The current debate over college students, millennials, and free speech is really heated. Because arguments about “kids these days” are, and always will be, so irresistible, there are a lot of sweeping generalizations being made. And since certain ideas about 21st-century young people have become more or less solidified, at least in some quarters — college students are incredibly coddled and fragile (nope), they’re uniquely anti-free-speech when it comes to Americans (probably not), and they’re too promiscuous (nope) — it often feels like this whole area of coverage has turned into a gloppy stew of confirmation bias.
For a very telling example, check out an article by Jonathan R. Cole, a sociologist at Columbia University, that ran in The Atlantic last week under the headline “The Chilling Effect of Fear at America’s Colleges.” Cole decries the “recent outcry against free expression on campus” and attempts to add on to prior, psychological arguments for this ostensible trend with some “cultural, institutional, and societal explanations.”
It’s a paragraph where he tried to buttress his argument with statistics that caught my eye:
There’s hardly consensus among students on the forms or appropriateness of these restrictions on speech. Today, nearly half of a random sample of roughly 3,000 college students surveyed by Gallup earlier this year are supportive of restrictions on certain forms of free speech on campus, and 69 percent support disciplinary action against either students or faculty members who use intentionally offensive language or commit “microagressions”—speech they deem racist, sexist, or homophobic. According to a free-speech survey conducted by Yale last year, of those who knew what trigger warnings are, 63 percent would favor their professors using them—by attaching advisories to the books on their reading lists that might offend or disrespect some students, for example—while only 23 percent would oppose. Counterintuitively, liberal students are more likely than conservative students to say the First Amendment is outdated.
Several things jump out. The first is that you should be immediately suspicious of any survey result about free speech that isn’t presented alongside some context explaining how these numbers have looked at other points in time. Oftentimes, numbers that reflect long-time, rather stable trends in public opinion can appear jarring out of context.
I should know, since I myself fell for this last year — I posted an article highlighting some sexy new findings about how anti-free-speech those darn millennials appeared to be, specifically that “40 percent of millennials would be in favor of government bans on speech offensive to minority groups.” Whoa! Except, as I realized and explained in a mea culpa follow-up, this number is well within the standard range of responses when not just college students but Americans have been asked if some forms of speech should be banned by the government. Despite what we’d like to tell ourselves about Americans’ fierce love of free speech, a lot of us have basically always supported “banning” its offensive varieties. What has changed, over time, is which categories of speech make people the most ban-happy — in the past it’s been pro-Communist propaganda, and more recently it’s been racial slurs. But at no point has it been unusual for something in the ballpark of “nearly half” of Americans to be in favor of government bans on some form of speech, so the fact that this same percentage of students are “supportive of restrictions on certain forms of free speech on campus” tells us exactly nothing. But, hey, without all that pesky context it sure does sound damning.
Then there’s the little matter of what happens when you click through to the Gallup survey itself:
Recall that Cole writes that “69 percent support disciplinary action against either students or faculty members who use intentionally offensive language or commit ‘microagressions’ [sic]—speech they deem racist, sexist, or homophobic.” But the table clearly shows that the 69-percent number referred to those who agree that “Using slurs and other language on campus that is intentionally offensive to certain groups” should be a restricted form of speech on campus. It’s not a stretch to think that many respondents interpreted the question as asking whether there should be disciplinary action if, say, one student hurls a racial slur at another — again, “intentionally offensive” speech.
But Cole describes the question, totally inaccurately, as referring to intentionally offensive language or microaggressions, which are small linguistic slights that are usually conceived of as stemming more from ignorance than malice, and which just about everyone views as far less serious than racial slurs (though they take an aggregate toll on minority students, particularly on campuses that aren’t diverse, the argument goes). Plus, right there, it says that almost three-quarters of students are against restricting students from expressing political views that are “upsetting or offensive”! It’s literally in the same table as the item he’s freaking out about!
