Can words be ‘violent’?
‘Sticks and stones may break your bones, but words will never harm you!’ How many times did we hear that as children when the playground teacher intervened to break up a conflict that was about to boil over? Fast forward a few decades, and one wonders if these words of wisdom have been somewhat revised? The much-discussed question of whether words and speech are violence has re-emerged again. In the immediate aftermath of the assassination of Charlie Kirk in September 2025, there was a discussion on MSNBC about the death of this outspoken advocate of freedom of speech. Matthew Dowd, the Senior Political Analyst for MSNBC chose to defend the assassination of Charlie Kirk on the basis that he was ‘divisive’ and ‘pushing a kind of hate-speech’. He said that it always goes back to: ‘hateful thoughts lead to hateful words which then lead to hateful actions’. The perturbing thing about this syllogism was that he seemed to be implying that the utterance of certain words might justify an assassination. Had the culture wars really stooped so low?
In preparing this brief essay, I ended up listening to several hours of Charlie Kirk’s talks. What is clear is that Charlie Kirk did not use ‘hateful words’ (however we may wish to define them). And, we are not talking about verbal threats of violence or any indication that violence may have ensued. Charlie Kirk supported traditional values, freedom of speech and heterosexual relationships that lead to marriage and children. These values and the words he chose to couch them in could hardly be described as ‘hateful’ in any objective understanding of the word. We are instead concerned with an articulation of a set of views which are held to be ‘offensive’ by another group. What Matthew Dowd was implicitly saying was that Charlie Kirk deserved to die because his opinions were ‘offensive’ to some. In this essay, I don’t wish to discuss the comments and counter-comments that emerged after his tragic death. Instead, I want to delve into the question of whether words or speech can be inherently ‘violent’ as so many of Charlie Kirk’s opponents have claimed. This question has already been much discussed and by a number of respected scholars, but I am afraid to say that not only are their conclusions wrong but their analysis is flawed.
Presumably, one of the reasons why Dowd and those who share his views claim that ‘words are violent’ is because their understanding of violence is to ‘cause harm to a person’ as opposed to being an explicitly physical act. The problem with the ‘causing harm’ interpretation is that anyone can potentially claim that they have felt ‘harm’, that they have been ‘offended by someone’s words’. In the worst-case scenario where ‘causing harm’ is legislated against, you end up with a society where all kinds of discussion are closed down because people are frightened of being accused of ‘hate-speech’. Paradoxically, it results in a ‘closed’ society ostensibly founded on notions of openness but whose understanding of openness has been reconfigured to comply with woke norms. Dowd conflated conveniently ‘harm’ and ‘violence’, but even the use of the word ‘harm’ is suspect. Words and speech can obviously ‘offend’, but do they actually cause ‘harm’?
There is some evidence to suggest that words could cause ‘physical harm’ if they resulted in long-term stress for the hearer, but that doesn’t make words ‘forms of violence’. Feldman Barrett, a researcher on emotion, has explored thoroughly this topic. It is clear, I think, that the notion of ‘harmful speech’ is inextricably linked with the hyper-sensitivity culture that wokeism has brought about. If you consider speech to be ‘harmful’, then that is just one small step away from calling it ‘violent’ and if speech is ‘violent’, then you are more likely to be able to legislate against it because there is plenty of legislation that deals with physical violence. The purpose of the conflation is of course firstly to create an ideological environment that is favourable to those who wish to claim offence, and secondly to close down discussion that airs alternative views hic et nunc.
According to the Oxford English Dictionary, violence means: ‘the use of physical force intended to hurt, damage or kill someone or something’. Violence is then axiomatically physical, and thus to say that ‘words are violent’ is to make an invalid inference. In turn, if words are violent, then this flawed thinking might result in the view that ‘lexical’ violence can and should be met with physical violence and thus you arrive at Dowd’s veiled attempt at justifying the assassination of a free-speech advocate. All that Dowd had to say is that some people are offended by other peoples’ views which is banal and nothing new. That is how democracies work: different parties represent different political viewpoints and you vote for the one which most closely resembles your own.
