The way I read the article is that yes in every society some speech is deemed acceptable and some is not. But 80% of respondents think the pendulum has swung a but too far.
Is that what Yascha Mounk thinks or what his data says? My argument is that the two things are not the same. For instance:
"Nowadays, too many ordinary behaviors are labelled as sexual harassment"? 51/49 in favor, not 80/20.
"Many white people today don't recognize the real advantages they have"? 52/48 in favor.
"Many people nowadays are too sensitive to how Muslims are treated"? 49/51 against.
"Today's feminists fight for important issues"? 54/46 in favor.
Yes, a clear majority agree with "Political correctness is a problem in our country". But a similarly clear majority, plotted on the same graph, also agree with "Hate speech is problem is our country". And, again, "political correctness" isn't defined and is a fuzzy concept.
To complicate matters even more, people's overall perceptions are the result of distinct experiences across two different mediums, the physical and the digital, each with its own characteristics. I think studies ought to differentiate between the two. The barrier to online speech is substantially lower.
It sounds to me like yes, the problems like racism and sexism do exist, but the way they are used to bully and shame one's political opponents is unacceptable.
Or may be I simply have fallen into this exact trap: put my own meaning into a fuzzy article and then hastly proceeded to passionately agree with myself.
So while I don't really agree with you (I would be "somewhat disagree" in an opinion poll on that question), I think it's a totally legitimate premise to have.
My objection is to the article's (really, two separate layers of articles!) claim to have clear empirical support for that premise, when the data is anything but clear.
The way I read the article is that yes in every society some speech is deemed acceptable and some is not. But 80% of respondents think the pendulum has swung a but too far.