Tuesday, 14 December 2010

Bad but common arguments for free speech absolutism

"Freedom of speech" is, of course, an extremely important right. It's not, however, necessarily the most important right, and like all rights, it is not absolute.

The documented harm that certain forms of speech can cause - especially dehumanisation and other forms of hate speech - is serious enough for me to believe that it is proportionate to restrict it in these cases. Whether that restriction best takes the form of a state ban or private actions in the form of boycotts, protests or other criticism, will obviously depend on the details. (You might not consider the latter form a restriction on free speech, because it isn't, but see the "exaggeration of scope" section below)

Arguments against restricting freedom of speech in this way tend to mostly be based on a few extremely flawed themes. 1

The slippery slope

"If we criminalise a particular form of speech that we don't like, then we're giving power to the state to criminalise any form of speech it wants, and it won't always agree with us."

The first problem with this argument is that it can trivially be generalised to this argument.

"If we criminalise a particular form of behaviour, then we're giving power to the state to criminalise any form of behaviour it wants."

This is true, and there are various anarchist critiques of state power that are based around this (and I'm not addressing those here, because that's an entirely different argument - and a much more consistent one), but for non-anarchists, people who are quite happy for the state to criminalise murder or theft and apply a range of punishments for these crimes, but trust it not to extend this to criminalising "walking faster than 2 metres per second" or "owning a lettuce" or - more seriously - "being a member of the Opposition political party", there seems no real reason to treat speech with more cautiousness than physical actions.

The second problem is that the state already criminalises (or allows civil cases to be brought) for certain forms of speech and the vast majority of people are happy with this principle. Some examples:

  • Conspiracy to commit a crime.
  • Libel or slander (the UK laws do need a serious rewrite, but very few are arguing that it's a breach of free speech to have libel laws at all)
  • Knowingly or recklessly giving false advice on certain legal or financial matters.
  • Speaking via a megaphone between certain hours in residential areas.
  • Misuse of emergency services numbers.

The principle that the government can and should restrict or forbid certain forms of speech is well-established even in so-called "liberal democracies". There doesn't seem to be a particular risk of slipping into a totalitarian dictatorship as a result. Therefore we should be able to discuss exactly which forms of speech should be restricted or forbidden by the state without "all of them" or "none of them" being the only possibilities.

I do think we need to be very cautious about enforcing restrictions on speech, but that doesn't make them automatically wrong.

The marquee inconsistency

One of the major usability complaints about scrolling text banners on the web - bear with me for a moment! - is that they're inconsistent about importance. As the poster there puts it:

What information do you have that is not important enough to have a place on the screen at all times, but is important enough that a person will wait until it appears again if they missed it?

Arguments against restriction of free speech seem to fall into a related inconsistency. Restriction of free speech, it's accurately said, is one of the ways in which governments suppress criticism. Exercise of (certain forms of) free speech is powerful because it keeps the state accountable to its citizens and residents.

Having established that "the pen is mightier than the sword" and other such cliches, that speech is powerful and can have massive real-world effects, the argument that particular forms of speech should not be forbidden, restricted, or even particularly considered carefully by their users, because they can't be causing real-world harm, doesn't make a lot of sense.

(The argument that the real-world harm being caused to people who aren't the debater is a price worth paying to have free speech for other purposes, I think is just a variation on the slippery slope theme above)

Naturally this only seems to apply to speech that hurts the non-privileged. The harm caused by speech that hurts the default is well recognised and such speech is discouraged by various methods.

Exaggerating the scope

There are various forms of restriction on speech. One can, with varying degrees of difficulty - and I make no claims that this list is comprehensive:

  1. Forbid it magically and omnisciently.
  2. Forbid it to the full extent of the state's actual enforcement powers.
  3. Allow it under restricted circumstances but not more generally.
  4. Allow it under general circumstances but not in particular restricted ones.
  5. Allow it but with penalties for its use. (Non-state penalties, such as being boycotted, rather than actions via the civil or criminal law)
  6. Allow it but with situational penalties for its use.
  7. Allow it but disapprove of it.

It seems very popular to pretend that people suggesting options 5 to 7 are actually suggesting 2 (or somehow 1) and then argue against them on that basis, or to merge this with the slippery slope arguments and say "Well, you might only be recommending we disapprove of this, but that's the first step to throwing people in jail for it".

It's not an argument as such, but it seems a common overreaction to mild suggestions.


1 I'm not saying here there aren't good arguments against restrictions: I'm saying these aren't those arguments, and that they're commonly used enough to be worth noting.