Abstract
The aim of this review article is to explore some theoretical issues regarding the nature and scope of violence. There are two ways of thinking about violence: in terms of an act of force, or in terms of a violation. Those who define violence as an intentional act of excessive or destructive force endorse a narrow conception of violence (the Minimalist Conception of Violence or MCV), while those who see violence in terms of a violation of rights champion a broader conception of violence (the Comprehensive Conception of Violence or CCV). The strengths and weaknesses of both approaches will be assessed.
The twentieth century has appropriately been defined as the long century of violence. 1 The number of victims, the magnitude of devastation and the new technologies of destruction are without historical parallel. 2 It has been calculated that war-related killings in the twentieth century exceeded 105 million people, including 62 million civilian victims (Steger, 2003, p. xiii). And of course, the dead are not the only victims of violence. For every person killed by an act of violence there are many more who have survived torture, brutality, persecution, the loss of loved ones, or (if they are lucky) the loss of all their belongings and livelihood. One can safely speculate that every person living today will experience, directly or indirectly, some type of violence.
The persistent proliferation of political violence should not come as a surprise, given that violence is, and has always been, the essence of politics. Writing in the seventeenth century at a time when the modern state, and modern philosophy, acquired its present identity, Thomas Hobbes famously depicted the pre-political ‘state of nature’ as a place of violence, where everyone endeavours to destroy or subdue one another, making life solitary, nasty and short. Even John Locke, who entertained a much rosier picture of the state of nature, admitted to a problematic level of violence, which made life ‘inconvenient’. But if violence is the problem, violence is also the solution. We escape the pre-political state of promiscuous violence by forming a political society under the rule of a centralised authority that, to paraphrase Max Weber's famous dictum on the State, claims a monopoly over the legitimate use of violence. The force used by state institutions may be legal, even legitimate, but it is still violence. 3 And when the state is incapable of providing protection for its subjects, private agencies will take over the role of dispensing violence and security, as in the case of the Russian mafia (see Varese, 2001).
Following in the footsteps of Hobbes and Weber, today the vast majority of political theorists see collective violence as integral to political life. 4 No one is immune from political violence, whether living in a wealthy Western democracy, or in a country undergoing late economic development (Weinberg and Rapoport, 2001; Bates, 2001). Understanding political violence is therefore a principal preoccupation for many social scientists and political theorists, which explains why in recent years there has been an influx of books on the topic of political violence, widely defined. The empirically-driven output from sociologists, psychologists, anthropologists and political scientists has been relentless, impressive in its quantity if not always its quality. The bulk of these publications tend to be edited volumes based on ethnographic case studies rich in the exposé of violence in all its forms, from hate crime to genocide. 5 These publications are to be welcomed, as each case study is informative and revealing, yet the reluctance of the editors to ground the disparate ethnographic case studies on a shared definition of violence takes away from the theoretical value of these books. 6 We may learn a great deal about specific acts of political violence occurring within certain cultures in different parts of the world, without necessarily having a better understanding of the meaning of violence as a universal concept. At best these volumes remind us of the complexity of violence, they don't help us to understand it. In the last analysis, knowing what makes a certain violent act an act of violence, 7 deciding whether omissions that cause harm should count as acts of violence (Glover, 1977; Harris, 1980; Hondrich, 1989, 2002), determining why violence is bad and prima facie wrong, 8 or establishing when and how violence can be justified (Gert, 1969; Audi, 1971, 1974; Grundy and Weinstein, 1974; Geras, 1990), demands a much closer inquiry into the concept of violence.
In what follows the extensive empirical research on political violence will be largely ignored. 9 Instead, the scope of this review article will be much more modest, restricted to a philosophical analysis of the concept of violence. 10 As always the best place to start is by investigating the etymology of violence. The word ‘violence’ is derived from the Latin violentia, meaning ‘vehemence’, a passionate and uncontrolled force. Yet because acts of excessive force frequently result in the violation of norms, rights or rules, the meaning of violence is often conflated with that of ‘violation’, from the Latin violare, meaning ‘infringement’. Indeed most attempts to define violence tend to combine the idea of an act of physical force with a violation. Thus for example Ted Honderich (2002, p. 91; 2003, p. 15) defines violence as ‘a use of physical force that injures, damages, violates or destroys people or things’, while Manfred Steger (2003, p. 12) points out that violence ‘comprises a range of meanings, including ‘to force’, ‘to injure’, ‘to dishonor’, and “to violate” (see also Riga (1969) and Wade (1971)).
