Dialogue: listening, disagreeing, thinking together
How to listen and disagree to some purpose, when a group is smarter than one person, and what changes when the other party is a machine.
A course for anyone who has to reach agreement with others, and for those who talk more and more to a machine and less to people. You will sort out which communication techniques have been tested and which are merely popular, be able to argue usefully and run a group discussion, and understand how a chatbot differs from a person beside you.
So far there is only this table of contents. Lessons appear one at a time, and we promise no dates. Feed of new lessons (Atom)
17 lessons, 4 parts
Part 1 Listening
Popular listening techniques have been tested less than people say, which does not make them useless.
- coming soon
Active listening: what the experiments showed
People feel better understood, yet are no more satisfied with the talk than after advice: what to take from that into life.
- coming soon
Nonviolent communication and its limits
Observation, feeling, need, request: where Marshall Rosenberg’s four steps fit, and where they sound like a trick.
- coming soon
The machine that always listens
Why a chatbot seems attentive, and how the feeling of being heard differs from someone actually caring.
Part 2 Disagreeing to some purpose
A good argument changes something, and there is now data on how that happens, not only advice.
- coming soon
State their position first
Restate the other person’s argument until they agree with your version, and only then object: how that works and where it breaks.
- coming soon
Ten minutes at the door
Can a short conversation shift a prejudice for months: a door-to-door experiment, and the fabricated paper that came before it.
- coming soon
Arguing with a machine
A talk with a model can weaken belief in a conspiracy theory, which means the same persuasive power can also work the other way.
- coming soon
When not to argue
A gap in power, exhaustion, someone who needs rest: how to notice an argument has stopped leading anywhere, and how to leave it.
Part 3 Thinking together
A group can be smarter than any of its members or dumber than all of them, and the conditions for each are better known than you might think.
- coming soon
Galton and the ox
How a crowd at a livestock show came close to guessing an ox’s weight, and why different people’s errors cancel out.
- coming soon
When a crowd gets dumber
Even mild social influence undermines the wisdom of a crowd, and a group often fails to pool the facts only a few of its members hold.
- coming soon
A little debate, then a count
Why a few small debates can give a better answer than a large crowd, and what that means for your next meeting.
- coming soon
Does a group have a general intelligence
Does a group have a general factor, as one person does: who found it, who doubts it, and what to take from that in practice.
- coming soon
Citizens at one table
How Ireland’s citizens’ assembly took up abortion before the referendum, and what such a deliberation needs in order to work.
Part 4 Beside machines
A partner who always agrees and is always available changes how we think and who we end up with.
- coming soon
Why the model agrees with you
Why assistants tend to tell you what you like, what your own ratings have to do with it, and what all that agreement does to your judgment.
- coming soon
How to get a model to object
How to ask for objections, test advice by asking the question the other way round, and tell real criticism from politeness.
- coming soon
A friend in an app
A chatbot for company as an adult: what it gives, what it does not, and why a link with loneliness does not give you the cause.
- coming soon
Living people nearby
What a convenient partner cannot replace: what is known about human ties and health, and what here is our hypothesis, not data.
- coming soon
A look ahead: a go-between in every conversation
Models write your messages and answer other people’s: what becomes of conversation when machines speak on both sides.