Keywords: child-to-parent violence, limits, hierarchies and family secrets.

Introduction

Over the past three years, the media have reflected a dramatic increase in what we may call the third type of intrafamily violence. Historically, attention was first paid to the violence of parents towards children, then to violence between partners; today child-to-parent violence (CPV) is emerging.

The judicial reports of these years record a marked increase in complaints from parents assaulted by their children: they multiplied eightfold in Catalonia over the period 2000-2004¹, threefold in the Basque Country between 2002 and 2003², and fourteenfold from 2000 to 2004 in the Valencian Community³.

This does not mean that it did not exist before; but it is emerging before public opinion as if it were a new phenomenon, and professionals, like society as a whole, are beginning to pay attention to a problem that was conceived, in the past, as one of those that accompany moderately severe pathologies.

In reality, there is nothing strange about this process: the same thing happened with the other types of intrafamily violence. Child maltreatment and partner violence are both situations rooted for many years within the family. Only their definition as inadequate and harmful, together with the effort to bring them to light, changed the fragmented view that was held of them, favouring the social emergence of a hidden problem. In the same way, CPV remained concealed as one of the conflicts presented by a family carrying other dysfunctions.

But another factor has been decisive in this sudden appearance of CPV: the emergence of a new profile of violence, located in apparently normalised families, exercised by children who previously presented no problems, and who are responsible for this dramatic increase in judicial complaints.

We will describe these two types of violence further on but, first, we will define what we mean by violence, and by CPV.

Definition

The World Health Organization defines violence as “any intentional action or omission which, directed at a person, tends to cause them physical, psychological, sexual or economic harm” (WHO, 2002). We are therefore faced with various types of violence (physical, psychological, neglect, sexual abuse, economic maltreatment, vandalism), which can furthermore be classified according to the context in which they occur: intrafamily, political, school, workplace, etc.

It is necessary to draw a distinction between aggressiveness and violent behaviour. The former, relational in nature, is part of human experience and does not necessarily cause harm to the other party. Violent behaviour, on the other hand, is not part of one's way of being: it is a learned behaviour and, as such, it can be changed. Moreover, it is natural in its own context and only becomes disruptive when it is decontextualised (Abeijón, 2006). Thus, for example, rugby players can tackle one another violently for the whole duration of the match without their behaviour being deemed intolerable. But if those same players, once the match is over, reproduce the same behaviour elsewhere, it is very likely that the police will end up intervening.

We take a systemic starting point, in which we consider both a linearity and a circularity of violent behaviour. Violence must be addressed legally, in a linear context, where a differentiation is established between victims and perpetrators. But it must also be understood technically, and it is for this purpose that we introduce circularity, as a useful tool for understanding the process in which the participants in the violent behaviour interact (Lamas, 2000).

Definition of CPV

We understand this type of violence as the set of repeated behaviours of physical aggression (blows, shoving, throwing objects), verbal aggression (repeated insults, threats) or non-verbal aggression (threatening gestures, breaking cherished objects), directed against the parents or against the adults who take their place (Pereira, 2006).

Threats and insults are therefore included, whether conveyed through gestures or words, as are physical assaults of any kind, or the deliberate breaking of objects the assaulted person cares about. In addition, the violence must be directed against the parents or against the parental figures who replace them: guardians, educators, etc.

Occasional violence, without antecedents and which is not repeated, therefore does not fall within this definition. This excludes, almost as a general rule, parricide, which presents particular characteristics distinguishing it from this type of violence and which is often a single episode, without recorded antecedents.

Also excluded are sexual assault on parents and premeditated attacks with lethal weapons, considered as belonging to a different profile, as well as violence that appears in a state of significantly diminished consciousness (autism or severe intellectual disability) and which is not repeated once that state ceases: violence during intoxication, organic mental disorders, disorders of the form or content of thought, etc.

The violence that presents in these excluded cases belongs to that traditional model we referred to above: for example, assaults on parents committed by a drug addict in search of money or drugs in the throes of withdrawal, the attacks of a schizophrenic driven by delusions or hallucinations, or assaults produced in a twilight state are some illustrations of it.

We also include in this traditional CPV:

• adolescents and young people who defend themselves against assault, who react to being victims of sexual abuse or of humiliating treatment unsuited to their age;
• violence exercised to protect another family member who is being assaulted;
• what is called “retaliation”: children who suffered maltreatment or abuse in childhood, or who were the object of severe neglect or abandonment, and who return the maltreatment when the roles are reversed and it is they who care for their parents.

