Direct violence

The predominant usage of the concept of violence in IR is to mean either direct violence, or violence as the equivalent of military force. This is especially the case for those texts that use the concept only rarely.

Several traditional IR texts almost completely avoid using the concept of violence, even when they are discussing phenomena which clearly come under the rubric of violence. When they do use the concept, it often appears to be used interchangeably with the concepts they are using, whether that be ‘military force’, ‘war’ and so on. A clear example is Kenneth Waltz in Man, the State and War, the theme of which is the causes of war, a clear example of violence.Footnote ⁹ In this book Waltz only uses the word violence 11 times. In many of these cases the word is used to mean the same thing as inter-state war. For example: ‘If violence among states is caused by the evilness of man, to aim at the internal reform of states will not do much good. And if violence among states is the product of international anarchy, to aim at the conversion of individuals can accomplish little.’Footnote ¹⁰

Morgenthau also uses the concept of violence only relatively rarely in Politics Among Nations.Footnote ¹¹ Again, he uses it predominantly to mean direct violence. In most cases, however, Morganthau feels the need to specify that he is referring to physical violence, perhaps an acknowledgement of the contested nature of the concept. For example: ‘Political power, however, must be distinguished from force in the sense of the actual exercise of physical violence. The threat of physical violence in the form of police action, imprisonment, capital punishment, or war is an intrinsic element of politics.’Footnote ¹² One of the reasons Morgenthau refers to violence only rarely is because he draws a line between the threat of violence and its actual use. He discusses the threat predominantly, without getting into the unsavoury aspects of the actual use of violence by a state. The whole topic sounds very clinical in this text. For example, Morgenthau writes ‘The political aim of military preparations is […] to make the actual application of military force unnecessary by inducing the prospective enemy to desist from the use of military force’.Footnote ¹³

Bull is an exception to this trend, as he uses the concept of violence often. He also uses the concept predominantly to refer to direct violence, both by state and non-state actors. In this way, he is a good example of the use of the concept of violence as advocated in this article. He confines his understanding of violence to a narrow, instrumental definition, yet does not hide from using the concept, and thus the harm that is caused to individuals when violence is used. At the same time, he does not fall into the trap of using violence only for illegitimate actors. An example illustrates this: ‘it is clear that the monopoly of legitimate international violence long enjoyed by sovereign states is being challenged on the one hand by non-state political groups employing so-called “low-level” or “terrorist” violence on an international scale, and on the other hand by the assumption by international organisations of a right to use violence’.Footnote ¹⁴ In this example Bull uses the concept of violence equally, regardless of who the actor is. One can also assume that he is referring to direct, intentional, instrumental violence.

These few examples have shown that, whilst many traditional IR texts do not use the concept of violence often, where they do use it, it tends to be violence of a direct nature. Although this is a question of interpretation, the context within which the concept of violence is used in these texts, including the substitution for the terms ‘military force’ or ‘war’, make it clear that these texts are referring to direct violence. The fact that the concept of violence is used only rarely is not necessarily a criticism, as some authors use more specific terms, for example war. War is a specific form of violence, but not all violence is war. Being more specific with ones terminology is a positive thing. However this is not always the case. Sometimes less specific terms are used, and these cases can be accused of euphemism, or of constructing a discourse where violence in international politics is perceived as a normal part of the functioning of the international system.

Violence as illegitimate

Before discussing the way that ‘violence’ is used to categorise something as legitimate or illegitimate, the meaning of legitimacy needs to be determined. Of course, this is another contested concept, and one which is used to refer to many different things. It draws ‘on concepts of (moral and epistemic) right, legality, custom, tradition and popular approval’.Footnote ¹⁵ In the end, it refers to many different things, and is impossible to define simply. What is clear with the concept of legitimacy is that it conveys a meaning of approval, and this is key to the meaning of legitimacy being used here.Footnote ¹⁶ The aim is not to describe what is legal, but rather to convey the idea that something is perceived (or socially constructed) as the right, or justified thing to do.

Often, ‘violence’ is used only to refer to its use by non-state actors, or those the author does not approve of. In much of the IR literature investigated here there is a consistent approach, as although there are references to illegitimate or non-state actors using violence, state actors are also referred to on occasion as using violence. However, one could question why it is that traditional IR scholars in particular rarely use the concept of violence. It has to be noted that they predominantly discuss the use of violence between states – ‘legitimate’ actors, and thus if violence were to be restricted to illegitimate cases, then that could explain the infrequent use of the concept.