It should by now be clear that there is a conventional (and largely shared) understanding of the meaning of language and there is an ideological version of what certain words mean. As an example, ‘hate’ is conventionally understood as having an intense dislike for someone or something. Or, at least that has been the dictionary definition of the word for decades, if not centuries. However, ‘hate’ is nowadays ideological speak because there is an ideological understanding of the word which corresponds to ‘a disrespectful attitude shown to a minority group’. Occasionally, there is leakage between the conventional and the ideological meaning of words and we end up in this case with ‘hate-speech legislation’ which has little to do with ‘speech that shows an intense dislike for someone or something’, but instead refers to speech by the ethnic majority aimed at the ethnic minority and that shows disrespect to the ethnic minority. This leakage is convenient for the woke ideologues because it means they cannot only demonise, but also legislate against those holding a view which challenges the hyper-sensitive interpretation of events. The conventional meaning of the word ‘hate’ has been corrupted, and it is frankly alarming that this kind of ideological code is readily employed in drafting parliamentary bills. There is no dictionary definition of the word ‘hate’ meaning ‘to show disrespect to an ethnic minority’. And note that legislation in the UK and elsewhere makes it explicit that only a representative of the ethnic majority can be guilty of hate-speech. This in itself should tell you that we are dealing with a piece of legislation that has arisen out of an ideological context. But, I digress…
In a discussion of how language functions as communication, Wittgenstein (Philosophical Investigations) said that ‘there must be agreement not only in definitions but also […] in judgements’. I would not disagree and I think it is clear that the majority of the people ‘judge’ violence to equate to physical harm. The minority that believe otherwise are, I suspect, largely the youth ‘educated’ on university campuses where they have been fed a propagandistic meaning of the word. Of course, we should not think of language in terms of permanence and fixity. Meanings of words change over time and it might be that in time the majority subscribe to the interpretation that ‘violence’ is synonymous with ‘offence’. One example of this kind of semantic shift might be the word ‘gay’. I can remember my grandmother saying ‘doesn’t the garden look gay?’ and she had no idea why her assertion was met with laughter.
Of course, communication alone cannot be considered a form of conduct. Words convey a message, but the context in which the words are delivered determines to what extent these words are enacted. This is the point that the linguistic philosopher, Austin (1962: 14-15), and subsequently Searle (1969: 65-7) kept emphasizing. Irony, jokes and banter all serve to muddy the waters too. In the case of Charlie Kirk, we are clearly not concerned with a context of verbal threats of violence, bullying or any other kind of veiled threat. Charlie Kirk went out of his way to listen and engage with people who held opposite views to him. He was considerate and respectful in all of the footage that I have watched.
Now, let me get to the crux of my argument. Feminists and critical theorists (Butler, 1997; Langton, 2019; Morrison etc.) tend to employ Austin’s (1962) theory of performativity to support the ideas that words can be ‘violent’. In her Nobel Prize lecture, Toni Morrison tells us ‘oppressive language does more than represent violence; it is violence.’ Austin developed the theory that ‘you can do things with words’ by which he meant certain set phrases produced in certain set contexts can bring about a concrete change in circumstances. So, when the priest at a wedding says ‘I hereby declare you man and wife’, then that is an Austinian speech act for this so-called ‘performative’ utterance has brought about a change in the circumstances of the couple. They are now married. It turns out that the critical theorists who advocate for the notion that ‘words and speech are violence’ have largely misunderstood Austin’s work or (more likely) manipulated what he wrote to suit their ideological goals.
‘Saying is doing’ does not apply to hate-speech, as so many of these critical theorists have implied. Austin had nothing to say about ‘hate-speech’ and would, I imagine, not approve of his theory being misread in this way. Hate-speech is not a performative act for the reason that it is not ‘felicitous’ to use Searle’s (1969) term. In order to be ‘felicitous’ or ‘successful’, a speech act such as in the example above needs to comply with the ‘preparatory condition’. That is to say, the speech act needs to be embedded in a conventionally recognised context (for example the wedding ceremony) where the listeners mutually agree on the ‘performativity’ of the words uttered. A ‘performative’ utterance that is not a speech act does not exist, as Austin tells us. So-called ‘hate-speech’ cannot possibly constitute a speech act because there is wide disagreement as to what specifically ‘hate-speech’ actually refers to and it is seen by many as an ideological approach to language which seeks to perceive utterances on a discriminatory basis. Moreover, the context in which hate-speech does or does not occur in no way reaches Austin’s definition of a conventionally recognised context.
