Unfortunately, deception rarely takes the form of blatant or outright lies, which could be detected by basic fact checking etc. Also, because we – along with professionals in forensic settings, and law enforcement – process information through the lens of a truth bias, we are not very good at detecting deception. It also appears that those working in law enforcement, as “professional” lie catchers are little better than college students at detecting deception. In short, we are often no better than “chance” at detecting deceit. We may tell ourselves otherwise, and that we know certain “tells”, such as liars avoiding eye-contact etc., however those who engage in deception also know these things, and will make sure that they never avert their gaze. In reality we are better at detecting deception through verbal, rather than non-verbal, cues, however we rely on them both when detecting deception e.g., if someone tells us that they are really looking forward to something but their body language says otherwise we are not likely to believe them. With online deception we are deprived of these non-verbal cues, and we are also usually deprived of context and setting e.g., we are on our keyboard, and the deceiver is somewhere else, on theirs. This often leaves us with only with verbal cues when we are analyzing and evaluating digital statements. In this article I want to look at some of the linguistic differences between truthful and untruthful digital statements.

When somebody is engaged in digital deception, they must construct a story that is believable to their intended audience, and manipulate language in order to communicate this narrative. Digital deception, like all deceptions, consists of two components: the withholding of truthful information (such as the deceiver’s real/true identity), and the promotion of untruthful information (such as selling a product online that they don’t have/own). This means that digital deceptive statements tend to be more vague and possibly evasive e.g., the person(s) engaged in deceit being unclear and ambiguous about who they really are etc., and shorter in length, than truthful ones. There are several reasons why deceptive statements tend to be shorter than truthful ones, with the main one being that the more “facts” which are presented, the more likely it is than one will be proved wrong; it is easier to manage a single lie/untruth than many. This is why someone’s initial statement might be believable but as they give out more information they lose track of the individual lies they’ve told, and start to contradict themselves. To counter this reduction in statement length, someone engaged in digital deception may engage in superfluous repetition in order to “pad” their statement(s) out. This also means that deceptive statements tend to contain fewer distinct words, than truthful ones. For example, someone responding to a chat on a dating site, might say, “I like your hair” (statement one), or, “I really like the way you’ve done your hair” (statement two). Statement one has fewer distinct words than statement two and is simpler and less complex i.e., statement two from the perspective of detecting deception is likely to be judged more truthful. However, certain caveats may apply e.g., the use of fewer distinct words may reflect a more limited vocabulary, or an inability to express, rather than be a sign of deceit, and so this shouldn’t be used as a “Pinocchio’s Nose”.

Another linguistic clue that someone may be engaged in deception is a lack of specificity. Truth tellers tend to make definitive statements, such as, “I didn’t do that”, as opposed to statements, like, “I don’t think I did that.” With a definitive statement, if it is found to be untrue, the person has no way out/back, however with a statement like, “I don’t think I did that”, there is room for error, and the person engaged in deception can correct themselves, if/when challenged on something. Like everything, context is important. There is a big difference between someone who creates a deceptive online dating profile as part of an eventual scam, where they ask money for a plane ticket so they can meet you in person, and someone sending you an email informing you that you have won the Norwegian Lottery, and your bank details are required in order for the money to be paid. In the first example (the dating profile), it is likely that the profile description, will contain many “small” lies (as has been shown in numerous studies), whereas the second example is likely to contain one or a few bigger ones. One of the warning signs that has been found concerning fake/deceptive dating site profiles, is that the profiles contained fewer self-references. An example of a self-referencing statement would be, “I like good food, wine and great conversations, so if you are looking for a nice evening out, we should get along great.” i.e., the person behind the profile is referencing the things they like, hoping to attract someone who likes the same. Deceptive dating statements tend to lack these types of statements, as the person engaged in deceit is in some way not wanting to reveal too much about themselves, distancing themselves from the profile. Also by self-referencing themselves they might later contradict something they’ve said in a chat/conversation e.g., “I don’t drink”, “I’m not really into long conversations, I’m more of the quiet type” etc. Another feature of such profiles is a lack of self-deprecation and negative emotions i.e., they are trying to present something/someone who is too perfect.

Detecting digital/online deception is difficult, as often the mode of communication is somewhat different to how we would communicate face-to-face e.g., texting/messaging uses a mix of shorthand and emojis etc. There is also the possibility in certain forms of digital communication, to revise and alter what is being said, after writing it. The fact that it is possible to read something before “sending” it means that someone who is engaged in deception, can look at previous messages/texts first, and check that what they are saying complies with what they’ve said before (long pauses and long response times may indicate someone is multi-tasking but it could also mean they are verifying previous statements). However, communication is communication, and putting our “truth bias” to the side, especially when there is a potential safety risk, and looking at statements from the perspective of being truthful or untruthful, may allow us to hesitate before committing to something that may put us in danger.