The prevailing synthesis between ‘violence’ and ‘violation’ is not necessarily to be welcomed, as it may lead to growing confusion rather than clarity (see van den Haag, 1972). In fact, while acts of physical force often entail some form of violations, there are times when a violation occurs without the need of any physical force, or, alternatively, acts of physical force may take place without anything or anyone being violated. For this reason, in what follows the relationship between violence and force, and between violence and violations, will be reviewed independently of each other. Separating issues of physical force from issues of violations has the virtue of identifying two competing perspectives on violence, where the approach taken will determine whether the concept of violence ought to be defined narrowly (violence as force) or more broadly (violence as violation).
Violence as Force
Coady (1986) is probably right when he reminds us that the normal or ordinary understanding of the term ‘violence’ is in terms of interpersonal acts of force usually involving the infliction of physical injury, which suggests that the concept of violence cannot be understood independently from the concept of force. This strong affinity between the terms ‘violence’ and ‘force’ would appear to be vindicated by the Oxford English Dictionary, where violence is defined as ‘the exercise of physical force so as to inflict injury on, or cause damage to, person or property’.
Notwithstanding its linguistic roots and common perception, the relationship between violence and force is one of the most debated issues in the literature on violence. On one side there are those who have no qualms about defining violence in terms of force. 11 One of the earliest and most authoritative representatives of this group is the American pragmatist John Dewey. In a series of articles originally published in 1916, Dewey (1980, p. 246) argues that violence is force gone wrong, or, put another way, force that is destructive and harmful: ‘energy becomes violence when it defeats or frustrates purposes instead of executing or realizing it. When the dynamite charge blows up human beings instead of rocks, when its outcome is waste instead of production, destruction instead of construction, we call it not energy or power but violence’.
It is important to emphasize that Dewey is not saying that force and violence are synonymous. Clearly not all force is violence (rescuing someone from drowning or preventing someone from harming themselves), just as not all acts of violence require the use of force (murder by poisoning). Instead, according to Dewey it is only when force becomes destructive and harmful that it turns into an act of violence. 12 Apart from being destructive, it has been suggested that an act of force must also be intentional or deliberate in order to count as an act of violence. Thus, for example, Thomas Pogge (1991, p. 67) claims that ‘a person uses physical violence if he deliberately acts in a way that blocks another's exercise of her legitimate claim-rights by physical means’, while Steger (2003, p. 13) argues that ‘violence is the intentional infliction of physical or psychological injury on a person or persons’. 13 There is also a debate whether excessive force, apart from being intentional or deliberate, should also be unwanted. 14 The advantage of adding this clause is that it would rule out acts of intentional, excessive force that are clearly not acts of violence, such as surgery, although the risk is that clear examples of violence, such as footbinding or voluntary labial infibulations, would not be recognised for what they really are. 15
Relating violence to an intentional act of excessive force is intuitively appealing, but on further reflection there appears to be an underlying problem with any attempt to define violence in terms of force (see Miller, 1971). Robert Paul Wolff (1969, p. 604) defines force as ‘the ability to work some change in the world by the expenditure of physical effort’, and Hannah Arendt (1969, pp. 44–5) points out that ‘Force, which we often use in speech as a synonym for violence, especially if violence serves as a means of coercion, should be reserved, in terminological language, for the ‘forces of nature’ or the ‘force of circumstances’ (la force des choses), that is, to indicate the energy released by physical or social movements’. These two definitions underscore a fundamental qualitative difference between ‘force’ and ‘violence’. First, force is a dispositional concept, that is to say it refers to an ability or potentiality. Violence on the other hand refers to the action itself, or as Audi (1971, p. 50) points out, ‘violence in this sense is always done, and it is always done to something, typically a person, animal or piece of property’. Second, violence is an evaluative concept, perhaps even a normative concept, while force is not. 16 It is perhaps the moral neutrality of the concept of force that lead Hannah Arendt (1969) to dismiss it as a useful path into the meaning of violence, suggesting instead that we focus on the relationship between violence and power. 17
Violence as Violation
Apart from its affinity with the notion of force, violence can also be conceptualized in terms of the verb ‘to violate’, meaning to infringe, or transgress, or to exceed some limit or norm. Newton Garver (1973) goes as far as to suggest that the idea of violence is much more closely connected with the idea of violation than it is with the idea of force. 18 Following Garver, many contemporary theorists of violence have converged on the idea of defining violence in terms of a violation, although there seems to be some disagreement about what exactly is being violated when an act of violence takes place (see Waldenfels, 2000, 2005).
