Panics over apparently novel threats to freedom of speech are nothing new. Yet, much like anxiety over the state of the youth, the latest free speech crisis is always presented as unprecedented. And as with other moral panics, those sounding the alarm often turn out to have little genuine interest in what they claim to be protecting. The most vociferous opponents of ‘cancel culture’ have been strangely silent – or worse – in recent weeks on the suppression of speech in support of the Palestinians or critical of Israel. ‘Free speech’ has a political function. Elaborate twists, somersaults and pirouettes are performed to make this most sacrosanct of freedoms do the bidding of those who invoke it.
The results can be discombobulating. Censorship is not the answer to speech we don’t like, we are told: fight speech with more speech. But critical speech itself is routinely cast as a form of censorship. Racist speech is protected; speech which criticises racist speech is an attempt at suppression. Meanwhile, it is permissible, indeed routine, to say that ‘the left’, ‘feminists’ or the ‘trans lobby’ are bigots – the real racists, sexists, antisemites – without setting off the free speech alarm. What this amounts to in practice is that calling someone a bigot is an offence against ‘free speech’ when it’s true, but not when it isn’t.
Underlying this discourse is a common-sense idea of free speech, one with roots in liberal theory. What it means to have freedom of speech, in this conception, is simple: nobody stops you from speaking (for example through a law banning or penalising the expression of certain opinions). Restricting speech, in the traditional liberal view, is justifiable only in rare and circumscribed conditions: where speech poses an immediate risk of dire harm; and in particular situations, such as meetings, where we may temporarily stop someone from speaking in order to allow someone else to be heard (the so-called ‘chairman principle’). By contrast, speech must not be restricted on the grounds of its propensity to cause offence (distinguished with somewhat artificial sharpness from the category of ‘harm’), however warranted or painful. While there may be grey areas, this model is attractive because it seems to make free speech a ‘default’ situation: we have it until someone interferes to prevent us from speaking.
But things are not nearly so tidy as they seem. The liberal conception trails loose threads: pull at any one of them and the whole thing threatens to unravel. One of these concerns the chairman principle. The right to speak without being drowned out by the speech of others only makes sense in specific contexts where you have a legitimate expectation of speaking and being heard – in a debating chamber, perhaps, but not during a concert or a firework display. This suggests that the apparently pristine principle of free speech is something far more pragmatic and conventional. Also implicit in the chairman principle is an acknowledgement that freedom of speech is more than the freedom to make noises: others need to be able to hear us. Freedom of speech is the freedom not just to make words but (to borrow from the philosopher J. L. Austin) to do things with those words.
Here is another loose thread. To be able to do anything with our words, we need not only to be heard but to make ourselves understood, in at least the minimal sense of that term. And how we are understood may be affected by, among other things, what people say about us (or people like us). Doesn’t speech about women, immigrants or benefits claimants – not to mention anarchists, Marxists, even social-democratic ‘Corbynistas’ – do this all the time, in numerous subtle and not-so-subtle ways? As Catharine MacKinnon has put it, ‘speech acts, acts speak’, and one of the things speech does is affect what others can do with their speech. Our sayings and doings constantly combine to restrict, as well as enable, each other’s speech. Our ability to ‘speak’ in the sense that matters – meaning not the mere production of words but the achievement of some further, inherently social act of communication – is in fact always dependent on power relations.
Seen this way, freedom of speech is not a default condition that prevails unless it is obstructed or suspended. Against this backdrop, we may still discuss the merits and demerits of different ways of managing speech, from censorship and libel law to boycotts or no-platform policies. But the point is that there is no coherent scenario of unrestricted speech, understood as the absence of interference. This is so at odds with how we are accustomed to think about freedom of speech that it will seem to many that it simply cannot be right, that it must be a form of hyperbole to say that speech can restrict speech. Yet if speech is about more than uttering words, it becomes possible in principle to speak of freedom of speech being infringed by interventions that limit what our speech can do. Caroline West makes this point vividly using a thought experiment: suppose a government were to install a chip in the brains of its citizens, enabling it to switch off those citizens’ ability to understand language whenever certain views are being expressed (or when members of certain groups are speaking); we would not hesitate to call this a restriction on free speech, regardless of the fact that, in this scenario, citizens can still say whatever they like – and even, albeit in a thin and useless sense, be ‘heard’.