Maybe Cole disagrees that his home university of Columbia should be able to discipline a Catholic student who calls a Jewish one a “kike” — I’d be interested to hear his argument if he does — but that survey question about “intentionally offensive” speech has nothing to do with microaggressions. Of course, the term microaggression does sound a bit silly and is an endless target of ridicule among the folks clanging the alarm bells about free speech on campus (“Kids are freaking out about microaggressions! Ugh!”), so Cole gets a lot more outrage-mileage out of describing this survey finding by affixing the term “microaggression” inaccurately to it. Again, it sure does sound daming.
Then, there’s Cole’s claim that “[a]ccording to a free-speech survey conducted by Yale last year, of those who knew what trigger warnings are, 63 percent would favor their professors using them—by attaching advisories to the books on their reading lists that might offend or disrespect some students, for example—while only 23 percent would oppose.” He doesn’t provide a link, but must be referring to the McLaughlin & Associates National Undergraduate Study (here’s the slide-deck PDF) which was published in October, and which got attention from groups like the Foundation for Individual Rights in Education.
The survey indeed found that of the 55 percent of students who have heard of trigger warnings (page 30), 76 percent favor their use. But this isn’t the whole story on trigger warnings. In December I wrote about a survey conducted by the National Coalition Against Censorship in which a group of professors affiliated with the Modern Language Association and the College Art Association were asked about their experiences with trigger warnings. While the survey, as I noted, wasn’t nationally representative of college professors’ experiences, it offered a decent first-pass attempt at understanding a phenomena that had been mired in overblown anecdotal accounts.
From the report:
Although fewer than 1% of survey participants reported that their institution had adopted a policy on trigger warnings, 7.5% reported that students had initiated efforts to require trigger warnings on campus, twice as many (15%) reported that students had requested warnings in their courses, and 12% reported that students had complained about the absence of trigger warnings. Despite a media narrative of “political correctness,” student requests concerned a diverse range of subjects from across the ideological spectrum.
In other words, these professors were not getting a lot of trigger-warning requests. So which finding should we give more weight to: that college students, forced into a yes/no answer over trigger warnings, say ‘yes,’ or that professors seem to only be rarely getting active requests for them? Seems like an easy answer to me.
Perhaps more importantly, Cole ignores a whole swath of findings from the Yale study which cut against his argument. The slide deck notes on page 9, for example, that “[e]ight in ten [students surveyed] believe that freedom of speech should either be less limited on college campuses or there should be no difference compared to society at large.” On page 14, the authors write that “When given a list of choices, just one in ten believes colleges should regulate free speech more.” And on page 29, the authors note that “Greater than six in ten say political correctness on college campuses is either a ‘big problem’ (19%) or ‘somewhat of a problem’ (44%)” — though, to be fair, this can be interpreted both as evidence that college students respect free speech and of evidence that they think that PC culture is a problem on their campuses (though they don’t appear to be too exercised about it).
For Cole to have ignored these findings is really bad cherry-picking. He can’t use the survey to support his argument that free speech is under attack by plucking out a rather specific, context-dependent finding suggesting he’s right, and then sweeping under the rug questions which offer countervailing evidence — especially when those questions are a lot more straightforwardly worded and therefore, arguably, easier to interpret than the trigger-warning one. When you actually page through the slide deck, there’s solid evidence that college students agree with Cole on this stuff. How does a mention of that not get into his story, if he’s interested in carefully grappling with the survey data?
People should embrace some complexity on this issue. It would be foolish to assume that there are no free speech issues on campuses — for the love of God, a (small) group of Emory students said they were traumatized by the fact that someone had scrawled Donald Trump’s name on sidewalks there, and the university’s administrators were sufficiently cowed by that group’s protests to claim, extremely creepily, that they’d try to track down the perps with security cameras (nothing happened) — but it would also be foolish to simply give in to every narrative about kids these days being worse than ever, about college campuses mercilessly tightening the free-speech clamp.
There are problems worth discussing here, yes, but we need to be able to talk about them in a calm, adult way. It may well be the case that campus-protest-culture overreaction is more visible than it ever was before due to social media, for example, but that things aren’t as dire as people say. Either way, survey data can only help us sort through all of this if it’s presented and interpreted in a good-faith way.