Judith Butler and others like to exaggerate the agency of language and frame their arguments in Austin’s (1962) theory of performativity, but it seems have not in fact carefully read his works. Langton’s (2019) article published on the ABC Australia website is particularly ignorant in this regard. She starts her piece by saying, ‘when we speak, we do things with words’. No. That is not at all what Austin said. There are in fact relatively few bona fide instances of ‘doing things with words’ in the Austinian sense that she invokes here. A few examples might be: ‘I hereby name this ship the Queen Elizabeth’; I hereby declare this meeting adjourned’; ‘I hereby sentence you to 12 years in prison’; ‘I bequeath my car to my brother’. These are ‘performative’ Austinian speech acts because their utterance in the conventionally recognised context results in an immediate change of circumstances. Austin’s theory has been purposefully misconstrued so that it looks like the critical theorists’ argument has some kind of sound, theoretical underpinning when in fact it has nothing of the sort. In this very muddled piece by Langton, the author even claims that propaganda is hate-speech, but conveniently excludes from this very dubious claim her own rather radical propaganda. Needless to say, we should not engage in any such overdetermination of the ‘performative’ because language per se simply seldom enjoys that kind of agency.
Instead of ‘hateful thoughts lead to hateful words which then lead to hateful actions’, Charlie Kirk’s death could be summed up by saying: ‘open dialogue led to threats which led to murderous actions’. He would no doubt agree that ‘speech is not violence’. It should be borne in mind that the body is anterior to language, and language does not wield its own violence. We do many things with language: we please, displease, encourage, plunge into depression, offend, compliment but we don’t actually cause harm or violence through language alone in the way that these critical theorists claim. Language can only be a representation of violence. Any other interpretation represents an intellectual dishonesty to meet a certain ideological end. For speech to be considered ‘violent’, there would need to be a semantic corruption and that is what has been attempted with hate-speech legislation in Europe. The idea that speech can be violence is an attempt at deligitimising and silencing any critique of woke ‘values’, but people should not be fooled. Such ideologues think that speech can be hateful, but they are not universalist about their claims. It only applies to those who critique them.
The notion of hate-speech and the tendency to see language as ‘performative’ (either implicit or explicit) in the Austinian sense go hand-in-hand. So-called ‘hate-speech’ is perceived as being always ‘performative’ when in fact it is nothing of the sort. It should be obvious that this is a false premise because, as we have seen, juridically hate-speech is based on a unilateral assumption that only the ethnic majority can be guilty of this crime. If speech that challenges the woke agenda is ‘performative’, then words can be considered ‘violent’ and if speech is ‘violence’, then you can legislate against it. It is simply an attempt at framing the argument in woke tropes in order to pedal and universalise the ideologues’ own intolerance.
By ‘hateful’ words, Dowd meant undoubtedly words that ‘trigger’ such as Charlie Kirk’s assertion that there are only two sexes or that marriage and family make for a sound societal bed-rock. This is the problem with the word ‘hate’; its ideological meaning represents such a woolly dilution of its original, conventional meaning that the word has become rather meaningless. Students who have been ‘triggered’ (i.e. exposed to a view that they disagree with) speak of ‘hate’ and the speaker being ‘dangerous’. These claims are of course objectively absurd. They are simply saying in ideological speak that they strongly disagree and they would prefer the speaker to be silenced. Ironically, it is their ideological speak which is ‘dangerous’ for it has formed the basis of hate-speech legislation which is increasingly being used alongside quasi-censorship legislation (such as the Online Safety Bill in the UK) as a threat to silence people holding alternative (non-woke) views.
Those who promote the idea that words are violence are not only making an error in judgement, but are exploiting the notion that an alternative point-of-view can serve to enhance through metonymical slippage the putatively injurious power of words. It should be clear that this is a suspect argument because the proponents of this idea do not accept that their own ‘words are violent’. The idea that we should ascribe such efficacy to words is the product of the culture of hyper-sensitivity which tends to blur rational judgement because sentiment has to be sovereign. The prioritisation of sentiment over reason seems to leave the proponents of these views believing that they are justified in assuming a militant approach to these kinds of questions.
So, in conclusion: the notion that words are violence is based on a fundamental misreading of pragmatic theory. It is right to frame the objection in this way for the proponents of the notion that ‘words are violent’ couch their assumptions in these precise terms. Words are not violence and in a non-ideological context at least, there can be no such thing as hate-speech.
References
Austin, J. L. 1962. How to do things with words: the William James lectures delivered at Harvard University in 1955. London: Oxford University Press.
Butler, J. 1997. Excitable Speech: A politics of the performative. Routledge: New York; London.
Langton, R. 2019. Words that wound: Understanding the authority and effect of hate speech. https://www.abc.net.au/religion/the-authority-of-hate-speech/10478626
Searle, J. 1969. Speech acts: an essay in the philosophy of language. Cambridge: CUP.