Yet it might be argued that West’s example only counts as a case in which free speech is being infringed because a definite entity (the government) is intentionally blocking the understanding or ‘uptake’ (in Austin’s lexicon) of speech. But where agency is diffuse and cumulative (as, often, with speech about immigrants, women or Corbynistas), so that no particular intervention or agent can be identified as solely responsible, it cannot sensibly be said that anyone’s freedom of speech has been compromised. This reveals another hanging thread in the liberal understanding of free speech. The so-called ‘negative’ conception of liberty first distinguished by Isaiah Berlin identifies freedom with the absence of external restraint. As applied to speech, this negative conception holds that speech is free so long as it is not subject to any external restriction, such as a government ban, as opposed to an ‘internal’ restriction such as inarticulacy. So far, so common-sense.
But the line between being prevented and being unable is not so clear in practice. Someone who is too ‘inarticulate’ to make herself understood, for example, might alternatively be described as having been prevented (by education policy, say) from acquiring the skills that would render her speech intelligible and effective, or as being excluded by hegemonic standards of linguistic ‘correctness’. And why should external interference have to take the form of a one-off act by a clearly identifiable agent? If several companies pollute a river over many years, no single act can be blamed for the death of the fish (or the residents of a town downstream), but that does not make it tenable to claim that they merely ‘died’ from natural causes.
The distinction, central to the idea of negative liberty, between being impeded and being ‘merely’ unable, is often used to maintain the semblance of freedom in situations of powerlessness: nobody is stopping you from travelling – too bad you can’t afford the ticket; nobody is stopping you from sleeping indoors – too bad you can’t afford the rent. In the case of freedom of speech, the question becomes: why care only about whether we are ‘stopped’ from speaking, rather than asking what we can do about a society in which – in large part thanks to the concentration of wealth and access to the means of dissemination of opinion – some people are able to do a great deal with their words while others can do very little?
Those on the political right, along with most liberals, overwhelmingly purport to endorse the standard, narrowly negative conception of free speech. They typically have no truck with talk of megaphones and the marginalisation of certain groups and perspectives in the media, or with the idea that the speech of some might ‘silence’ that of others, all of which they regard as just more soft-brained snowflakery. Yet in the fray of the free speech wars, the orthodox notion of free speech underlying these responses is often forgotten or discarded. For the right, almost anything – criticism, protest, factual correction – can be construed as an attack on ‘free speech’. Suddenly, it seems, speech can silence after all. Something similar occurs on the left, too. Whereas the right operates with a narrow conception of free speech, but drops it when convenient for ideological combat, some on the left disdain the narrow conception, but then invoke it in battle with the right. In the face of the right’s claims to have been ‘silenced’ or ‘cancelled’, the reply is often: boohoo, so you had your book contract withdrawn / got called nasty names. Nobody’s actually stopping you from speaking . . .
This may draw attention to the right’s double standards, but the riposte risks falling into an inconsistency of its own and endorsing the dominant model of free speech as absence of restriction rather than effective power. On this alternative understanding of free speech, by contrast, pointing out that nobody has actually been prevented from speaking when they are ‘no-platformed’, ‘cancelled’ or simply criticised, cannot be the end of the story. If I find that, due to my criticism of Israel, many institutions and organisations are unwilling to employ me, invite me to speak, or publish what I write, then it would be entirely appropriate to see this as a threat to my and others’ freedom of expression: I cannot, in this scenario, speak freely on certain subjects without subjecting myself to an unacceptable risk of unacceptable penalties; and nor can others, to whom my fate may serve as an example of the costs of dissent. Even if I could eventually find a job or a publisher, this might or might not wholly restore my power to do things with my words, and it would not mean that this power had not been damaged in the first place.
Similarly, on this more capacious conception of free speech – understood not merely as the absence of external restrictions, but as the ability to intervene effectively in the world with words – the idea that certain kinds of critical speech might restrict their targets’ ability to communicate successfully is not inherently confused, mistaken or metaphorical. Speech can and does silence: for example, through lies, smears and distortions that can make it difficult or impossible for us to make ourselves heard. The upshot of this need not be sympathy for those styled as free speech martyrs, many of whom see their public profiles and general ability to do things with their speech grow. But the left response is on firmer ground when it highlights the disingenuousness of certain claims of silencing by working with an expanded conception of freedom of speech and insisting that the withholding of platforms to speak be seen in this wider context. That people sometimes become more influential as a result of persecution does not mean they were not really persecuted. Nevertheless, when a purported instance of silencing leads to its victim becoming a cause célèbre, this can tell us something about the position of that person or their views in relation to power (even if what it tells is not always straightforward).