The new CPV shares with the other models the search for control and power within the family, although it differs from them in the pursuit of objectives. The new profile refers to assaults carried out by apparently normalised children, adolescents and young people, from every social background, with more or less extensive violent behaviours, which always include the family setting and, very frequently, are confined to that context. The age range of the aggressors is wide, although CPV is more frequent in adolescence, and it is not unusual for them to be boys and girls who, not only are not aggressive outside the home, but even display over-adapted behaviours in other contexts. Substance use is usual, but no more so than in the population of their age (Romero Blasco et al., 2005).

The people assaulted are the adults responsible for their upbringing, of any age and gender, although it is more frequent among older parents, in single-parent families, and even more so among mothers than among fathers (Gallagher, 2004; Ibabe, 2007).

It is a violence that generally occurs in escalation: it usually begins with insults and disqualifications, moves on to threats and the breaking of objects, and ends with increasingly severe physical assaults. It is a process that can last for years, and it must be stressed that it has no predetermined end: the violence grows progressively and does not stop, even once absolute submission, complete domination and control through terror have been achieved.

There are no reliable studies of prevalence and incidence, but its constant increase in recent years can be observed, mainly through judicial reports. Data from the United States indicate that 25% of telephone calls to associations against intrafamily maltreatment come from parents assaulted by their children. Cyrulnik (2005) reports an estimated incidence that varies between 1% of French families, 4% of Japanese families and 6% of American families, and adds that “in a single generation, the phenomenon of battered parents has taken on worldwide dimensions”.

Social factors

The social changes that have taken place over recent decades in most of the world – and most particularly in Western society – have extraordinarily favoured the appearance of new dynamics that have altered the balances of power within the family, as well as within the other great educational system: the school.

From the second half of the twentieth century onwards, there was a shift from a clearly authoritarian system to another, democratic but poorly understood, in which democracy is equated with the absence of authority, and with equality when it comes to making decisions. In Spain, this happened later and in a more accelerated way, in parallel with the political transition.

The family and social hierarchical model is called into question, but only formally, since educational responsibility continues to be attributed, legally and socially, to parents and educators. They are not stripped of responsibility but, often, of authority, as well as of some of the means habitually used to maintain it.

Other social changes, occurring in parallel, increase the difficulty parents and educators have in maintaining their authority:

– The fall in the number of offspring, with a significant increase in only children, a situation that reaches its peak in China, where it is instituted by law. Children, increasingly rare, frequently become a treasure to be cherished and cared for very delicately, to whom attention must be paid always and everywhere: the kings of the house.

– Changes in family models. The nuclear family sees its presence progressively diminish, to the point of representing today less than 50% of family models in Western society. Other typologies, such as single-parent or reconstituted families, foster or adoptive families, take up more and more space. In all these new models, for reasons specific to each although fairly similar to one another, maintaining the authority of the parent or parents becomes more difficult.

– Changes in the family life cycle, with a progressive rise in the average age at which people have children, which produces older parents, with less energy to fight with the children in order to maintain discipline and set limits.

– Changes in the world of work, with the full entry of women into employment and the increase in the number of hours that have to be spent away from home – interminable working hours or long commutes –, which give rise to what are called latchkey children. The number of hours of contact with the children decreases, people come home tired and, in the little time spent with them, try to avoid situations of tension as much as possible. Acts that generate frustration are eliminated, in search of a family harmony that does not further overburden the tired parent. A permissive upbringing then takes hold, the consequence of which is difficulty in setting limits and commanding respect.

And when it is other members of society who attempt to set those limits, it is not unusual to observe that the parents indiscriminately side with the child, confronting, sometimes violently, the teachers who are trying to educate without the parents' collaboration. This confrontation between family and educational system has led to the frequent loss of an indispensable collaboration, as well as to the indifference and absenteeism of many teachers burnt out by the lack of support of every kind in carrying out their work.

– Society's evolution towards an educational model based more on reward than on sanction, on tolerance rather than discipline, has led to a significant restriction of educators' capacity to sanction. Teachers and schoolmasters were long ago deprived of the possibility of using almost all punishments. Now this capacity to sanction is being taken away from parents, who can no longer physically punish their children.

