Catherine Jones on what language reveals about us

It had been a typically robust exchange one Prime Minister's Question Time last December. All was going well for Gordon Brown until he tripped up. "And we not only saved the world, er, saved the banks ..."

But it was too late; the Commons had already erupted at this revealing slip. He secretly thinks he really has saved the world, everyone concluded. Anything Brown said next would be forgotten.

Sigmund Freud believed that some slips of the tongue are caused by the subconscious mind: a consciously repressed thought or belief that is unconsciously released. These "Freudian slips" were perhaps one of the earliest psychological explorations into what speech reveals about us: the way we speak to each other - what and how something is said, unsaid, what we hear and misunderstand - reveals much more than just a glimpse of hidden motives lurking in our subconscious.

Speaking in code

While Freud is celebrated for pioneering the talking cure, it is others, many from outside the field of psychology, who have made significant contributions to the study of human communication. It was linguist Ferdinand de Saussure's Course in General Linguistics, published posthumously in 1916, that familiarised the study of language as a formal system made up of linguistic signs, the signifier, the signified and the referent. Another model of communication was developed in the late 1940s by two US telephone engineers. Claude Shannon and Warren Weaver were interested in finding a way for communication channels to work more efficiently and so developed the Shannon and Weaver model, depicting communication as messages that are passed between people as a linear process, starting with an information source (the speaker) at one end, passing the message along a channel (conversation, by phone or fax, for example) to a destination (the listener).

While this model may sound obvious, it is of significant value to psychologists who use it to talk in quantitative terms about each stage of the process, setting up experiments accordingly. Shannon-Weaver is an important model, however, it merely depicts language as a code comprising words and phrases; it cannot capture the true complexity of human communication. Conversation is always a two-way street, where each of us must try to listen as much as we speak: the speaker receives signals such as verbal clues from the listener the moment he starts speaking, modifying his message accordingly.

There is ambiguity and misunderstanding to overcome too. Language philosophers put forward an intentionalist view of language that, unlike Shannon and Weaver's model, acknowledged that a word or phrase can take on a variety of different meanings, making any message potentially ambiguous. So, depending on the speaker's intention: "Can you drive my car?" can be interpreted as either a request or a question of ability.

It's not just what we say, it's how we say something that tells our story. Our opinions of others are based in part on how they sound. No wonder then that some politicians, most famously Margaret Thatcher, opt for elocution lessons to help lose an accent or lower their tone for added gravitas. While most of us could confidently identify a person's gender by their voice, psychologists have proven that voices can reveal much more than that. Professor Robert Krauss of Columbia University's Human Communication Lab randomly stopped people in Central Park and asked them to read test sentences that were recorded on to tape. Back in the lab, another group was asked to guess each speaker's height, weight, age and gender. Everyone guessed gender correctly, and performed well on every other category, Krauss explains, with the exception of estimating women's weight. He is still unsure why.

Breaking it down

Today the study of verbal psychology is entering a new era; one in which computer programmes are able to unravel our subconscious so forensically it would make Freud blush.

Just over a decade ago the social psychologist James Pennebaker at the University of Texas developed a computer program called Linguistic Inquiry and Word Count (LIWC). This programme can analyse millions of text files, computing the number of pronouns, adjectives, words about emotions, work, inhibition and so on.

His analysis has revealed that women tend to use more pronouns - such as "I", "we" or "you" - and refer to others more than men, who are more likely to use longer words, articles and prepositions. Meanwhile, older people tend to refer to themselves less often, use more positive emotion words, more future tense verbs and fewer past tense verbs. Our language even gives our status away. The fewer emotion words and first person singular pronouns we use, the higher our social class.

Pennebaker has discovered that our word choice reveals more about us than we think. But rather than focussing on conspicuous phrases or unintentional slips, he looks at function words. These include pronouns, prepositions and articles such as "I", "you", "to", "of", "on" or "that".

The average vocabulary of a native English speaker is 100,000 words, yet we rely on just 400 function words for half our daily speech.

While we may consciously pick up that someone sounds depressed, insincere or arrogant, it has taken a computer programme to reveal that this is in part down to whether that person relies on certain pronouns, Pennebaker explains. His research has shown that our use of function words, specifically "I" and "we" can reveal a remarkable amount of information. When people tell the truth, they are more likely to use first person singular pronouns and exclusive words, that distinguish between what they did and didn't do, such as "except", "but" and "without". Similarly, people who are depressed, suicide-prone or experiencing a traumatic event tend to use "I" more.

