Humanity Hallows Issue 4 Out Now!
Pick up your copy on campus or read online

How good are you at telling if someone is lying? How many times a day do you think you’ve been lied to? Humanity Hallows talks to Manchester Met’s Professor Dawn Archer about her linguistic approach to spotting lies.

By Nima Khorramrooz

Professor Dawn Archer is the Director of Research and Knowledge Exchange in the Department of Languages, Information and Communications at Manchester Metropolitan University. Professor Archer’s areas of expertise include pragmatics, corpus linguistics and the discursive practices of the English courtroom.  She also has a growing interest in the language of emotion and more recently, has been investigating the linguistic performance of emotion, credibility and deception in a range of contexts. This includes work with the Emotional Intelligence Academy (EIA).

In recent years, you have been working on projects such as ‘Truth telling’ and ‘How to spot a lie’. Can you share your findings with us?

One thing worth highlighting, from the outset, is that there isn’t always agreement across researchers as to the specific details of how to spot a lie, and what the truth looks like. However, a growing number of us are concluding that it makes sense to look across the communication channels (i.e., the words, voice, face, body language, gestures and the autonomic nervous system). If we’re communicating with someone face to face, for example, we can focus on what their face and gestures might be telling us, as well as their words and tone of voice. If we’re close enough to them, we might be able to pick up on other signals too. Importantly, we have to make sure that we’re not confusing nervous behaviour with deceptive behaviour (as they can overlap). What we do, then, is to pay attention to all available information (from the communication channels), and “notice what we notice”. We’re especially interested in mismatches across the channels. The words might be saying one thing and the face might be indicating something else, for example.

‘Lying is a cooperative act. A lie has no power by its mere utterance; its power emerges when someone else agrees to believe the lie. In other words, if at some point you got lied to, it’s because you agreed to get lied to.’ Do you agree with this statement?

The statement is deliberately contentious. It’s picking up on the idea that there are different types of lie. By this, I mean, there are times when we are part of the lie. We might choose to ignore things about a partner because we don’t want or don’t know how to deal with the potential consequences if we call them on something. This said, there are also lies that are covert. With this type of lie, people say something or do something in such a way that they’re not found out. There are people who are very good at lying to us in such a way that we don’t know we’ve been lied to.

How early do we learn to lie?

We’re socialised into telling lies – albeit white lies, which are meant to protect other people’s feelings. We teach people to lie, from a very early age, for all kinds of altruistic reasons. A child is meant to say thank you for presents they don’t like for example – because it’s polite, and shows appreciation. As we get older, we learn that, if someone asks us whether their bum looks big in an outfit, they’re looking for us to reassure them. Unless you have one of those relationships where being brutally honest is ok, then you may well lie in such circumstances.

Some readers may not think of these as lies, because, for them, lies have an element of mal-intent behind them. Lies with an element of mal-intent involve people deliberately omitting, hiding or not providing information they have access to, for self-centred purposes. These are the types of lie that tend to be frowned upon in society. They’re also the types of lie that we tend to research.

How many times do people lie in a day and why is it easier to lie than to tell the truth?

People give different numbers in respect to how many times we lie in a day. I don’t know how you prove or disprove such figures. How many times you lie goes back to how many times you try to be nice to others versus trying to deceive them. How many times you lie probably depends on the context you’re in too. Is it different with family? Probably. Is it different across various professions? Absolutely. If your profession is to spy, you’re probably lying a lot (and may be so good at it, you’re believed to be telling the truth). The thing to note, here, is that people will quote numbers to you. However, often those numbers are based on particular research studies, which are then used as though they apply to every context (which is misleading). In respect to the second part of your question, I don’t think it is easier to lie than to tell the truth. This shouldn’t be too surprising, given that societies work, in part, on the assumption that most of us are telling the truth most of the time.

There is a principle within Linguistics and it relates to some work by Paul Grice who talked about the Cooperative Principle (CP). He believed that we can’t communicate unless there’s a level of cooperation between speaker and addressee(s). As part of the CP, he identified 4 maxims. One of those maxims was to do with the quality of the information we give. This equates to what we believe to be true – and that’s important because we might sometimes say something that we believe to be true, but then we find out later, it wasn’t true. Now, technically we haven’t lied because we said it believing it to be true.

Do you think social media makes people lie more?

Social media allows us to say things that might not be true, strictly speaking – and deliberately so. We can give a better version of ourselves. We can exaggerate who we are, and even take on roles that are fictitious. Are we doing that all of the time – rather than telling the truth – when using social media? I don’t know.

Another maxim proposed by Grice was to do with the quantity of information we give, when we communicate. If we’re sticking to the CP, we give the right amount of information, for the context. We don’t give too much or too little information. You can see how we can make use of maxims like this to determine when somebody provides a lot or too little information – and what this might mean.

Thanks to technology, we now have specialised eye trackers, infrared brain scans and MRIs that can help us spot a lie, but such tools can be fooled and are not reliable enough to be admissible in court. If these tools can’t be fully trusted, then how can we spot a lie?

