Double Empathy
For a long time, people said that autistic people had trouble understanding others. And there is something true in that sentence. But it is only half the sentence.
Here is the other half: non-autistic people also have trouble understanding autistic people.
The gap runs both ways.
When two people from very different places try to talk to each other, it is not one of them who fails. They both struggle. They carry different habits, different signals, different ways of showing they are listening or interested or pleased. Neither set of habits is the right one. They are just different, and the difference belongs to both of them.
This is what researchers call the double empathy problem. Not a problem with one person. A problem that belongs to both people in the room.
Here is something interesting. Put two autistic people together, and often the gap shrinks. They find each other easier to read. They understand why the other person needs to explain the fish in detail, or why the pause before answering isn't rudeness but thinking. They share something that is harder to put into words than it is to experience.
This is the double empathy problem working in reverse: the closer two people are in how their minds work, the less translation the conversation needs.
The concept was named in 2012 by Damian Milton, an autistic researcher and sociologist, in a paper whose title politely buries the point: "On the ontological status of autism: the 'double empathy problem'." The argument was that much of what gets described as autistic social difficulty is not a property of autistic people alone. It is a property of the interaction between two people whose ways of moving through the world are different enough that each finds the other hard to read.
The word double is doing real work. Autism had historically been framed around the single empathy failure: autistic people struggle to understand the thoughts and feelings of others. The research Milton drew on, and the research that has built up since, adds the other side. Non-autistic people also misread autistic communication. They rate autistic adults as less likeable in first impressions, even when the content of what's being said is identical. They miss the warmth and interest in an autistic person's face or voice because they are reading a different set of signals. The failure, such as it is, is mutual and symmetric.
The most striking evidence for this failure is what happens when you take neurotypical people out of the room. Studies comparing autistic-to-autistic interactions against autistic-to-non-autistic ones consistently find that autistic adults communicate more successfully with each other: higher rapport, more effective information transfer, more accurate reading of intentions. The difficulty is not in autistic communication per se. It is in cross-neurotype communication, which is a different thing entirely.
Why does this matter? Because if the difficulty is one-directional, the remedy is to change the person with the difficulty. And for decades, that is exactly what many educational and therapeutic approaches aimed at: teaching autistic children to make better eye contact, hold conversations in the expected format, pass as more neurotypical. The approaches were sometimes well-meaning. The effect, accumulated over years, was usually corrosive.
If the difficulty is bidirectional, the remedy looks different. It becomes a question of building shared understanding rather than one-sided compliance. It asks non-autistic people to do some of the learning too: to notice when they are misreading, to check what a silence or a blunt sentence actually means, to stop evaluating a person whose social style differs from theirs as somehow deficient.
This is harder than it sounds when the default assumption has been, for decades, that one side of the gap is where the problem lives. But the research is consistent, and the people who have lived on the misread side of the gap have been saying this for longer than the research has existed.
- Asking what a person means, rather than assuming you know from how they said it.
- Recognising that flat tone, no eye contact, and blunt sentences are often neutral rather than negative signals.
- Learning some of the communication patterns that autistic people use with each other, rather than expecting translation to always run one way.
- Explaining your own signals when you can ("when I go quiet it means I'm thinking, not angry").
- Treating every misunderstanding as the autistic person's error to fix.
- Teaching autistic children that their natural way of communicating is wrong.
- Measuring connection only in eye contact and small talk.
- Assuming that because a communication style is unusual, there is no warmth or interest behind it.