to move beyond intuition. We have to get a little bit intimate with the science.