Imagine you are a juror on a murder trial. A married couple have been found shot dead. The defendant, a man known to them, denies the charge. You’ve heard the prosecution’s evidence and you’ve heard his testimony. But you and your fellow jurors are unsure if you should believe his protestations of innocence. At the hotel in the evening, another juror makes a novel suggestion: contact the spirits of the dead couple to find out if the defendant is lying. In agreement, you all sit around a crudely constructed Ouija board and call upon the spirits of the dead couple to ask: “Who killed you?” The board spells out the name of the defendant. The next day, you return a guilty verdict to the court.
Sounds too absurd to be true? Well, in 1994 an English jury did consult a Ouija board (a retrial was ordered, and the defendant was found guilty again). But it is no less absurd than a jury being directed by the courts to use an assessment of body language to make a judgment. Judicial directions in Scotland advise jurors that they can “look at the content of witnesses’ evidence, [and] their body language in giving it”. Similarly, in England and Wales, jurors are instructed not to take so many notes during a trial that they are “unable to observe the manner/demeanour of the witnesses as they give their evidence”. It appears that the UK’s judicial system is no different from most of the population in assuming there is a clear association between body language or demeanour and deception – while being ignorant of the fact that looking at these to determine an individual’s honesty is not trustworthy.
The most untrustworthy body-language cue to identifying a deceptive speaker is the one that says they will not look you in the eye. But liars are very good at looking listeners in the eye, as they need to gather immediate feedback from the listener’s expressions to inform or adapt the content of their lie or how they say it. Another whopper in the body-language lie-detection repertoire is that the direction of a speaker’s eyes can be used as to determine if they are lying. Similar tells lacking scientific basis are regularly repeated, gaining common acceptance. These include behaviours such as mouth covering, nose scratching and even the barely imperceptible raising of an eyebrow.
Tests designed to measure lie-detection accuracy also show that we are strongly influenced by a person’s appearance. This type of facial-feature bias is found to affect the sentencing of defendants, meaning attractive defendants are more likely to be found not guilty of a crime or to receive shorter sentences.
The continued focus on dubious indications of deception from body language and demeanour assessments ignores the fact that lying is primarily a linguistic act. Our language exposes our inner thoughts, intentions, memories, experiences and emotions, and it is no different when we lie.
Research into language has revealed many indicators of deceptive speech. Liars may have control over the content of what they say, but they are less in control of the selection of words, the grammar, the length, and the construction and style of what they say. This is because lying is cognitively challenging. There is much that the liar must attend to: telling the lie, seeing how the lie lands and preparing to adapt it, if necessary, while also deciding what details to leave out or include. This is all undertaken in real time, which leaves the speaker vulnerable to linguistic cues slipping out.
One way that this occurs in language is when a liar is trying hard not to display too much self-focus. This usually takes the form of a sudden self-correction, replacing the first-person singular pronoun “I” with the plural one “we” instead. For instance, someone saying: “I … we thought”, or “I … we didn’t”. Liars may do this if they haven’t specifically thought through what it is they are going to say in detail, and they try to create distance from the content of what they are saying by hiding in numbers.
The language we use not only signals if we are trying to distance ourselves from a situation but also gives us away when we are inventing something, disguising how we feel or describing something that did or did not happen. For instance, imagine you are a juror again. You listen to one witness using language that includes sensory-perceptual details such as what they saw, heard and felt (“flashed past me … screeching tyres … felt my heart beating”), whereas a second witness does not include these types of words but refers to their internal thought process (“I remember thinking … I tried to recall …”).
When we live through a traumatic or dramatic event, even a minor one, we will tend to relate our experience using sensory-perceptual details. But the language of liars, who are fabricating an experience, will often lack this descriptive type of language and include claims relating to what they were thinking or reasoning instead.
Liars do not all lie in the same way, of course, and the linguistic cues to deception will also be context-dependent: the speaker could be being interviewed in a casual conversation or an official setting. The context, along with their goals and intentions, are all crucial in consideration of whether they are speaking the truth or not.
There is much that we can detect by listening out for linguistic signals of deception. But if our courts continue to encourage jurors to use judgments of truth based upon defendants’ and witnesses’ body language or demeanour rather than the content of what they say, they may as well let them consult the supernatural, too.
