Two Reasons Conversations Go Wrong: Nonviolent Communication and the Curse of Knowledge

When the words arrive but the meaning doesn’t

Conversations go wrong for many reasons, but two come up often. One is speaking in a way that makes the other person defensive; the other is assuming they know what you know.

Nonviolent Communication: observations, feelings, needs, requests

In Nonviolent Communication (Rosenberg, 2003), the psychologist Marshall Rosenberg proposed speaking in four steps instead of blaming or judging.

1. Observation: say only what happened, without evaluation. Instead of “You’re always late,” say “This week you were 30 minutes late three times.”
2. Feeling: say how you felt. “I was upset and a little anxious.”
3. Need: say the need behind the feeling. “I want to be able to plan around our time together.”
4. Request: ask for a specific action the other person can take. “If you’re going to be late, could you let me know ten minutes ahead?”

Rosenberg said that everything people say can ultimately be heard as either a “please” or a “thank you” — meaning we can look for the unmet need even in angry words.

The evidence is still limited. A small study found that empathy scores rose among parolees after NVC training (Marlow et al., 2012), but large randomised studies are lacking. Still, the principle of speaking with observations instead of evaluations and requests instead of demands is a useful starting point in conflict. You can look at your own conversation style with OIYO’s communication style test.

The curse of knowledge: once you know, you can’t go back

The curse of knowledge is the difficulty of imagining the perspective of someone who doesn’t know what you know. Camerer, Loewenstein and Weber (1989) first named it in economic experiments: better-informed people couldn’t ignore their own information when predicting the judgements of less-informed people.

The most famous example is Newton’s (1990) “tapping” study. One person tapped out the rhythm of a well-known song on a table, and another tried to guess the title. Tappers predicted that listeners would guess about half the songs, but listeners actually got only 3 of 120 (2.5%). The tappers heard the melody in their heads; the listeners heard nothing but knocking.

Some ways to explain better:

• Start with a concrete example: show one case before the abstract idea.
• Check in their words: instead of “Got it?”, ask “Could you tell me how you understood it?”
• Try it on a newcomer first: before finalising an explanation, run it past someone who doesn’t know the field.

Summary at a glance

Concept	Source and research	How to use it in conversation
Nonviolent Communication	Rosenberg: observations, feelings, needs, requests	Observations over evaluations, requests over demands
Level of evidence	Mostly small studies	Use as a principle, not a cure-all
Curse of knowledge	Camerer et al. (1989); tapping study 2.5%	Start with examples; check in their words

---

References

• Rosenberg, M. B. (2003). Nonviolent Communication: A Language of Life (2nd ed.). PuddleDancer Press.
• Marlow, E., et al. (2012). Nonviolent communication training and empathy in male parolees. Journal of Correctional Health Care, 18(1), 8–19.
• Camerer, C., Loewenstein, G., & Weber, M. (1989). The curse of knowledge in economic settings: An experimental analysis. Journal of Political Economy, 97(5), 1232–1254.
• Newton, E. L. (1990). The rocky road from actions to intentions (Doctoral dissertation). Stanford University.