‘Sticks and stones may break your bones, but words will never harm you!’ How many times did we hear that as children when the playground teacher intervened to break up a conflict that was about to boil over? Fast forward a few decades, and one wonders if these words of wisdom have been somewhat revised? The much-discussed question of whether words and speech are violence has re-emerged again. In the immediate aftermath of the assassination of Charlie Kirk in September 2025, there was a discussion on MSNBC about the death of this outspoken advocate of freedom of speech. Matthew Dowd, the Senior Political Analyst for MSNBC chose to defend the assassination of Charlie Kirk on the basis that he was ‘divisive’ and ‘pushing a kind of hate-speech’. He said that it always goes back to: ‘hateful thoughts lead to hateful words which then lead to hateful actions’. The perturbing thing about this syllogism was that he seemed to be implying that the utterance of certain words might justify an assassination. Had the culture wars really stooped so low?
In preparing this brief essay, I ended up listening to several hours of Charlie Kirk’s talks. What is clear is that Charlie Kirk did not use ‘hateful words’ (however we may wish to define them). And, we are not talking about verbal threats of violence or any indication that violence may have ensued. Charlie Kirk supported traditional values, freedom of speech and heterosexual relationships that lead to marriage and children. These values and the words he chose to couch them in could hardly be described as ‘hateful’ in any objective understanding of the word. We are instead concerned with an articulation of a set of views which are held to be ‘offensive’ by another group. What Matthew Dowd was implicitly saying was that Charlie Kirk deserved to die because his opinions were ‘offensive’ to some. In this essay, I don’t wish to discuss the comments and counter-comments that emerged after his tragic death. Instead, I want to delve into the question of whether words or speech can be inherently ‘violent’ as so many of Charlie Kirk’s opponents have claimed. This question has already been much discussed and by a number of respected scholars, but I am afraid to say that not only are their conclusions wrong but their analysis is flawed.
Presumably, one of the reasons why Dowd and those who share his views claim that ‘words are violent’ is because their understanding of violence is to ‘cause harm to a person’ as opposed to being an explicitly physical act. The problem with the ‘causing harm’ interpretation is that anyone can potentially claim that they have felt ‘harm’, that they have been ‘offended by someone’s words’. In the worst-case scenario where ‘causing harm’ is legislated against, you end up with a society where all kinds of discussion are closed down because people are frightened of being accused of ‘hate-speech’. Paradoxically, it results in a ‘closed’ society ostensibly founded on notions of openness but whose understanding of openness has been reconfigured to comply with woke norms. Dowd conflated conveniently ‘harm’ and ‘violence’, but even the use of the word ‘harm’ is suspect. Words and speech can obviously ‘offend’, but do they actually cause ‘harm’?
There is some evidence to suggest that words could cause ‘physical harm’ if they resulted in long-term stress for the hearer, but that doesn’t make words ‘forms of violence’. Feldman Barrett, a researcher on emotion, has explored thoroughly this topic. It is clear, I think, that the notion of ‘harmful speech’ is inextricably linked with the hyper-sensitivity culture that wokeism has brought about. If you consider speech to be ‘harmful’, then that is just one small step away from calling it ‘violent’ and if speech is ‘violent’, then you are more likely to be able to legislate against it because there is plenty of legislation that deals with physical violence. The purpose of the conflation is of course firstly to create an ideological environment that is favourable to those who wish to claim offence, and secondly to close down discussion that airs alternative views hic et nunc.
According to the Oxford English Dictionary, violence means: ‘the use of physical force intended to hurt, damage or kill someone or something’. Violence is then axiomatically physical, and thus to say that ‘words are violent’ is to make an invalid inference. In turn, if words are violent, then this flawed thinking might result in the view that ‘lexical’ violence can and should be met with physical violence and thus you arrive at Dowd’s veiled attempt at justifying the assassination of a free-speech advocate. All that Dowd had to say is that some people are offended by other peoples’ views which is banal and nothing new. That is how democracies work: different parties represent different political viewpoints and you vote for the one which most closely resembles your own.
