Human-centered design pioneer Don Norman, who coined the term ‘user experience,’ explains why he’s not convinced by the current obsession with empathy and what we should do instead.
I approve of the spirit behind the introduction of empathy into design, but I believe the concept is impossible, and even if possible, wrong. The reason we often talk about empathy in design is that we really need to understand the people that we’re working for. The idea is that, essentially, you’re in a person’s head and understand how they feel and what they think.
In my opinion that’s impossible, and here’s why. If I’m designing a medical rehabilitation device for a unique person, I could argue it’s crucial to really understand their likes and dislikes, their personality and issues, and how they approach the world. But that’s relatively rare. Most of the time, in our field, we’re devising products and services that are being used by lots of people — hundreds, thousands, sometimes millions. In Facebook’s case it’s even billions. This means that understanding the individual person is actually not going to be very useful.
Instead we must really focus on the activities that people are trying to carry out. We must also understand people’s capabilities and their points of view and how to support them. That requires us to understand the wide variety of abilities that people have. Let’s look at reading, for example. More and more of our texts have become invisible, by which I mean that they are technically present, but illegible. I simply can’t read them. They have been taken over by graphic designers who say ‘text is ugly and gets in the way.’ So they make it prettier by using a really tiny font with very low contrast. If you’re a 20-year-old programmer, you can read it, but an 83-year-old grouch like me can’t. I always carry a flashlight with me, so I can shine it on the text, which increases the contrast. Maybe I can then read it. What a pain.
You have to create things that are adjustable and variable, but even then, if you exclude the top five or ten percent, you can still be excluding a million people. So it’s really important that we address people’s real needs and abilities. We can’t get into the heads and minds of millions of people, and moreover we don’t have to: we simply have to understand what people are trying to do and then make it possible. We should not fool ourselves into thinking we can get into their heads.
Why do I think it impossible to have empathy?
Here are two examples. Firstly, I often don’t understand the things I do myself. It’s amazing how little empathy I have for myself. In part, that’s because most of our behaviors creep in subconsciously. The conscious mind then watches over it and tries to rationalize and make sense of it. Often the next day we wonder “why did I say that?” or “why did I do that and not this other thing?” My conscious mind has little or no empathy with my subconscious.
Secondly, the person I’ve been with the longest — 50 years at this point — is my wife. We get along really fine, but we often don’t understand each other that well. We often laugh about how she tells me something and I hear and understand it, but later on, it turns out that what she meant and what I understood were completely different! And that’s very natural in people.
It also hasn’t really got anything to do with people’s location. Sure, a designer in the Bay area may not understand the people in India, but people in Bangalore don’t understand people in rural India either. People may not even have been to a part of the city they live in, where people are poor or even homeless. They’ve never been there, or talked to the people, so of course, they don’t understand their problems. You don’t have to be very far away not to understand. It’s a cultural issue. There’s a cultural gap between the educated designers and the people in the streets they’re trying to design for.
It sounds wonderful but the search for empathy is simply misled.
The type of design that I’ve been advocating for years, the human-centered design process, means that we actually must go out and understand what the real issues are and what people’s real needs are in order to design something. So we send out anthropologists or design researchers, and the designers then try to figure out appropriate solutions constrained by the technology that’s possible, the costs, time frame, etc. Eventually, we build something and go through a series of iterations to make sure it really works for the population we care about.
That’s still an appropriate procedure, especially when you have a set of products and are trying to make it better, and when you’re dealing with mass markets. But in my current work I’m trying to look at different kinds of problems—the major issues that societies are facing in the world, like hunger, education, health, and security. A lot of people have tried to solve these issues over the last decades, usually following a human-centered design approach. Invariably, the history of success in these projects is dismally low, however. They either completely fail, take a lot longer than expected, or cost much more than had been budgeted.
Experts coming in and telling people what to do is also really paternalistic and doesn’t work. Experts don’t understand the local culture. They don’t understand the abilities, skills, and real needs of the people they’re trying to help. But there are really creative people in every community, and if you go there, you will always find people who already have solutions to the problems they’re facing.
However, local people can only solve the symptoms, not the underlying problem. Their interventions are relatively small. If you just solve the symptom and not the cause, it leads to the renewal of the symptom. So we’re proposing to combine experts and community workers. Instead of recommending solutions, experts should be facilitators, guides, and mentors. We need an approach that’s top-down, the expert knowledge, and bottom-up, the community people. This method will have to differ from community to community around the world.
There is actually more to our approach than the combination of community expertise with domain expertise. We still have the difficulties of tackling large, complex systems without getting all tangled up in political fights. We still have to ensure that domain experts do not patronize community experts, but rather facilitate and mentor. However, the total story is worthy of a book, not a few sentences.