The idea that the vast majority of communication occurs nonverbally is quoted everywhere: 53% face, 38% voice, 7% words. The figures come from two studies conducted by Albert Mehrabian in the mid-1960s. A bit like Pavlov with his dogs I’d never really thought about the practicalities of how one goes about measuring nonverbal communication; I blindly accepted the “facts” and got on with my poems. My reasoning was that because there were only the words on the page then 100% of the communication had to be verbal. I’ve always had issues with reciting poems.

 

In reality Mehrabian’s tests were very basic indeed. As he says on his website:

I’ve no idea what B. saw in my face when I spoke to her. Probably what she expected to see, a friend. And when I spoke she’ll have heard a friend. We see what we want to see and hear what we want to hear. Like most scientists Mehrabian never thought to factor in observer bias; from a scientific point at least we are not all equal. Would Mehrabian’s results have been different if the speaker was that pretty girl from the coffee shop on the corner or some Japanese holidaymaker they’d coaxed in off the street?