At the same time, and inconsistently, society is increasingly permissive towards the violence of children. Violent messages multiply in the media, television and video games teach children and young people the thousand and one ways of solving their problems through violence, while the age of criminal responsibility is pushed back and legal responsibility is relativised.

– Finally, one cannot fail to point out the social drift towards a growing hedonism and nihilism, with a significant general disorientation as to the path to follow and a loss of the value references that would guide family members in the difficult work of upbringing.

We are therefore faced with a social model that attaches ever greater importance to childhood and youth, the latter being designated as the best age of life, which people try to prolong indefinitely. Permissiveness towards the behaviours of these ages increases while, simultaneously, those responsible for education are deprived of the coercive instruments necessary to exercise it. The private family space shrinks, and the social and legal systems – the public – intervene more and more in what was previously the family sanctuary. There is no doubt that this invasion of the private by the public has positive aspects, such as the fight against maltreatment and sexual abuse for example; it does, however, also carry some side effects: by withdrawing from educators and parents the social delegation for the administration and use of coercive instruments, these remain in the hands of the Administration alone.

It is therefore not surprising that we encounter, more and more frequently, educators and parents who give up their function and delegate it to the State. It is becoming common for social or judicial services to find themselves facing parents who explicitly renounce parental authority and hand their children over to the administration to take charge of them. Overwhelmed by a complex situation, with few resources to face it and held responsible by a society that blames them for their children's misbehaviour, they end up capitulating and passing the hot potato to the one who has reserved for itself the exclusive right to disciplinary and coercive capacity.

Parenting styles

Alongside these social changes, various authors have attempted to correlate parenting styles with CPV.

Peek and Fisher (1985), in a large study of a sample of students in the final years of American high schools, questioned the students about situations of CPV that had occurred over three years and about the styles of exercising authority and control (power style) in their families, and they found a clear relationship between the two factors. They then correlated their observations of CPV with four parenting styles: a) punitive and non-strict style; b) strict but non-punitive style; c) neither punitive nor strict style; d) violent style. Their results showed a clear relationship between the violent style (widespread violence in the family, exercised habitually) and the punitive but non-strict style (that is, more arbitrary and without clear rules) and CPV.

Gallagher (2004) proposes a similar model when he correlates CPV with two parenting styles: the permissive-liberal and the authoritarian-violent.

In line with these proposals, we find, in our caseload, a clear correlation between these parenting styles and CPV, although we can establish differences between the new and the traditional CPV as regards parenting styles.

The new CPV would be more closely linked to the permissive-liberal (or punitive non-strict) style, in which clear rules are not established and there is no consistent exercise of authority. When attempts are made to apply rules or punishments, they turn out to be arbitrary and generate increased frustration. The children are often overprotected, they are given everything they ask for almost without demands, so that they end up becoming tyrants, with low self-esteem and a low level of frustration tolerance and empathy.

This arbitrariness in the imposition of rules and authority, like the immediate satisfaction of the children's wishes, may stem both from the existence of a fusional interaction with the parent, in which the relationship takes precedence over upbringing, and from an inability to establish consistent rules because of disagreements between the parents or conflicts between them.

The authoritarian and violent parenting style often gives rise to multi-violent families, in which assaults serve to resolve conflicts. Any rebellion is sanctioned with further punishments, threats, humiliations or violence. Authority is not negotiated and is exercised rigidly and inflexibly. CPV appears here as defensive, protective, retaliatory, or linked to severe psychopathology.

A third parenting style, noted by Ibabe et al. (2007) citing Laurent and Derry (1999), would be the neglectful-absent style, in which the parents abdicate their role, openly declare that it is not up to them to set the rules and that the children (regardless of their age and cognitive abilities) are already old enough to know what they should do. They generate in the minor a pseudo-independence, when it is not a parentification. The children use violence to try to impose themselves or, on the contrary, to reject the role in which they are placed. Both types of CPV, the traditional and the new, are found here.

In summary, the new CPV is linked to permissive and neglectful parenting styles; it is mainly connected with the emotional fusion between the violent young person and the assaulted parent, and is conceptualised as a primitive attempt at distancing in a relationship where upbringing (understood as setting limits and rules and, therefore, generating frustration) is sacrificed to maintaining the relationship (extremely close, fusional, created and sustained by both).

Traditional CPV would be associated with authoritarian and neglectful parenting styles, with prior intrafamily violence or severe psychopathology in one or more family members. Violence appears as a defensive behaviour, as the learned way of resolving conflicts and lowering tension, or as an expression of psychopathology or of a transitory diminution of consciousness.