A good example of this is Mearsheimer, The Tragedy of Great Power Politics.Footnote ¹⁷ This text does not use the term violence at all, despite the fact that there is much discussion of the need to use ‘military force’, including many case studies of states that have done so.Footnote ¹⁸ Rather, the violence between states is portrayed in such a way as it becomes clear that this is legitimate action, and we cannot doubt that it is typical and justifiable state behaviour. For example ‘Japan took advantage of almost every favourable shift in the balance of power to act aggressively’,Footnote ¹⁹ Italy ‘focused its aggressive intentions on five different areas’,Footnote ²⁰ and ‘Stalin was also determined to conquer territory’.Footnote ²¹ This does not mean that there is no recognition within this text of the brutality of violence. But it is an interesting example that within a text of this size, about the subject of violence, the word violence is not used. This at least hints at the fact that ‘violence’ is seen as illegitimate, and the use of military force between states is seen as legitimate. After all, the purpose of the text is to set out the case for offensive realism, which claims that great powers attempt to maximise their power through offensive strategies, and that war is the main strategy used to do this.Footnote ²² This is borne out by the fact that one of the very rare occasions that Mearsheimer uses the concept of violence (in a separate article) is in discussing ethnic violence in the former Yugoslavia, where he refers to ‘the interethnic violence’. This is contrasted in the same article with ‘the Spring 1999 military offensive’, referring to ‘legitimate’ violence.Footnote ²³ Snyder does a similar thing when he refers to violence by non-state actors, but a ‘military response’ by state actors.Footnote ²⁴ For example he refers to ‘the violence that is vexing the experiment with democracy in Iraq’.Footnote ²⁵

This way of ‘hiding’ state violence as the justified way of countering illegal, private violence has been linked to modernity, dating back to at least the seventeenth century, so is not a new phenomenon with IR scholars.Footnote ²⁶ This link between modern thought and the justification of violence for the spread of modernity and civilisation has been widely recognised as damaging.Footnote ²⁷ This is still important today as modern Western thought continues to influence what seems to be common sense, and distinctions between civilisation and barbarism continue to inform the distinctions we make between legitimate and illegitimate violence.Footnote ²⁸ There are prominent examples of this sort of discourse continuing. Meyer points to examples of Tony Blair using the danger of the ‘uncivilised other’ to argue that Western states need to spread freedom and democracy by sending in military forcesFootnote ²⁹ – in other words to stop the violence of the extremists by using military force (‘justified’ violence). Meyer comments, ‘[s]ince military action is performed in the name of the common good of humanity, it must not be represented as an act of violence, but celebrated as an intervention that counters violence and war’.Footnote ³⁰ In this way acts of state violence perform an ordering function in the interests of the powerful and Western values, and as such are presented as positive.Footnote ³¹

This has an impact on the way that actors behave in international politics. Coleman, for example, points out how the legitimisation of state violence is reproduced in development policy in Columbia. In its National Development Plan 2002–2006, Columbia represents criminals as engaging in violence, but not the state engaged in (violent) repression: ‘Violence is thereby set apart from the state. What the state does is not violence: violence is its other.’Footnote ³² This is not only important for the way we discuss and understand international politics, but the naming of violence in this way constitutes it as such, so that non-state actors become violent and state actors are non-violent, regardless of how similar their actual actions are.

These examples have shown that although many traditional scholars use the concept of violence to mean direct violence, and use it to refer both to legitimate and illegitimate cases, there are some who prefer to leave the usage of violence for cases which the author perceives to be illegitimate. This is problematical as it constructs a discourse of legitimacy around all state violence, and it hides the violence of state action.

State of violence

Violence can also be used in IR to refer to a state of violence, as the opposite of peace or order. This is violence where there is no actor, but where the general state of the world or a region and so on is said to be violent. Although this interpretation of the concept of violence is less common in the literature, one could put the interpretation on some statements. For example, the quote used to illustrate Waltz's use of the direct violence concept above could also be interpreted as referring to a general state of affairs, with no actor – ‘violence among states’ does not necessarily refer to specific violent actions by states, but could refer to a state of affairs. Waltz states that ‘in the state of nature conflict and violence reign’.Footnote ³³ If this refers to the fact that in the state of nature actors are always using violent tactics, then we can accept this as direct violence. But if we interpret it as nebulous actors with indeterminate actions in a state that seems violent, or as a way to characterise disorder, then this interpretation is less useful. It blurs the distinction between the actual use of violence and the potential use, fear or threat of violence.