The most popular answer to the question ‘violation of what?’ is ‘violation of rights’. Unfortunately, the immediate appeal of this answer is misleading. If violence is the violation of rights, then naturally one ought to say something about the nature of rights being violated. This next step is imbued with difficulty. There are at least three distinct ways of conceptualizing the set of rights that are being violated by an act of violence. First, we could be talking about the violation of personal rights, or those rights essential to personality. Garver (1968, 1973) is the best-known exponent of this position. He argues that rights are of two kinds, referring to either the body or the dignity of the person. 19 Second, we could be talking about the violation of the right to ourselves, widely defined. This answer finds favour with those political philosophers who endorse the thesis of self-ownership, thus for example Nozick (1974, p. ix) famously tells us that ‘individuals have rights, and there are things no person or group may do to them (without violating their rights) … any more extensive state [beyond a minimal state] will violate persons’ rights not to be forced to do certain things, and is unjustified’. Finally, we could be talking about the violation of human rights, widely defined to include any obstacle or impediment to the fulfilment of a basic need. Galtung (1969) is perhaps the best-known advocate of this position, which is still very popular especially amongst radical opponents of globalization. For example Jamil Salmi (1993, p. 17) defines violence as ‘any avoidable action that constitutes a violation of a human right, in its widest meaning, or which prevents the fulfilment of a basic human need’. He goes on to say that each time human beings starve or are undernourished because of social or political reasons, it is legitimate to consider these people as the victims of social violence.
The appeal of defining violence in terms of the violation of rights is evident, but once again further analysis exposes at least two intrinsic problems with this line of reasoning. Albeit rare, there are cases of violence occurring without rights being violated. This point has been made by Audi (1971, p. 59): ‘[while] in the most usual cases violence involves the violation of some moral right … there are also cases, like wrestling and boxing, in which even paradigmatic violence can occur without the violation of any moral right’. Another objection regards the all-inclusive concept of rights being violated. Apart from the violation of our basic rights, such as the right to life, to personal security and to liberty, violence is understood to include also the violation of our socioeconomic rights. Yet the broader our definition of human rights, the more pervasive and inescapable violence becomes. Almost any act can be said to violate someone's rights, making violence ubiquitous and therefore meaningless. As Joseph Betz (1977, p. 341) points out: ‘If violence is violating a person or a person's rights, then every social wrong is a violent one, every crime against another a violent crime, every sin against one's neighbour an act of violence’.
Minimalist vs Comprehensive Conceptions of Violence
So far two different ways of approaching the idea of violence have been investigated: violence as excessive force and violence as violation. The first approach leads to a narrow conception of violence, while the second approach leads to a broader conception. Defining violence in terms of excessive or destructive force has the important advantage of delineating clear boundaries around what constitutes an act of violence, avoiding therefore the tendency to use the term violence as synonymous for everything that is evil or morally wrong. 20 This in part explains why Norman Geras (1990, p. 22) suggests that violence be defined simply as ‘the exercise of physical force so as to kill or injure, inflict direct harm or pain on, human beings’. Definitions of violence that emphasize the notion of physical force deliberately used to cause suffering or injury will be referred to as the Minimalist Conception of Violence (MCV). 21
It is usually advisable to strive for precise, tight definitions of key concepts, therefore from a purely analytical point of view the MCV is to be welcomed. The problem with the MCV is that by restricting acts of violence to intentional, direct, physical acts against other persons the MCV misses out on too many other important dimensions of the phenomenon of violence. For example, Audi (1971, p. 59) reminds us that an act of violence can be physical or psychological, aimed at persons, animals, or property, as suggested by his definition of violence: ‘Violence is the physical attack upon, or the vigorous physical abuse of, or vigorous physical struggle against, a person or animal; or the highly vigorous psychological abuse of, or the sharp, caustic psychological attack upon, a person or animal; or the highly vigorous, or incendiary, or malicious and vigorous, destruction or damaging of property or potential property’. One of the major virtues of Audi's definition is the acknowledgement that violence has a psychological dimension. This has particular resonance within the growing literature on family violence, where the psychological impairment that comes from living under constant threat and fear of violence is recognized as being part of what constitutes domestic violence. 22 Similarly the growing literature of testimonies from genocide survivors also suggests that psychological violence may be the worst aspect of an act of violence, even worse than physical violence. 23
Another problem with the MCV is that it seems to be oblivious to the most pervasive and destructive form of violence: structural or institutional violence. In his ground-breaking article ‘Violence, Peace, and Peace Research’ Galtung (1969) distinguishes between ‘direct violence’, where the instigator of an act of violence can be traced to a person or persons, and ‘structural violence’, where there may not be any person who directly harms another person. In structural violence the violence is built into the structure, and shows up as unequal power and consequently as unequal life chances. As Galtung reminds us, structural violence is more deadly and destructive than direct violence. 24
The attempt to broaden our understanding of violence, either along Audi's or Galtung's lines, may be referred to as the Comprehensive Conception of Violence (CCV). 25 As we have already seen, advocates of the CCV maintain that there are some notable advantages with going beyond the MCV, yet there are also serious problems with the CCV that must be confronted. For example, by introducing a psychological component Audi would appear to offer only a more obscure and less precise definition of violence. It is true that according to Audi not all psychological abuses are acts of violence, but only ‘vigorous’ psychological abuses, yet as Audi himself recognizes, the term ‘vigorous’ is inherently vague, and perhaps even subjective. 26 A problem of an analogous nature troubles Galtung's notion of structural violence. When Galtung (1969, p. 168) explains that ‘violence is present when human beings are being influenced so that their actual somatic and mental realizations are below their potential realizations’, he fails to appreciate that this definition is much too inclusive, its scope much too broad. Not surprisingly Galtung has attracted much criticism, for example from Keane (1996, p. 66) who takes a very dismissive stand: ‘attempts (such as John Galtung's) to stretch its meaning to include ‘anything avoidable that impedes human realization’ effectively makes a nonsense of the concept [of violence], linking it to a questionable ontological account of ‘the satisfaction of human needs’ and making it indistinguishable from ‘misery’, ‘alienation’ and “repression” (see also Coady, 1986).
