Contemporary liberals often have a more 'nuanced' approach to speech than classical liberals. Sure, free speech sounds great in theory BUT, they will say, as Jeremy Waldron does, it can also be used to drown out minority expression, or to demean vulnerable people in society.
The issue is, of necessity, a little more complicated than giving everyone a free rein to say whatever they like. We don't necessarily give a mafia boss the benefit of the doubt when he 'implies' that someone should be beaten or killed. We don't demand that only the people pulling a trigger are actual criminals. So it is clear that various threats, incitements and conspiracies (that might all be forms of expression) might indeed be subject to criminal sanction.
Problems arise when we start to extend potential restrictions on speech to content that attacks someone's basis of self-respect, or their dignity. It sounds rather subjective and could amount to the de facto outlawing of offence, which liberals have traditionally defended (how else can society progress but without the occasional controversy?). When Canada introduced human rights against 'hate speech', their various Human Rights commissions investigating speech complaints developed a near perfect record of convictions. That sounds awfully like a purely victim-definied notion of 'hate speech'.
Victim thinks it is hate speech = it is hate speech
Obviously, this isn't any decent way to legislate. Almost any expression outside accepted norms could be subject to prosecution if a 'victim' can be found. Controversial speech could be strategically silenced by opposing activists, and soon no debate will be able to happen freely.
I put this sort of problem to Jeremy Waldron at a recent free speech discussion. I used the example of pornography and asked what kind of evidence would we need to say that a particular instance of pornographic expression was having harmful effects (usually, supposedly, on the interests of women). I find that current evidence surveys, at least those which include aggregate studies examining the effects that pornography appears to have at the societal level, indicate that pornography has either no relationship with harm to others, or may even have a moderately beneficial effect on others (slightly reducing sex crime and improving people's attitudes towards relationships).
Waldron responded that there was clear testimony from women, amongst others, for pornography's harmful effects. It ruins relationships, it creates impotence, it makes women feel objectified by men when they go out on a dates with them. It is responsible for violence against women.
The problem, I find, with this sort of testimony (this book is a massive collection of the stuff), is that it presents narratives that connect pornography with social harms without actually making the logical or empirical argument that pornography causes these harms. Many of these testimonies look transparently bizarre to anyone not already buying into the 'pornography causes rape' memes. This applies especially to the narratives offered by male sex offenders who are delighted to hold their pornography use as decisive in encouraging them to rape and abuse others. It is only in this instance when people seem suddenly willing to take rapists and sex offenders at their word.
It is true that victims of abuse also sometimes (but far from in every case) associate pornography with their abuse. But, once again, it is a narrative that tries to find some sort of explanation for what is really just morally weak people doing incredibly evil things. The vast majority of people can look at pornography without being adversely effected. And overall, pornography is not associated with any increases in sex crime. This suggests the problem lies with the people, not with the media they are watching.
Waldron suggested that, in any case, there should be no presumption of liberty here. It is up to defenders of pornography to demonstrate its beneficial or non-harmful effects in order justify its continued expression. This seems to make free expression a hostage to testimony. I am sure people in the past blamed rock n'roll, jazz or Catholicism for many of those social and personal ailments that are now blamed on pornography, and offered pursuasive narrative arguments for it too. I am sure a quick dip onto the Internet would find people happy to blame a Jew-dominated media for the breakdown in relationships between men and women. What we won't find is reasonable independent evidence to back up these sort of claims, just as with pornography. Should we pull Jerry Springer and David Baddiel off the airways anyway? After all, I don't think I can provide any reasonable independent justification for allowing either Jerry Springer or David Baddiel to speak on TV, so perhaps we should take some anti-semite's word for it, or at least test their hypothesis for a few years.
Traditionally things like impotence and sexual immorality were blamed on the devil. The testimony of people like sex offenders is just a repetition of 'the devil made me do it' hysterics. Pornographers and porn users are just the current generation of witches. What we need to know is how contemporary liberals would design their 'hate speech' codes to avoid the occasional repetition of the Crucible.
In this respect, Waldron drops another clanger from the perspective of any robust conception of the rule of law. He suggests that a senior prosecutor in a state should have a veto on whether to action a free speech case. This doesn't seem much of an answer. It is an admission that speech codes are probably too ambiguous to be trustworthy guides to what can be prosecuted in any given case. So it will just be some public official with a 'I know it when I see it' approach to sift through the possible cases. It replaces settled law with the opinion and preferences of one official. Speech and expression deserves more protection than that.