Are no-platforming and other restrictive measures all right so long as they are cases of ‘punching up’ – deployed against the powerful and those with the loudest voices? This isn’t satisfactory, either. Flat-earthers are marginalised and ridiculed; is it wrong to deny them speaking invitations since this is ‘punching down’? Only at the cost of absurdity. Is free speech then a matter of getting the platforms you deserve? The notion that the power of speech should be proportionate to its merit sits uneasily with the way in which we conventionally think about this freedom, because it begins to blur the fêted distinction between how we evaluate the content of an expression and our defence of the liberty to express it – one summed up in the renowned gloss on Voltaire (often mischaracterised as a direct quotation), ‘I disapprove of what you say, but I will defend to the death your right to say it’. At the extreme, we end up saying that free speech is the freedom to say true things, or things judged to fall within the bounds of the (politically, morally or empirically) ‘reasonable’.
Yet if we instinctively reject that position, it’s not clear that the supposed content-neutrality of freedom of speech can survive scrutiny – nor that we really believe in it as much as we say we do. Is it a coincidence that those who defend the ‘free speech’ of those accused of racism tend to be those who are unconvinced that the speech is racist? Or that part of the point of defending ‘free speech’ on Israel is to resist charges of antisemitism regarded as cynically employed to suppress criticism? This pattern suggests that Voltaire’s separation, between how we evaluate speech and what we say about freedom, is not hard and fast. What we make of a given statement cannot only influence how we regard the prospect of the restriction of a speaker’s power to express it. It can also affect our judgement about whether a given response counts as a restriction, as an instance of ‘silencing’, in the first place.
Take smears. Clearly, the power of a person’s speech can be diminished if they are shunned and condemned, and this is so regardless of whether they really are a racist, a crank, or merely the victim of a witch-hunt. But it makes a difference whether the case is a smear, or one of legitimate criticism. This affects our view not only as to whether the loss of power is warranted, but also on the extent and nature of the loss. The notion of ‘uptake’ is central here. If my speech is misrepresented in a smear, then I may become unable to achieve basic uptake, to communicate my meaning effectively: my speech is received through a distorting filter, however hard I may try to circumvent it or compensate for it. By contrast, critical speech which exposes genuine racism (for example) does not block its reception – and might actually enhance the speech power of its target in a sense (albeit not one the target is likely to find pleasing), insofar as it contributes to that target’s speech being understood for what it is.
Trading the artificial simplicity and false neutrality of the liberal view for one attentive to the workings of power might seem like a risky move; the prospect of departing from the safety of liberal norms always gives rise to fever dreams of totalitarian bogeymen, even when those norms are in reality anything but safe for most. As with any nightmare, this is not an ex nihilo invention: it is true that the language of power relations and oppression, ‘silencing’ and ‘safety’, is sometimes invoked to justify practices which are at best strategically unwise and at worst politically and humanly destructive. Yet far from a lapse into authoritarianism being the price of transcending the liberal model of free speech, there are grounds for regarding the malaises associated with a politics of ‘safety’ as symptomatic of the uncritical embrace of liberal modes of thinking which centre on the sovereign individual, and which treat problems like racism and sexism as contaminants of individual souls.
Thinking about freedom of speech as an effective power does not have determinate practical implications, but that is not the same as entailing indeterminacy or committing us to a fuzzy, case-by-case approach. There may be good reasons, in various areas of life, to operate with firm, practical principles. But we need not – and have no good reason to, and have good reason not to – rest these on unexamined or misleading concepts. It may be that thinking about free speech with proper attention to questions of power leads us toward practical postures which resemble or overlap with the relatively ‘hardline’ positions advocated (at least in theory) by some on the right, by libertarians of right and left, as well as by some liberals. If anything, an approach appropriately attuned to power will be less willing than the liberal approach to countenance restrictions of speech by entities like the state, which, once no longer viewed through the idealising lens characteristic of liberal political philosophy, reveals itself as in no way to be trusted.
The apocryphal Voltaire should be supplemented with an extension of the apocryphal Emma Goldman: if speech changed anything, they’d make it illegal. But that liberal rights like the formal freedom of speech and the right to vote are hollow does not mean we don’t have every reason to protest their removal or erosion. You can deny that the UK, for example, is a ‘democracy’ in the full sense of the word, yet still object when the government disenfranchises people through voter ID laws. Equally, we cannot really be said to have freedom of movement in a country where many people simply cannot afford to go anywhere; but that doesn’t mean we shouldn’t object if the government attempted to enact coercive restrictions on our movement of the kind that Israel has long imposed on Palestinians. The point – which means both that erosions of formal freedoms are to be resisted and also that formal freedoms are not enough – is to bring closer a world in which we have real power over our lives, a world that seems ever further away.
Read on: Lorna Finlayson, ‘Rules of the Game?’, NLR 123.