Mehrabian’s rule

Albert Mehrabian’s 7-38-55 rule is almost always misinterpreted. These misinterpretations probably aren’t too far from the truth, however, and they can probably explain why the Internet is such an obstacle to communicating effectively. Mehrabian’s 7-38-55 rule actually says that when we communicate face-to-face, how well we like the person that we’re communicating with depends on three factors: the words used, the tone of voice used, and the body language used. He first described this in his 1971 book Silent Messages, where he estimated that 7 percent of the overall level is due to the words used, 38 percent is due to the tone of voice used, and 55 percent due to the body language used. 

This is often generalized to saying that in any face-to-face communication that 7 percent of the information that we send is verbal, 38 percent is sent in our tone of voice and 55 percent is sent through body language. Mehrabian’s research doesn’t actually support this generalized result, but that hasn’t stopped people from calling this conjecture “Mehrabian’s rule,” or inaccurately attributing the generalized result to Mehrabian.

It’s probably the case that most communication is non-verbal, even if it doesn’t follow the 7-38-55 rule, and that’s why the Internet causes so many problems. Even if the exact fraction of the information that’s lost when we communicate on-line isn’t exactly 93 percent, it’s probably a significant part of it, and this causes many more misunderstandings than you would ever get face to face. Here’s an example of how this recently affected me.

There are now three IETF RFCs that describe identity-based encryption and how to use it in secure email. While getting one of these standards through the IETF bureaucracy I had to get three new media types defined for the types of data that get transported over HTTP when IBE is used: IBE public parameters, an IBE private key request and the IBE key that a of key server returns to a user.

There’s a mailing list where you propose new media types and other list subscribers get to critique your proposal. In my case, we had a heated debate over my proposal that turned out to be just over a slight misunderstanding in how one particular parameter was defined. If we were sitting down face to face, this misunderstanding would have been obvious almost immediately. But because we were only communication over email, lots of information was lost. Maybe it wasn’t exactly 93 percent of it, but it was a significant amount. This led to wasting several days in a debate that wouldn’t have taken more than a few minutes to settle face to face.

Lots of communications over the Internet seem to be plagued by the same problem, and because there’s no easy way to indicate tone of voice or body language over the Internet, we’re probably stuck with the imperfect communication that it allows. I suppose that you could write an IETF standard of some sort that might help fill the gaps in communication that the Internet creates, tags like <humor> and </humor> that you could use to indicate where you’re not being serious. But because you’d have to develop this standard over the Internet, it would be much harder to do than it needs to be.

Leave a Reply

Your email address will not be published. Required fields are marked *