For definitional purposes, it is crucial to delineate the boundaries of what constitutes an act of violence. At present there appears to be no agreement regarding this issue, with some arguing for a narrower scope (MCV), while others defend a broader scope (CCV). While this debate cannot be settled here, it is important to emphasize that behind this debate there is a disagreement of a different nature, namely whether violence should be defined from the point of view of the perpetrators (violence as intentional, destructive force) or the victims (violence as a violation). An alternative approach, yet not fully explored in the literature, is to define violence from the point of view of an impartial spectator or third party. 27
Conclusion
The problems of violence may be cardinal to a proper understanding of political life, yet the concept of violence remains elusive and often misunderstood. It was 1906 when Georges Sorel (1961, p. 60), the French social theorist of anarcho-syndicalism, remarked: ‘the problems of violence still remain very obscure’. 28 Writing 60 years later, Arendt (1969, p. 35) commented: ‘what Sorel remarked sixty years ago … is as true today as it was then’. We can confidently say that what Arendt remarked 40 years ago is also as true today as it was then.
Compared to Sorel's time, or even Arendt's time, today we have access to a much greater pool of empirical case studies on violence. The many different forms of political violence, including terrorism, civil disobedience, genocide, war and revolutions, all enjoy their own growing and specialised bodies of literature, while philosophical investigations of the concept of violence seem to be lagging behind. 29 In undertaking a preliminary exploration of the idea of violence, this review article has focused on two ways of thinking about violence, the minimalist approach sees violence as an act of intentional, excessive force, while the comprehensive approach sees violence as a violation. The ideological and methodological gulf between these two approaches is so fierce and deep-seated that one is left wondering whether there are not one but two concepts of violence.
Footnotes
The research for this review article was made possible thanks to funding received from the Irish Research Council for the Humanities and Social Sciences. An earlier version of this paper was presented at the Centre for Values and Social Policy at the University of Colorado, Boulder. I am grateful to the editors of Political Studies Review, two anonymous referees, and the participants of the seminar in Boulder for their challenging comments. In particular I would also like to thank Eve Garrard and Alison Jaggar for their thought-provoking views on the meaning of violence.
1
See Keane (1996, p. 3): ‘Genocidal wars, firebombed cities, nuclear explosions, concentration camps, spreading plagues of private blood-letting: this century has seen more than its share of planned and unplanned violence, with prospects of still more to come.’
2
See Wolin (1963, p. 20): ‘This is the century of secret police, concentration camps, gas chambers, labor camps and the calculated ‘liquidation’ of specified classes, races and whole populations.’
3
See Brady and Garver (1991) on the relationship between violence and law: violence is stemmed by laws, yet laws also justify and legitimate violence. 
argues that when violence is legitimised, it is no longer violence. Violence applies specifically to the illegitimate act of imposing one's will upon another, whereas the imposition is held to be legitimate in the case of force. According to Macfarlane the state uses force, not violence.
4
See Ladd (1991); Hardin (1995); Hall and Whitaker (1999); Summers and Markusen (1999); Barkan and Snowden (2000); ; Conteh-Morgan (2004).
5
Recently published edited volumes on violence include Apter (1997); Turpin and Kurtz (1997); Besteman (2002); Worcester, Bermanzohn and Ungar (2002); Stanko (2003); .