Individual factors

Besides the social and family factors – which we will address next –, there is a series of individual variables associated with the aggressors in CPV.

A series of personality characteristics has been noted that appear more frequently among aggressors: low self-esteem, egocentrism, impulsivity and absence or reduction of the capacity for empathy. The last two, frequent but not always present, are usually associated with certain disorders that we will see further on.

Low self-esteem and egocentrism are two of the characteristics most frequently perceived, though not exclusively, in violent young people, and they are linked, according to H. Omer (2007), to changes in the educational model. According to this author, the evaluations carried out – in the 1980s and 1990s – on the evolution of education towards a more permissive model, centred on positive stimulation more than on overcoming difficulties, showed that young people's self-esteem had decreased. It seems that self-esteem is linked more to the perception of being capable of facing obstacles and overcoming them by oneself than to the positive connotation of achievements. In other words, by way of example, automatically moving up to the next year at school without having passed one's exams does not generate high self-esteem, but quite the opposite; and the more things are made easy for children and adolescents, the less effort they have to make to face problems, the less self-esteem they feel.

Another factor influencing the emergence and maintenance of CPV is substance use. Although it is not a central variable – as we have already pointed out –, it favours the appearance of violent behaviours at any age and in any context.

CPV is also associated, with a certain frequency, with the presence of various psychopathology:

Attention deficit hyperactivity disorder

It is currently fashionable and is the object of strong pressure for its diagnosis to be extended to adults. It presents a series of characteristics frequently associated with CPV: impulsivity, low frustration tolerance, frequent substance use, etc.

When diagnosing this disorder in CPV, it is necessary to distinguish carefully whether its presence manifests exclusively in family relationships or in any context: work, school, the therapy room, etc. In the first case, we would have indications that the behaviour responds only to specific interactions within the family nucleus.

Personality disorders

They are frequently diagnosed in this type of phenomenon, although it must be stressed that they are not necessarily linked to it: in some cases of personality disorder, CPV will appear, and only some individuals who exercise CPV will be able to receive such a diagnosis.

Faced with these cases of antisocial personality, we must pay particular attention to the existence of a pattern of general violence or criminal behaviour, or to whether this occurs exclusively within the family. If that is the case, and although all interventions require a family approach, it would become evident that the intervention must focus on the violent relational circuit more than on the individual aggressor.

Anxiety-related disorders

In our experience – which coincides with that of certain authors such as Mouren et al. (1985) –, the new CPV is more frequently associated with anxiety-related disorders, in particular phobic and obsessional ones: children and young people who are demanding and tyrannical at home, but docile and fearful outside – above all at school. It is not unusual, in these cases, to see compulsions or phobic symptoms appear, such as agoraphobia or school phobia.

Family factors

In our view, in families presenting the new profile of CPV, a series of characteristic relational dynamics can be observed, which we propose to describe below.

To understand them, it is necessary to point out that the systemic model conceives family violence as the result of a particular interaction between the different family members, in the course of which the roles of victim and aggressor can be exchanged, as the leading systemic authors on family violence have described. As we have explained, in order to intervene with a greater chance of success in a psychotherapeutic context, we must try to understand the process by which violent behaviour is arrived at. This is achieved by introducing the concept of circularity, that is, the continuous interaction between the participants in the relationship, in a communicational exchange that has neither beginning nor end and that is mutually conditioning. If we manage to know the family interactions around the violent behaviour, we will be able to decipher it and find a meaning for it, which will allow us to talk about it and look for alternatives.

A series of family factors that favour the appearance of CPV has been described:

– A prior family experience of using violence to resolve conflicts. If what children learn in the family is that disagreements – of whatever kind – lead to an impasse, to a situation with no way out, and that violence is a resolving resource that discharges the tension created, it is likely that as they grow up, when they find themselves in similar situations, they will repeat the mechanisms learned in childhood to resolve them. In other words, families where the children have witnessed frequent maltreatment more easily produce children who maltreat – their parents, or their partners and their children in the future. In these cases, violence can appear in three different ways: a) generalised, in families where the violence is of everyone against everyone; b) directed against the aggressor, when the identification is with the victim of the earlier maltreatment; c) directed against the victim, when the identification is with the aggressor. This last case is more frequent than the previous one: the children end up mirroring the behaviour observed, and join in the maltreatment of the victim designated by the family system.

