The words we choose give away more than we realise.

The words you choose give away more than you realise.

Along time ago, I read that political ideologies – such as social democrat, fascist and conservative – don’t represent points of a continuum from left to right or liberal to conservative, but might be better described in two dimensions: the importance assigned to the principles of freedom on the one hand and equality on the other. Communists should like equality much more than freedom, and free-market conservatives the reverse.

I tested this in the New Zealand context by photocopying all the speeches in what is called the Address in Reply debate that started the 1995 parliamentary term. I scanned them into my clunky Windows 95 PC, converted them to text and searched for the words “equality” and “freedom” and their synonyms. Sure enough, National Party politicians paid more tribute to “freedom”, and Labour MPs showed a preference for “equality”.

This sort of thing is handy, because if I were to approach our politicians and ask them to tell me what they really think, most of them would tell me (a) to go away, if they didn’t just ignore me, or (b) whatever they think presents them in the most positive light.

You can do much more with politicians’ (or anyone’s) utterances than just add up how many times they say a particular word, but it will take you a while. So thanks to Professor Jamie Pennebaker of the University of Texas at Austin for giving us Linguistic Inquiry and Word Count, or LIWC – pronounced “luke” – for short.

Pennebaker developed the computer program to gain insight into the psychology of someone by analysing what they say or write. He’s not the first person to do this: computerised text analysis goes back to the early 1960s at least, but LIWC does rather more than those early attempts. It was originally designed to automate the process of counting emotion-related words as part of a research project to look at how one’s writing about negative life events might predict one’s subsequent health outcomes. Early retrospective investigations showed, for example, that the language used in bodies of poetry reliably differentiated suicidal poets from their more light-hearted peers – for example, through the frequency of different types of pronouns, consistent with hypotheses derived from at least one theory of suicide.

At the recent Innovations in Health Psychology conference, held in Wanaka, Pennebaker spoke about the use of LIWC to analyse Donald Trump’s first State of the Union speech. You will be entirely unsurprised to hear that Trump’s language scores him highly on the dimension of confidence (or “clout”), which is reflected in plural pronouns (“we” rather than “I”) and frequent use of negations (“no” or “not”, among others). It would surprise few to hear that he displayed less evidence of analytical thinking; his ways of structuring ideas were simple and intuitive.

The surprise is that although some commentators would like to believe that Trump is particularly puffed-up and simplistic, his address style is following a trend that’s been developing since Franklin D Roosevelt was in the White House. Pennebaker has done the numbers on every State of the Union since then and concluded that US presidents have become steadily less analytic and more confident in their language.

The same holds true for British prime ministers, so the Brits shouldn’t feel too self-righteous. In short, American and British leaders are adopting increasingly uncomplicated visions of their world and taking stances on the issues of this simplified globe.

Nobody has ever done this analysis in New Zealand, however, and until someone does, we won’t know whether we can tell ourselves that we have bucked this particular trend.