One could also interpret Hedley Bull's use of the concept of violence as a state which is the opposite of order. After all, his Anarchical Society is primarily a treatise on order, and violence is often referred to as the opposite of order, for example ‘The liberation of African and Asian peoples from European empires has been accompanied by violence and disorder’.Footnote ³⁴ However, Bull's consistent approach of referring to direct violence, regardless of the actor, as violence, leaves little room for doubt that he means direct violence in each case, and not a state of affairs.

Although violence as a state of nature is not an interpretation used as often as direct violence, it is still one that is used on occasion. It is quite possible that authors often use the term in this way when they are referring to many acts of violence, but it is not always clear that this is the case, and it can also be interpreted as a state of disorder, or more generally a state of affairs that the author does not approve of.

This analysis of how traditional IR uses the concept of violence has shown that authors most often use the concept of violence to mean direct violence. Scholars tend to use the concept only rarely, and there remains a question about whether they avoid the concept due to its implication of illegitimacy when they are referring to the violence used by states which they perceive as being legitimate.

The normative problem

Some authors claim that the concept of violence is not useful due to its normative content.Footnote ³⁵ Although most people will agree that violence should be condemned, the problem comes when one asks whether all violence should be equally condemned. Most people will also believe that some form of violence can be justified in certain situations. Violence that is perceived to be legitimate is often referred to in a different way, leaving the term violence to imply illegitimacy. For example, the use of violence by ‘legitimate’ state troops is often referred to as ‘force’. In effect, this is merely a euphemism for violence. But for another person, military violence is never or rarely justified. This person is more likely to refer to the same thing as violence rather than force. This problem is very similar to the problem with using the word terrorism. Whereas it is easy to label Al-Qaeda as using terrorist tactics, one will find more resistance to labelling a liberal democratic state's action as terrorist, even if its methods and aims fit within a typical definition of terrorism. In the same way, the description of Al-Qaeda's 9/11 attacks on the twin towers as violence is common, whereas the description of US attacks on Afghanistan or Iraq is often termed as force. Choosing the word force implies a certain legitimacy for the act.

It has been claimed that the term violence is simply applied to whatever the person using the term does not like.Footnote ³⁶ Betz calls it a

For the purposes of academic analysis of violence, this type of rhetoric is not useful, and uses that aim to label something as wrong by applying the label of violence need to be highlighted. This is not to say that we can have an objective definition of violence, as normative values will always affect our understanding. But the normative implications of the use of the term need highlighting, and we need a definition that allows some consistency of use. Although violence has a negative connotation, this is not as strong as the values that are apportioned to some concepts. For example, it is not easy to assert that legitimacy is anything other than positive, and yet the idea of legitimate violence is not an oxymoron. Thus, although violence conveys a normative element of illegitimacy, this is not inherent or essential, and is dependent on context. It is possible to discuss the core meaning of violence upon which notions of legitimacy can lie.

Violence and legitimacy

There are some definitions of violence that are formed to a high degree on normative values, and which are weak because of this. For example, Wolff and Honderich see violence as illegitimate, but the use of force by a legitimate authority as not violent.Footnote ³⁸ For Wolff, this means that the whole concept of violence is incoherent as his anarchist viewpoint also sees legitimate authority as incoherent.Footnote ³⁹ There are some problems with this. This concept of legitimacy is based only on the legitimacy of the actor. The concept of legitimacy is widely contested. The view that it is only about actors often becomes conflated with the assumption that the only legitimate actors are states. However there will always be disagreement about what authorities, or what uses of violence, are legitimate. For example, a non-violent protestor may see a police officer's use of force as illegitimate and therefore violent. But a government official may see the same police officer's actions as a legitimate use of force to maintain order (and possibly to stop the protest turning ‘violent’).

Some concepts of legitimacy restrict it to actors. The legitimacy of an actor is often measured by its democratic credentials, representativeness, moral authority and so on, and the legitimacy of an act is often measured according to the legitimacy of the actor.Footnote ⁴⁰ Sometimes legitimacy is assumed to belong to states and only to those non-state actors that the state decides to grant it to.Footnote ⁴¹ But legitimacy is more complex than this and can include the perception or assumption that an action or actor is desirable or appropriate, the belief that a rule or institution should be obeyed or conforms to the law, that something has received wide approval or external recognition.Footnote ⁴² It is socially constructed and varies according to time and place.Footnote ⁴³ Legitimate violence will often be seen as being carried out by a democratic state, and be perceived as being legal. This idea that violence is legitimately carried out by states has a long history, including Weber's claim that states have the monopoly on legitimate violence, and just war theory. However violence carried out in the name of self-determination, representative of the desires of a people group in that territorial location, may also be perceived as legitimate. Liberation struggles like the African National Congress (ANC) in South Africa have been widely perceived as being legitimate. One could argue that once successful, uses of non-state violence become legitimate.Footnote ⁴⁴ There is also a view held by many that it is state violence that is illegitimate, and violence used to stop the oppression of the state is legitimate.Footnote ⁴⁵