The technology isn’t fool-proof, but it is getting better. This said, I don’t think it’s advisable to use technology to spot lies, without having a human operating the system, so that they can assess whether they’re being notified about everything they need to know, before making any decisions.

The best way to spot a lie, in my view, is to have people who are trained, so that they understand what the potential indicators of deception are across the various communication channels, and can hypothesise about the occurrence of these ‘points of interest’ when they do occur. This means not jumping to conclusions, because many of the indicators can point to things other than deception (especially if they occur in isolation).

In what ways can people improve their lie detection skills?

The first step is to check whether they’re believing some of the myths of deception detection. Unfortunately, some of the more popular books on deception detection seem to propagate these myths. This is true of some culturally-specific behaviours such as looking away, rather than having sustained eye contact with someone, for example. Looking away isn’t a reliable sign that someone’s lying. But it can be part of some people’s cultural practices, as a marker of, say, respect. We need to be careful, then, when we’re interacting with people from different backgrounds that we’re not misjudging them, because of a behaviour we have wrongly associated with deception. In fact, we suggest that people who are attempting to mislead may well maintain their eye contact with us, as a deliberate means of convincing us that they’re telling the truth. My interest in somebody rises, therefore, when they look too much me, without breaking eye contact, as I’m doing to you right now. Keeping eye contact for sustained periods starts to make many of us feel uncomfortable and we do what you did – we look away. Research suggests, further, that the hearer in a conversation looks at the speaker more than the speaker looks at the hearer, generally speaking.

The second step is to learn which indicators point to possible deception (and which don’t) – but then not go looking for them. Rather, they should let ‘points of interest’ find them. This means being attentive – in a way that signals genuine interest – and, when they become aware of changes in behaviour across the channels of communication, noting when – and hypothesising as to why.

What do people have to focus on when attempting to distinguish a credible comment from a lie?

There is a procedure called criteria-based content analysis that might be useful here. It’s made up of 19 criteria believed to be associated with credibility. It was originally developed to help assess the evidence of young children in alleged abuse cases. It’s used in some European courts today, as part of a wider process (known as statement validity analysis), but not in the UK. The process involves letting the other person give their account without interrupting them. Once that account has been made and transcribed, it’s a case of noting any incidences of the 19 criteria in the transcribed text. If you tell me a story, for example, and you put yourself in two places at once, it lacks coherence: as it’s impossible to be in two places at once. I thus have a very good indicator that what you’re saying isn’t credible. In contrast, if you make a mistake, whilst telling me your story, but you then correct your mistake, this spontaneous act will add to the credibility of what you’re saying. Another credibility features to do with spontaneity is spontaneously admitting that you can’t remember some things (as long as these things are periphery to the core of the story). One hypothesis is that truthful people are less concerned about admitting things such as poor memory recall than people who are fabricating their stories (because the latter want to be as convincing as possible).  One thing that you need to remember, though, is that credibility doesn’t mean truth. You can sound extremely credible but not be telling the truth.

You have an interest in the ‘language of emotion’. Can you explain to us what the language of emotion is and how it is used in detecting deception?

I’m most interested in emotional talk, that is, how the words and voice show different levels of emotion without an emotion being labelled, necessarily. If someone says, “I’m fine”, but their voice is clipped, and their jawline is particularly tight, I’m going to pick up that they’re not fine. I’m also going to infer that, for one reason or another, they probably don’t want to tell me anymore. In that case, I’m getting a mismatch between what they tell me and what I see. There are certain universal triggers of the face that we can look for in respect to emotions. Even micro facial expressions can be informative. You may have personal experience of this if you’ve been around an individual who has a tendency to get angry frequently. You may have even sensed the emotion before the person noticed it happening to them.

What language technique do we apply when we lie to others?

I’m not sure that most people think about lying to that level. What is more revealing, to me, are the linguistic and psychological concepts that can help us to better explain what people do, when they lie to others. Especially given that lies are most successful when we aren’t triggered to look for any meaning beyond the literal one stated. I often explore misrepresentation strategies, evasion strategies, equivocation, ambiguity, etc. Politicians are often cited as providing good examples of these phenomena. If someone asks them a question, they might give an answer, which we assume relates to the question, but it’s often more general and sometimes off-topic such that it doesn’t provide the specific answer sought by the question. There are interesting findings that relate to certain types of words such as pronouns too. I’ve done some research where people’s changes in pronouns (I, me, she, he, etc.) pointed to distancing. One of the reasons for wanting distance can be deception but it’s not the only reason. We use distancing techniques when things are too emotionally or psychologically painful for us, for example.

Is there anything else you’d like us to know?

Lie spotting isn’t as easy as some of the more popular books suggest. Where you can, start from the premise of believing others. Learn about potential indicators of deception – so that you avoid the trap of believing the myths — but remember that, even the indicators that are validated by the research are still potential indicators only. In which case, my advice is to not go looking for them. Instead, be genuinely interested in others – and then “notice what you notice.” When you do notice anything that appears to be a mismatch across the communication channels, remember to hypothesize. Deception should be the conclusion you come to when you’ve exhausted all other possibilities. Where you don’t have enough information, you should avoid making conclusions.