Pennebaker analysed the former New York mayor Rudolph Guiliani's press conferences from 1993 to 2001. In late spring 2000, a series of events occurred to him within a month: he announced the breakup of his marriage, his affair with another woman was made public, he was diagnosed with prostate cancer and he withdrew from the senate race against Hillary Clinton.

Text analyses of his press conferences in the months surrounding this period revealed that his use of first person singular pronouns increased. In the days and weeks after a socially shared disaster, such as 9/11, pronoun use changes again. Pennebaker recorded a drop in use of "I" and an increased use of "we" as people switched from thinking about themselves to focusing on friends and family.

Speech and status

Just as we "mirror" another's body language while talking with them, we do the same with pronouns. This mirroring happens within seconds of the conversation starting. How we shift our language depends, again, on status.

Researchers have noticed how people unconsciously shift their speech and voice style to more closely match those of people in powerful or authoritative positions. One study that analysed interviews on the Larry King Live television programme found that when King was interviewing prominent guests, such as Bill Clinton, his language style changed to match Clinton's. The opposite was true when interviewing guests of a lower social status.

Our unconscious will always shape much of what we say and how we say it, yet the growing interest in uncovering hidden meanings and messages makes Pennebaker uncomfortable. "The implications for this psychology are, frankly, creepy," he admits.

Technology in the age of Google means billions of words can be logged and analysed. Could intelligence services start examining the language of world leaders in press conferences to detect a country's true intentions? Will we require potential partners to sit a verbal test before dating them? "It's all in the hands of whoever uses this technology" says Pennebaker.

Clean communication

Given how personal and revealing our language and communication style can be, it's perhaps no surprise that another key theory comes from the world of therapy. Twenty years ago therapist David Grove developed a set of questions, as free as possible of metaphor, assumption and suggestion, designed to help his patients to explore the stories behind their own words and metaphors, with as little interference from the therapist as possible. In addition, using these questions in everyday life can make it possible to avoid many of the misunderstandings and ambiguities that appear in conversation.

"Clean language" is based on the principle that we each have our own notion of what a word means, uniquely personal to us. Think of the word "flower". What do you see in your mind's eye? A rose? A sepia photograph of a rose? A cartoon? Any word has as many personal interpretations as there are people, yet much of what we understand from everyday conversation is based upon the assumption that we all mean the same thing by the same word, according to Judy Rees, co-author of Clean Language.

Asking clean language questions gives the listener more clarity, detail and understanding. The two most common questions are: "What kind of [flower] is that?" and: "Is there anything else about [that flower]?" When someone at work asks you for a status report, for example, they may want a quick two-line email from you, or a bound 100-page dossier. Asking, "What kind of report?" can avoid a costly misunderstanding.

Word choice in speech is normally unconscious, says Rees. Our words express the metaphors which underpin our thinking, which in turn express who we are, our values and our life experience. She has observed how different groups reveal many of their values in the metaphors they use. For example, when asking senior management teams to describe what they're like when they're working at their best, they often use competitive, sporting metaphors - "we're like a gold medal-winning team" - because winning is important to them. Pose the same question to a room of teachers and the metaphors are startlingly different - "it's like tending a garden, or bringing up a family" - because nurturing is an important value for this group.

Other experts have studied our tone of voice, the pauses in our speech, the role we take in conversations and our use of fillers - for example, "um" or "you know" - to reach many more conclusions. Language is, after all, our most comprehensive method of communication, and yet there is still much to discover about how and why we say what we say and the true meanings that lurk behind our conversations.

Further reading

• Clean Language: Revealing Metaphors and Opening Minds by Wendy Sullivan and Judy Rees (Crown House)

• Influence: The Psychology of Persuasion by Robert Cialdini (William Morrow)

• Yes! 50 Secrets from the Science of Persuasion by Robert Cialdini, Noah Goldstein and Steve Martin (Profile)

• The Power of Persuasion by Robert Levine (John Wiley)

• Nudge: Improving Decisions About Health, Wealth, and Happiness by Richard H Thaler and Cass R Sunstein (Yale University Press)

• The Language Instinct by Steven Pinker (Penguin)

• The Stuff of Thought: Language As a Window Into Human Nature by Steven Pinker (Allen Lane)

Thanks to who have provided this article. View the original here.