The analysis of legitimacy based on the actor alone is incomplete, and not helpful for a definition of violence. Legitimacy is a much broader concept that also can be applied to ‘a resource to be held or traded, a perception to be described, an authoritative judgement, or a tool of imperialistic exclusion’ and so on.Footnote ⁴⁶ Using the legitimacy of an actor in order to determine whether something is violent not only simplifies the concept of legitimacy, but also unhelpfully narrows debate about a subject by differentiating between actors that can be doing the same thing. This can have an impact on the way we understand and act in the political world. If we think that only non-state actors are capable of doing violence, they are treated and understood in completely different ways than state actors who can be doing the same thing – this unnecessarily restricts the way we think about and conceive of these acts and plays a role in constituting some acts as violence and not others.Footnote ⁴⁷

By bringing the concept of legitimacy into the definition of violence, the term becomes a political tool. Governments are able to call something violent in order to condemn it, at the same time as using violence themselves to maintain the status quo where they have the power. Garver points out that President Johnson was telling the civil rights movement in the US that they could not win by violence at the same time as he was escalating violence in Vietnam.Footnote ⁴⁸ Presumably he would have used the euphemistic term force for the Vietnam War instead of violence. In a similar way, there are people who currently support the ‘War on Terror’ in order to stop the ‘violence’ of terrorism, often without naming the violence being used within the War on Terror. For example, a recent article in the Financial Times said both that ‘Not all Muslims – even among those prepared to use violence in pursuit of their cause – think alike’, and that ‘There are plenty of dangerous Islamists for whom the only response will be military force.’Footnote ⁴⁹ The presumption is that the terrorists use violence and are illegitimate, and state authorities are legitimate and so use military force.

Violence versus peace and order

Concentrating on the legitimacy of the actor engaging in violence is not the only way that normative values enter into the definition of ‘violence’. Some definitions also bring in normative values by contrasting violence with peace or with order. Galtung contrasts violence with peace in his famous article on the subject, Violence, Peace and Peace Research, which is aimed at setting out a research agenda for peace research.Footnote ⁵⁰ Although the majority of the article is spent articulating what Galtung understands by the concept of violence, the aim is in order to understand the term peace. Everything that is violent is the opposite of peace. Galtung sets out three principles to discuss the idea of peace, the third being ‘The statement peace is absence of violence shall be retained as valid’.Footnote ⁵¹ There are two problems with this approach. Firstly, the fact that violence and peace are not opposites in this way, and secondly the fact that Galtung uses his definition of violence to be able to encompass within his definition of peace the things he wants to be able to discuss in peace research. For Galtung, peace research should not only be about ending direct violence, but also ending social injustice.

In the case of the first problem, that it is wrong to conceptualise violence as the opposite of peace, Galtung clearly wants to establish the definition of peace as not only the absence of direct violence, but also the presence of social justice (or absence of structural violence – these two terms are interchangeable for Galtung). As I show below in discussing structural violence, by broadening the concept of violence to include social justice, Galtung makes the term incoherent. With a traditional view of peace, the problem with comparing peace and violence is that there are times when the two can be coexisting, or when neither is present. Acts of violence in international politics can be perpetrated during peacetime as well as during wartime, for example terrorist attacks or short campaigns not classed as war. There are also acts of violence that occur during peacetime, for example violence against women in the home, acts of murder, which under the usual conception of peace do not mean an end to peace. At the same time, two actors can technically be at war with each other, and therefore not at peace, and yet not be perpetrating any acts of violence. This could happen, for example, where a state instigates a blockade against another state. This is technically a declaration of war, but is not inherently violent. Galtung wants to get away from this conception of peace, a debate not within the scope of this article, but I would argue that one act of violence does not necessarily end the state of peace, although it brings a source of conflict into the peace. This conception of peace and violence highlights the question of whether violence can be a state of affairs, or a condition of world order. I argue below that violence is an instrumental act. As such it cannot be a state of affairs. If we were to agree that violence is an instrument it cannot be the opposite of peace which is a condition of world order.

