There has been a steep rise in the number of people labelled autistic over the last 20 years. Is that down to greater understanding – or overdiagnosis?

The home of 85-year-old Uta Frith does not exactly look like the latest front in the culture wars. Located in a serenely quiet corner of north-west London, it suggests a life still well lived, a set of deep interests and an idyllic kind of retirement. There are so many books that some are reachable only using a very tall ladder. The four hours I spend talking to her – and her husband Chris, an expert in schizophrenia – are an absolute pleasure. There is no sense of outrage or upset: even when I ask her about the stream of hostile and abusive emails she has recently received, she is relaxed and phlegmatic. “I tend to cope by just deleting them,” she says. “I don’t dwell on it.”

“It” is a reference to the controversy kicked up by her recently expressed views on a subject that sits at the heart of Frith’s academic career: what it is to be autistic, and who should or shouldn’t qualify for a diagnosis. Back in March, Dame Uta Frith – she was honoured with the title seven years ago – gave an interview to TES magazine (formerly the Times Educational Supplement) scattered with stark pronouncements. Among them was a claim that the so-called autistic spectrum has “gone on being more and more accommodating”, to the point that it “has come to its collapse”.

Then, in August, she published a piece in the academic journal Psychological Medicine claiming autism was in dire danger of being turned from a “precise, clinical” term into a “vague, culturally amplified identity category”.