It should by now be clear that there is a conventional (and largely shared) understanding of the meaning of language and there is an ideological version of what certain words mean. As an example, ‘hate’ is conventionally understood as having an intense dislike for someone or something. Or, at least that has been the dictionary definition of the word for decades, if not centuries. However, ‘hate’ is nowadays ideological speak because there is an ideological understanding of the word which corresponds to ‘a disrespectful attitude shown to a minority group’. Occasionally, there is leakage between the conventional and the ideological meaning of words and we end up in this case with ‘hate-speech legislation’ which has little to do with ‘speech that shows an intense dislike for someone or something’, but instead refers to speech by the ethnic majority aimed at the ethnic minority and that shows disrespect to the ethnic minority. This leakage is convenient for the woke ideologues because it means they cannot only demonise, but also legislate against those holding a view which challenges the hyper-sensitive interpretation of events. The conventional meaning of the word ‘hate’ has been corrupted, and it is frankly alarming that this kind of ideological code is readily employed in drafting parliamentary bills. There is no dictionary definition of the word ‘hate’ meaning ‘to show disrespect to an ethnic minority’. And note that legislation in the UK and elsewhere makes it explicit that only a representative of the ethnic majority can be guilty of hate-speech. This in itself should tell you that we are dealing with a piece of legislation that has arisen out of an ideological context. But, I digress…
In a discussion of how language functions as communication, Wittgenstein (Philosophical Investigations) said that ‘there must be agreement not only in definitions but also […] in judgements’. I would not disagree and I think it is clear that the majority of the people ‘judge’ violence to equate to physical harm. The minority that believe otherwise are, I suspect, largely the youth ‘educated’ on university campuses where they have been fed a propagandistic meaning of the word. Of course, we should not think of language in terms of permanence and fixity. Meanings of words change over time and it might be that in time the majority subscribe to the interpretation that ‘violence’ is synonymous with ‘offence’. One example of this kind of semantic shift might be the word ‘gay’. I can remember my grandmother saying ‘doesn’t the garden look gay?’ and she had no idea why her assertion was met with laughter.
Of course, communication alone cannot be considered a form of conduct. Words convey a message, but the context in which the words are delivered determines to what extent these words are enacted. This is the point that the linguistic philosopher, Austin (1962: 14-15), and subsequently Searle (1969: 65-7) kept emphasizing. Irony, jokes and banter all serve to muddy the waters too. In the case of Charlie Kirk, we are clearly not concerned with a context of verbal threats of violence, bullying or any other kind of veiled threat. Charlie Kirk went out of his way to listen and engage with people who held opposite views to him. He was considerate and respectful in all of the footage that I have watched.
Now, let me get to the crux of my argument. Feminists and critical theorists (Butler, 1997; Langton, 2019; Morrison etc.) tend to employ Austin’s (1962) theory of performativity to support the ideas that words can be ‘violent’. In her Nobel Prize lecture, Toni Morrison tells us ‘oppressive language does more than represent violence; it is violence.’ Austin developed the theory that ‘you can do things with words’ by which he meant certain set phrases produced in certain set contexts can bring about a concrete change in circumstances. So, when the priest at a wedding says ‘I hereby declare you man and wife’, then that is an Austinian speech act for this so-called ‘performative’ utterance has brought about a change in the circumstances of the couple. They are now married. It turns out that the critical theorists who advocate for the notion that ‘words and speech are violence’ have largely misunderstood Austin’s work or (more likely) manipulated what he wrote to suit their ideological goals.
‘Saying is doing’ does not apply to hate-speech, as so many of these critical theorists have implied. Austin had nothing to say about ‘hate-speech’ and would, I imagine, not approve of his theory being misread in this way. Hate-speech is not a performative act for the reason that it is not ‘felicitous’ to use Searle’s (1969) term. In order to be ‘felicitous’ or ‘successful’, a speech act such as in the example above needs to comply with the ‘preparatory condition’. That is to say, the speech act needs to be embedded in a conventionally recognised context (for example the wedding ceremony) where the listeners mutually agree on the ‘performativity’ of the words uttered. A ‘performative’ utterance that is not a speech act does not exist, as Austin tells us. So-called ‘hate-speech’ cannot possibly constitute a speech act because there is wide disagreement as to what specifically ‘hate-speech’ actually refers to and it is seen by many as an ideological approach to language which seeks to perceive utterances on a discriminatory basis. Moreover, the context in which hate-speech does or does not occur in no way reaches Austin’s definition of a conventionally recognised context.