The second problem with Galtung's conception of violence purely in reference to peace is that he ends up building into the concept of violence the opposite of everything he wants to be able to build into his conception of peace. This is in danger of leading to the approach discussed above of including everything that we do not like. The function of this definition of violence is primarily to provide a definition of peace, and one which enables a research project within peace research which encompasses both direct violence and social injustice, one which enables the researcher to look at both direct violence and development. Galtung backs this up by stating that a narrow definition of violence is not suitable as ‘If this were all violence is about, and peace is seen as its negation, then too little is rejected when peace is held up as an ideal.’Footnote ⁵² In some ways this is a creditable aim. There are clearly links between violence and social justice or development, and it is important that these links are recognised and researched. But keeping the labels separate enables greater clarity when looking at links between the two. Galtung recognises this point to some degree when he comments that instead of referring to the ‘absence of personal violence, and absence of structural violence […] For brevity the formulations “absence of violence” and “social justice” may perhaps be preferred’.Footnote ⁵³

Normative values are also brought into the definition of violence when it is defined as the opposite of order.Footnote ⁵⁴ In this approach, order is seen as a positive thing that should be sought after. However, order can also be seen as the maintenance of the status quo, rather than the development of a just and good society. It depends on one's priorities as to which one prefers. Many actors turn to violence in the belief that the resulting disorder is a valid price to pay if justice results. Civil wars often begin in order to change the status quo, to change the existing order.

So, normative values are introduced into the concept of violence by using legitimacy, or contrasting violence with peace or order. Whilst it is impossible to discuss politics, and the use of violence, without introducing ones values into the discussion, it is also important that we recognise and highlight where these values are. If we recognise the implications of the words and definitions we use, we can also make attempts to question our assumptions and question the impact these assumptions have on the way we think about the world and act in it.

The instrumental meaning of violence

One of the most common discussions about violence centres on means and ends. A good social end is sometimes said to justify the means of violence. An example was the argument that the US and UK governments should go to war with Iraq in 2003 in order to stop Saddam Hussein's atrocities. Gandhi on the other hand stated that the means must be consistent with the ends, as the means will affect the ends. This is a key argument for non-violence. It is important that violence is seen as a means and not an end. The use of violence is not the end position, but an instrument in order to achieve a certain aim.Footnote ⁵⁵ The aim is not to artificially rule out of debate an aspect that should or could be considered, or to artificially simplify the concept. Rather, it is to recognise that violence as an instrument is the core meaning of the concept, and stretching it to incorporate a state of violence brings in a vague concept unnecessarily. As I argued earlier, the concept of violence as a state of nature often refers to many acts of direct violence. Where it does not, it usually refers to a state that the author does not like, for example disorder.

It could be argued that a state of violence exists when there is a culture of using violence, or many acts of violence make violence ubiquitous. In this case the state of violence is no different to the instrumental meaning, in that the state of violence is referring to an aspect of instrumental violence – its frequent occurrence and so on. In the same way, one could argue that violence is part of human nature, and therefore we can talk about the state of violence within humankind. I would argue that it is true that violence can be a learned pattern of action, or a culture within society. But the act of violence still is perpetrated by an actor, who at some level has a choice about whether to engage in violence, and why. It remains important to restrict the concept of violence to mean an instrument. It is an instrument that can be used to gain many things, from the oppression of another person, to a general state of world disorder, but violence is always a means aimed at achieving a certain aim.

Frazer and Hutchings argue that it is not possible to separate ‘violent doing from violent being’ and that doing so ‘underestimates the levels of ideological and material investment needed to sustain violence as a repertoire for political action’ and overestimates the ability of actors to transcend the use of violence.Footnote ⁵⁷ This view should not be discounted. It is certainly important to recognise the structures and beliefs that allow the use of violence to continue so easily. However, this does not need to mean that violence is other than instrumental. Recognising the structures, institutions and beliefs that perpetuate violence can be done at the same time as recognising that violence is an act that is carried out by an actor. Although it should also be recognised that in the real world these links will be complex and the borderlines of violence may be fuzzy.

Violence cannot be compared to world peace, or social justice, as they are completely different – they are the end state towards which one may choose to aim. The study of violence in international politics should treat the concept as an instrument. The instrument of violence is always used in order to achieve another aim. This may be positive or negative, and include an aim about other people's use of violence, for example to create disorder, or to prevent another actor being able to use violence. But the use of violence is a tool nevertheless. Therefore the next question this article will look at is what kind of instrument it is, and what it includes.