6
7
On the important and often neglected distinction between violent acts and acts of violence, see Harris (1980).
8
On why violence is bad see Bufacchi (2004); on why it is wrong see Perry (1970); Wells (1970); Nielsen (1981); .
9
10
11
Apart from Coady (1986), see also Graham and Gurr (1969); Nieburg (1969); Audi (1971); Gotesky (1974); Geras (1990); Keane (1996); .
12
Ladd (1991, p. 27) goes as far as to say that ‘destructiveness is the essence of violence’.
13
Intentionality also plays a central role in the definition of violence given by Miller (1971) and Coady (1986). For a counter-argument see .
14
Keane (1996, pp. 66–7) defines violence as ‘the unwanted physical interference by groups and/or individuals with the bodies of others, which are consequentially made to suffer a series of effects ranging from shock, bruises, scratches, swelling or headaches to broken bones, heart attack, loss of limbs or even death’.
15
Simpson (1970) argues that in a hostile environment one may become inured to almost anything, including injury, therefore someone raised in the midst of turbulence or social decay may fail to recognise that they are being subjected to violence.
16
Robert Paul Wolff (1969) argues that force, unlike violence, is morally neutral; Holmes (1973) suggests that the concept of violence is evaluative, although this does not make violence wrong by definition; maintains that the concept of violence is a moral concept.
17
Arendt's analysis of violence in terms of power is also not very convincing. First, Arendt tells us that power corresponds to the human ability to act in concert. This is very odd. If we define power in Weberian terms as the ability to get outcomes despite resistance, it follows that power applies equally to single individuals as it does to groups. Secondly, Arendt tells us that power is an end in itself, whereas violence is by nature instrumental. Once again this is puzzling. Why is power an end in itself? As points out, the desire for power as an end in itself, perhaps because it is intrinsically gratifying, is often overestimated. Notwithstanding my reservations, Arendt's analysis of violence is still very influential. See Bar On (2002).
18
In 1973 Garver revised and expanded his original article ‘What Violence Is’ first published in 1968. See Garver (1968, ).
19
Garver argues that one has the right to one's body and the right to autonomy; Holmes (1971, p. 112) defends a position similar to Garver, although he emphasizes the non-bodily side of violence: ‘to do violence to someone … is to diminish him as a person, where what is central to this notion is the inflicting of mental harm’. See also .
20
Audi (1974, p. 38) warns us against using the emotively loaded word ‘violence’ when the grievance can be better described and treated under another name, for example ‘injustice’: ‘misnaming the disease can lead to the use of the wrong medicine – or none at all’. Platt (1992) makes a similar point. Lee (1996, pp. 79–80) puts forward the counter-argument that ‘both the outlaw and the maintainer of an unjust status quo are capable of great moral wrongs, and we should not let this point be obscured by reserving the strongest condemnatory language for only one of them’. See also .
21
The Minimalist Conception of Violence echoes what Coady (1986) calls the ‘restricted’ definition of violence, what Grundy and Weinstein (1974) refer to as the ‘observational’ definition, and what identifies as the ‘descriptive’ concept of violence.
22
For an overview of the voluminous literature on domestic and family violence, see Breines and Gordon (1983); French, Teays and Purdy (1996); ; Bar On (1998).
23
Taking the Guatemalan genocide as a representative case study, see Zur (1998); ; North and Simmons (1999).
24
See also Garver (1973); Galtung (1996); Lee (1996); Curtin and Litke (1999). For an overview of Galtung's contribution to the study of violence and peace, see .
25
The Comprehensive Conception of Violence echoes what Coady (1986) calls the ‘wide’ definition of violence, what Grundy and Weinstein (1974) refer to as the ‘expansive’ definition, and what identifies as the ‘polemic’ concept of violence.
26
At the same time it is debatable whether the term ‘vigorous’ in ‘vigorous psychological abuse’ is more vague than the term ‘excessive’ in ‘excessive physical force’.
28
On Sorel, see King (2003); Jennings (1985); .
29
It is worth mentioning that there is an excellent anthology in French that looks at the meaning of violence in the history of philosophy. La Violence, edited by Hélèle Frappat (2000), reproduces the relevant extracts from the usual suspects in the history of European thought, including Plato, Machiavelli, Spinoza, Hobbes, Rousseau, Kant, Hegel, Engels, Nietzsche, Sorel, Schmitt, Freud, Benjamin, Weber, Foucault, Arendt and Derrida. Each extract is preceded by a brief introduction, and at the end of the volume the reader will find a short philosophical analysis of nineteen key terms related to violence. Overall, this is an extremely useful book that inadvertently exposes a major lacuna in the literature in English on violence.