Principles › Hearing each other
Hearing each other
Check before assuming
Text loses tone. Ask what they meant instead of guessing.
For example
Porcupine · a tipRiley’s “fine” could mean a few things. You might ask what they meant before you answer.
How Porcupine uses it
- When a message could be read more than one way, Porcupine suggests checking before you react.
- Before rewording, it asks you (at most two questions) instead of guessing a fact it needs, like who someone is or what a reference means. A feeling your words make likely is different: it writes that in and points it out, so you can take it out (see Say what’s underneath).
- When you’re reading a message someone sent you, it offers a few possible readings, never a verdict.
- “I’ll say it in person” on any draft helps you get ready to talk, starting with a short text asking for the talk.
In practice
Offer a call or a talk in person when a text would lose the warmth or the tone, and let them choose. Not when they’ve asked for space.
Why it helps
Without a voice or a face, readers have to fill in the tone, and they can get it wrong, sometimes reading a message as more negative than it was meant. Senders, meanwhile, overestimate how clearly they came across. Imagining the other person’s view doesn’t fix this: across 25 experiments, it didn’t make people more accurate. Asking them did. A voice carries the tone that text loses: in experiments, people who disagreed understood each other better and clashed less when they talked than when they wrote, though they expected the opposite. So for something a text would flatten, offering a call or a talk in person can help, as long as the other person is free to say no. Porcupine slows both sides down: senders see their words before they’re sent, and readers are encouraged to check before reacting. Porcupine follows the same rule itself, asking before it guesses at a fact it needs.
What we don’t know yet
The studies show that readers misjudge tone in email. We haven’t yet seen research on whether a nudge to check first changes how people reply. In experiments, people who disagreed got on better when they talked than when they wrote, but in studies of couples, conflicts by text ended about as well as face to face, so text isn’t the wrong way to have a hard conversation. And no study we’ve found tests suggesting a call instead of sending a hard text.
See it in a conversation
- How to ask someone you’re dating where things are goingDating
- How to stay friends with your exExes
- What to do when a family member keeps bringing up the pastAnyone in the family
- How to ask a friend to pay you backFriends
- How to talk to your roommate about noiseRoommates
- How to push back on your manager’s deadlineManagers and employees
The research
Studied with adults in general
In five experiments, people overestimated how well their tone (sarcasm, seriousness, anger) came across over email. They “heard” the tone they meant, and found it hard to imagine reading the message without it.
Studied with adults in general
Across 25 experiments, imagining another person’s point of view didn’t make people more accurate about what they thought. In the final experiment, asking them did.
Studied with adults in general
In randomized experiments with 1,576 people who talked or wrote to someone who disagreed with them, the spoken conversations went better: more understanding, less conflict, and kinder views of each other. People expected the opposite, and preferred to write.
More research
Studied with adults in general
In experiments about polarizing opinions, people who heard someone explain a view they disagreed with judged that person as more thoughtful than people who read the very same words.
Studied with adults in general
In a field experiment and lab studies, people expected a phone call to feel more awkward than writing. It wasn’t: talking by voice built a stronger connection than email or text chat, with no more awkwardness.
Studied with adults in general
Two meta-analyses of negotiation and group decision experiments found that seeing and hearing each other, and talking in real time, helped people who came in neutral, made no difference for people already set on cooperating, and made things worse for people set against each other.
Studied with couples
In an experiment, 100 couples were randomly assigned to discuss a conflict face to face or by text chat. Distress, anger and how far they resolved it came out the same either way.
Studied with couples
A systematic review of 15 studies of couples found that handling conflict by text went with similar outcomes to face to face on average, though it depended on the people involved.
Theory
A theoretical review of how emotion travels by email argued that readers often take work emails as more negative, or more neutral, than the sender intended.
Library last updated October 2026.