Aspects of violence

A key debate in the definition of violence is whether it includes only acts of commission, or acts of omission also. Honderich illustrates this by asking if the following two scenarios are equally violent: I send a poisoned food parcel to India, and the people who receive it die; I do not give money to Oxfam, and people who could have been helped die. In the first example, the deaths are a direct result of my act. The only way for another person to stop those deaths would be to intervene to stop my act. If I had not been born the act would not have happened. In the second example, the deaths are a result of omissions by many people. Any one of them can act to stop the deaths. If I had not been born the deaths would still occur. Honderich concludes that the difference is intention. Honderich does not classify acts of unintentional or partly unintentional omission as violent.Footnote ⁵⁸ By accepting that all violence is intentional, it means that to call an event in international politics violent, we must be able to see some element of choice from the actor involved. This does not mean that structural forces are irrelevant to the subject of violence in international politics. There are clearly strong structural forces at work within the international system, and within relationships at all levels that affect international politics. However, although violence may happen as a result of the structures in society, the actual act of violence itself is intentional, in that it has a conscious actor, it is the result of a decision by the person perpetrating the act. This decision may well be made because the actor thinks he or she has no other option given the circumstances they find themselves in, but it is still an intentional act. Structural violence takes the opposite point of view, and is designed to incorporate into the definition of violence structural aspects which do not have an actor.

Structural violence

Galtung is the most prominent advocate of structural violence. Many papers referring to structural violence refer back to his paper Violence, Peace and Peace Research. He explains structural violence as the difference between the actual and the potential: ‘violence is present when human beings are being influenced so that their actual somatic and mental realizations are below their potential realization’.Footnote ⁵⁹ So according to this definition if something is avoidable and yet still happens, violence has occurred. If it was unavoidable then there is no violence. The example Galtung gives is that tuberculosis is now a preventable disease, and so if someone dies from it this is due to violence. In the eighteenth century however, tuberculosis was not preventable, and so someone dying from it was not a result of violence. Thus, violence is not necessarily dependent on there being an actor, or of the act being physical, or of there being an object that is hurt. For Galtung, we can discuss physical or psychological violence, negative or positive violence (where ‘a person can be influenced not only by punishing him when he does what the influencer considers wrong, but also by rewarding him when he does what the influencer considers right’),Footnote ⁶⁰ violence with or without an object, violence with or without an actor (direct or structural violence), violence that is intended or unintended, and violence which is manifest or latent.Footnote ⁶¹ This concept of violence has become popular, especially the notion of structural violence, where there is no actor but that violence is the result of structure and ‘shows up as unequal power and consequently as unequal life chances’.Footnote ⁶²

Much of the attractiveness of this idea of structural violence is that it broadens the remit of security studies, or of research into violence. Thus, economic issues and the damage done by poverty and so on become just as important as the damage done by direct, physical violence. Authors who work with the concept of structural violence aim to highlight the hidden structures in order to work towards their transformation.Footnote ⁶³ The problem with this is that it is defining a concept dependent on what we want to be able to study within its remit. A concept should not depend for its meaning on how we want to study it, but rather on what the concept means. The concept of structural violence is performing a similar role to the debate about broadening the definition of security. Within the security debate we are used to people posing the challenge of securitisation – that people call something a security problem in order to make it sound more urgent, more policy relevant and so on. One can pose the same challenge to structural violence. Rather than having a clear academic reason for stretching the concept of violence to incorporate other, equally bad, social ills, the main reason proposed is that these other social ills cause as much or more damage than the damage caused by violence. This may well be the case, and these issues should have urgent academic and policy attention. However, re-defining a concept like violence to incorporate these issues in order to gain that attention is a poor way of achieving this. A clear argument that states the reasons why wider social ills are more worthy of our attention does not need the further argument that we should also call them violence.

So, by calling something violent, or designating it as an issue of security, an author is claiming a certain importance for the issue, escalating it up the policy agenda, and allowing for extraordinary reactions. For example, by saying that extremists use violence, one designates it as a significant and bad problem, which allows an exceptional response of violence (called military force) to counter it. The aim of calling something violence in order to push it up the policy agenda, meaning it requires special, urgent action, can also lead to dealing with these issues in a different, exceptional way, outside the realm of ‘ordinary’ politics. This may or may not be beneficial in dealing with the issue in question, or for our politics in general.Footnote ⁶⁴

A similar move is made by Žižek when he claims that alongside subjective violence (direct, intentional violence), there is also objective violence, one form of which is systemic violence – the damaging consequences of the ‘normal’ functioning of the system. This systemic violence must be taken into account, according to Žžiek, in order to make sense of subjective violence.Footnote ⁶⁵ He argues that this systemic violence needs to be given prominence, despite the urgency attributed to direct violence which fights for our attention.Footnote ⁶⁶ Again, this argument is valid in that systemic problems cause more suffering in the world than direct violence. But there has to be a further reason to also call this violence. After all, the argument can be made without the need to label something as violence. One could argue that seeing as our concepts are important because they change the way we think about the world, and change the way we act in it, a definition of violence that incorporates these other social ills would be valuable. However, broadening the concept in this way also has the potential for being damaging. If we conceive of all these things as equally violent, in order to see them as equally important, there is an implication that we are also going to tackle the issues as problems of violence. Issues of poverty are not helpfully tackled in the same way as issues of direct violence.