Judith Butler and others like to exaggerate the agency of language and frame their arguments in Austin’s (1962) theory of performativity, but it seems have not in fact carefully read his works. Langton’s (2019) article published on the ABC Australia website is particularly ignorant in this regard. She starts her piece by saying, ‘when we speak, we do things with words’. No. That is not at all what Austin said. There are in fact relatively few bona fide instances of ‘doing things with words’ in the Austinian sense that she invokes here. A few examples might be: ‘I hereby name this ship the Queen Elizabeth’; I hereby declare this meeting adjourned’; ‘I hereby sentence you to 12 years in prison’; ‘I bequeath my car to my brother’. These are ‘performative’ Austinian speech acts because their utterance in the conventionally recognised context results in an immediate change of circumstances. Austin’s theory has been purposefully misconstrued so that it looks like the critical theorists’ argument has some kind of sound, theoretical underpinning when in fact it has nothing of the sort. In this very muddled piece by Langton, the author even claims that propaganda is hate-speech, but conveniently excludes from this very dubious claim her own rather radical propaganda. Needless to say, we should not engage in any such overdetermination of the ‘performative’ because language per se simply seldom enjoys that kind of agency.
Instead of ‘hateful thoughts lead to hateful words which then lead to hateful actions’, Charlie Kirk’s death could be summed up by saying: ‘open dialogue led to threats which led to murderous actions’. He would no doubt agree that ‘speech is not violence’. It should be borne in mind that the body is anterior to language, and language does not wield its own violence. We do many things with language: we please, displease, encourage, plunge into depression, offend, compliment but we don’t actually cause harm or violence through language alone in the way that these critical theorists claim. Language can only be a representation of violence. Any other interpretation represents an intellectual dishonesty to meet a certain ideological end. For speech to be considered ‘violent’, there would need to be a semantic corruption and that is what has been attempted with hate-speech legislation in Europe. The idea that speech can be violence is an attempt at deligitimising and silencing any critique of woke ‘values’, but people should not be fooled. Such ideologues think that speech can be hateful, but they are not universalist about their claims. It only applies to those who critique them.
The notion of hate-speech and the tendency to see language as ‘performative’ (either implicit or explicit) in the Austinian sense go hand-in-hand. So-called ‘hate-speech’ is perceived as being always ‘performative’ when in fact it is nothing of the sort. It should be obvious that this is a false premise because, as we have seen, juridically hate-speech is based on a unilateral assumption that only the ethnic majority can be guilty of this crime. If speech that challenges the woke agenda is ‘performative’, then words can be considered ‘violent’ and if speech is ‘violence’, then you can legislate against it. It is simply an attempt at framing the argument in woke tropes in order to pedal and universalise the ideologues’ own intolerance.
By ‘hateful’ words, Dowd meant undoubtedly words that ‘trigger’ such as Charlie Kirk’s assertion that there are only two sexes or that marriage and family make for a sound societal bed-rock. This is the problem with the word ‘hate’; its ideological meaning represents such a woolly dilution of its original, conventional meaning that the word has become rather meaningless. Students who have been ‘triggered’ (i.e. exposed to a view that they disagree with) speak of ‘hate’ and the speaker being ‘dangerous’. These claims are of course objectively absurd. They are simply saying in ideological speak that they strongly disagree and they would prefer the speaker to be silenced. Ironically, it is their ideological speak which is ‘dangerous’ for it has formed the basis of hate-speech legislation which is increasingly being used alongside quasi-censorship legislation (such as the Online Safety Bill in the UK) as a threat to silence people holding alternative (non-woke) views.
Those who promote the idea that words are violence are not only making an error in judgement, but are exploiting the notion that an alternative point-of-view can serve to enhance through metonymical slippage the putatively injurious power of words. It should be clear that this is a suspect argument because the proponents of this idea do not accept that their own ‘words are violent’. The idea that we should ascribe such efficacy to words is the product of the culture of hyper-sensitivity which tends to blur rational judgement because sentiment has to be sovereign. The prioritisation of sentiment over reason seems to leave the proponents of these views believing that they are justified in assuming a militant approach to these kinds of questions.
So, in conclusion: the notion that words are violence is based on a fundamental misreading of pragmatic theory. It is right to frame the objection in this way for the proponents of the notion that ‘words are violent’ couch their assumptions in these precise terms. Words are not violence and in a non-ideological context at least, there can be no such thing as hate-speech.
References
Austin, J. L. 1962. How to do things with words: the William James lectures delivered at Harvard University in 1955. London: Oxford University Press.
Butler, J. 1997. Excitable Speech: A politics of the performative. Routledge: New York; London.
Langton, R. 2019. Words that wound: Understanding the authority and effect of hate speech. https://www.abc.net.au/religion/the-authority-of-hate-speech/10478626
Searle, J. 1969. Speech acts: an essay in the philosophy of language. Cambridge: CUP.