The concept of structural violence is problematical as it means that the definition of violence becomes linked to the result of an act (or influence), and not to the intention or actual action of the actor. This illustrates a key difference in the way violence is conceived: one way sees violence from the perspective of the perpetrator, and sees it as intentional, destructive force; the other way sees violence from the point of view of the victim, and sees it as a form of violation.Footnote ⁶⁷ Bufacchi points out that these two concepts of violence stem from the Latin roots of the term. The root of the word violence is violentia, meaning a passionate and uncontrolled force, but the meaning is often conflated with ‘violation’, from the Latin violare, meaning ‘infringement’.Footnote ⁶⁸

Although the definition of violence is partly contingent on the result of an act, in that it requires that the intention is to physically harm the victim, it is also necessary for that result to be a means to an end, not an end in itself. Because the result of many different acts or situations is physical harm or death (or unequal power or life chances), many things become incorporated into the definition of violence if the concept of intention is not used, and if the instrumental nature of violence is ignored. Roberts goes some way to improving on the debate about structural violence by pointing out that structures are created by people, and thus structural violence can be prevented and does have responsible actors. He also operationalises it by referring to specific acts of structural violence, looking at avoidable civilian deaths. He argues that his way of looking at human insecurity enables the analysis of structures, institutions and human agency, but without the problems caused by broadening the debate to include Galtung's concept of realising full human psychosomatic potential.Footnote ⁶⁹ But this does not prevent the problem that structural violence still refers to anything an author wants it to. Interestingly, many of Roberts' choice of examples can be incorporated under the title of direct violence in any case (they are preventable female deaths: infanticide, maternal mortality, intimate killings (normal domestic murder, dowry murders and ‘honour’ killings), lethal female genital mutilation; and avoidable deaths in children under five).Footnote ⁷⁰

Some caution is needed in this approach, however. It is easy to think of examples where violence is not intentional, for example a natural disaster, or bombing an empty building that accidentally harms someone nearby. We have already recognised that the concept has fuzzy boundaries, and thus we can recognise that some violence occurs naturally. The point here is to discuss that violence which is most relevant for the study of international politics. This is not the same as accepting that everything that causes harm in some way can be captured within the term violence. It is also not going so far as to say that accidental acts of violence are not violence at all. Collateral damage is still violent. The point of including the idea of intentionality in the definition of violence is that it ensures the violence we are discussing in international politics has an actor and does not end up including indeterminate ideas like a state of violence. It does not just happen on its own. It is also an action that is done with the intention of harming, unlike acts such as a doctor causing some pain in order to heal.

Harris points out the difficulty in describing what constitutes a violent act. He says that a violent act can be done non-violently, for example poisoning someone, or using biological weapons in warfare instead of dropping bombs.Footnote ⁷¹ One could add that a non-violent act can be done violently, for example a surgeon using a knife to perform an operation.Footnote ⁷² Harris therefore asserts that the descriptive element is moved from the act to the consequences. The result is that violence is an instrument – more to do with the means, the actual act itself, than to do with the result. Yet it is also partly contingent on the result. It is an act that intentionally leads to harm, but with another aim. The harm is not the desired end, but a means to another end.

Structural or systemic violence focuses almost completely on the result of violence, and does not include a requirement of intention or the instrumental nature of violence. Therefore it ends up including a broad range of acts or occurrences that lead to physical or mental harm. This is not useful for the study of international politics. As Keane puts it, this ‘makes violence indistinguishable from experiences like “harm”, “misery”, “unhappiness”, “alienation”, “cultural discrimination” and “repression”’.Footnote ⁷³ This does not completely negate the insight the concept of structural violence can bring into the structural cause of injustice and so on. These insights have been important in bringing social injustice into the foreground. The study of IR needs to be careful not to prioritise the study of direct violence over other forms of suffering like poverty. However in conceptual terms structural violence needs to be seen as different to violence. In fact, as Derriennic points out, Galtung's concept of structural violence makes much of what is usually seen as nonviolence (for example, passive resistance, strike, civil disobedience) a violent act. Derriennic does not see this as a problem, as non-violence in this context refers to a particular tactic, and could easily be referred to as a less-violent action.Footnote ⁷⁴ However, this point does bring into question a definition of violence which cannot distinguish between violence and non-violence. We do not need to refer to all social ills as violence in order to recognise that they bring human suffering and need to be eradicated. Nor do we need to refer to structural forces as violence in order to recognise the role they play in instigating acts of violence. What we do need is an understanding of the complex nature of structure and agency regarding violence. Calling everything violence does not contribute to this understanding. The choice to use violence is conscious, but the choice is made from a series of socially acceptable choices, and structure plays an important role in defining what those choices are.Footnote ⁷⁵

Although the structural aspect of violence is an important issue, it moves too far away from the central meaning of the word. Galtung says ‘[i]n order not to overwork the word violence we shall sometimes refer to the condition of structural violence as social injustice’.Footnote ⁷⁶ Social injustice is not the same as violence. It may be a bad thing, but the aim is not to give the label violence to everything that one disagrees with.

Direct violence

So, why is direct violence the best way to approach its study in International Relations? This article has already established that other forms of violence are conceptual stretching and complicate any discussion of violence unnecessarily. This is not the same as saying that these phenomena do not exist or should not be studied. In fact, I would argue that links between direct violence and social injustice or global disorder are fruitful areas for research. But it is to say that these broadened definitions should be treated as separate phenomena from direct violence. Ultimately, many of these other conceptions of violence are more closely related to violation than violence. As such, they can refer to a very broad range of things of varying levels of importance, including civil rights, human rights and so on. The argument that these aspects are just as important as direct violence is compelling, but not for the re-definition of the concept. After all, the word violation is already available for these aspects.

Following on from this is the need to establish what direct violence involves. I have already demonstrated above that it is an intentional and instrumental act. It could also involve a range of things, from physical or emotional pain or injury, the threat of pain or injury to enforce an action,Footnote ⁷⁷ psychological violence, covert violence, a violation of someone's dignity.

In common parlance, the word violence is often used in a metaphorical sense. I argue that this needs to be avoided, or at least made clear that this is how it is being used, when analysing international politics, for the sake of clarity. When violence is used metaphorically, it tends to be used in a way that gives (negative) value to something. For example, someone might say ‘she let loose a violent attack on him’, meaning a verbal rather than a physical attack. This would be a metaphorical way of describing the way she talked, rather than an actual description of what happened. The term violence has been used metaphorically to such a degree that it is often difficult to ascertain the exact meaning of the word, and whether it means physical harm or an extreme case of something like an argument. This distinction is important in international politics.

A dictionary definition of violence is ‘[t]he exercise of physical force so as to inflict injury on, or cause damage to, persons or property’.Footnote ⁷⁸ This definition works well as a description of direct violence. But it is also important to remember the instrumental nature of violence. Keane provides a fuller yet similar description of the concept of violence: ‘the more or less intended, direct but unwanted physical interference by groups and/or individuals with the bodies of others, who are consequently made to suffer a series of effects ranging from shock, speechlessness, mental torment, nightmares, bruises, scratches, swellings, or headaches through to broken bones, heart attacks, loss of body parts, or death’.Footnote ⁷⁹

The only aspect these definitions do not account for is psychological violence. Betz argues that this is a metaphorical use of the term, although it can also be seen to result in serious harm, including physical harm. He claims that all violence is overt or physical.Footnote ⁸⁰ However, psychological violence can result in harm as damaging as physical, direct violence, and so it would seem wrong to exclude psychological violence from a definition of violence. In this context, however, it would need to be intentional psychological violence that was a means to an end.

In conclusion, I am suggesting a working definition for violence in the context of the study of international politics as being an intentional act designed to cause harm, which is direct and physical or psychological. It is instrumental, a tool in order to achieve a particular aim. The reason this is a good use of the concept is that it is close to what most people would conceive of as the core meaning of the term. At the same time, it is narrow enough to allow analysis without causing confusion about what exactly we are discussing. It is also applicable to any actor, whether we perceive their actions as legitimate or not. We must not forget that the fundamental basis of violence is that it does harm to an individual. Euphemisms that are often used in international politics can allow us to forget this. This definition of violence also allows violence to be researched at many levels, from an individual instance of violence like rape, to a large scale campaign of violence like war. This conception of violence does not hide the fact that we are talking about harm to individuals, whether it be through organised violence, violence of one person against another, violence perpetrated via the use of technology and so on.