In any event, it is important to stress that, while the repetition of the exercise of violence is frequent when one has experienced it, it cannot be claimed that this experience determines future aggressive acting.

– Excessively permissive, non-normative parents, who have explicitly set out to raise their children democratically – and have often announced it publicly in those terms – (Cyrulnik, 2005); who maintain a symmetrical parent-child relationship from the earliest age, in which rules, even those of the first degree – linked to the children's immediate survival –, are not imposed but negotiated, and where everyone's opinion carries the same weight.

– Overprotective parents who, for various reasons – a long-desired child, a late child, a fragile, fostered or adopted child, etc. –, are prepared to satisfy all their children's wishes. They announce publicly that they will love their children above all else, whatever they do. In other words, they give up not only coercive behaviour, but also the other important instrument for setting limits: if you do not behave well, I will not be able to love you.

– Parents dissatisfied with their roles, who frequently express that their life is empty, devoid of meaning, and that they do not know why they go on living. Parents who had their children accidentally or against their will, and who express their displeasure about it.

– Parents who have a highly conflictual relationship and are in the habit of disqualifying each other in front of the children. The conflict can lead to a triangulation, in which the minor is used to attack the partner, by seeking the child's alliance. To this end, there is no hesitation in disqualifying the other parent, especially if they have set a limit: punishments are lifted, what the partner does is belittled, they are ridiculed and discredited. This situation leads to inconsistency and disagreement about how to raise the children, or to the inadequacy of the means used: the same behaviours bring disproportionate punishments in some cases and go unnoticed in others. Arbitrariness becomes the norm and discredits any attempt to set consistent limits.

– Parents who, for different reasons, maintain an excessively close, fusional relationship with one of the children. These are generally single-parent families, or families in which one of the parents is absent, cold or distant. This characteristic is central to the appearance of the new CPV and is found, in our experience, in the majority of cases; at least – or above all – at the start of the violent behaviour.

Dysfunctional areas

Family dysfunction is fundamentally observed in three of the main areas of family functioning (Harbin and Madden, 1979):

1. Hierarchical organisation and rule-setting

The absence of hierarchy is the main characteristic of the functioning of these families, and it is constant, whether the family is multi-violent, single-parent or with both parents present. The difficulty in establishing rules and limits is what is most striking when they come for consultation, along with the admission of failure in this area and the request that someone from outside take charge of it.

In these families, we find that one of the parents – sometimes both – has abdicated their role, that their rivalry prevents them from developing procedures for setting rules, or else deprives these of all effectiveness. This does not prevent them from attributing, almost unanimously, the abdication of the educational role to the child's personality and violent behaviour. “Nothing can be done” is the leitmotiv that not only masks the parents' inability to take on a hierarchical role, but frequently produces a lack of collaboration when it comes to addressing the problem: if others solve it, then it was not impossible and, consequently, I bear a share of responsibility for it.

Often, the parents explicitly refuse to be the ones who impose the rules – and thus to make themselves responsible for their observance –, declaring that this is not their task, or that it falls to the school or to society. This attitude favours the parentification of one of the children, to whom the little remaining authority is delegated; this child, overwhelmed by the situation, develops violent behaviours as an attempt at control, applied first to younger siblings – if there are any – and then extended to the parent(s).

The failure or the renunciation of establishing a hierarchy that would allow rules to be instituted and maintained is linked to several of the factors we noted earlier: triangulations in which the child's alliance is sought, a fusional relationship between the violent child and one of the parents, conflicts and rivalry between the parents that allow neither of them to hold the family hierarchy, etc.

2. Protection of the family image

In CPV, the family image – that of the parents as well as that of the violent children – is damaged. The parents' sense of failure in upbringing, the shame of being assaulted by a child, the protection of the family image mean that almost all the families concerned deny the seriousness of the assault and minimise its effects, even when they are public and obvious. This constitutes a serious obstacle to the adequate prevention and treatment of these cases, which only come for consultation when the assaults become public for one reason or another.

The deterioration of the family situation brings about a reaction aimed precisely at presenting the opposite image: thus the myth of family peace and harmony is created, visible in most of these families until the evidence can no longer be concealed. To hide what is going on, a secret is progressively constructed around the family's functioning.

Although the creation and maintenance of this secret are typical of any type of intrafamily violence, in CPV denial is practically the rule and reaches serious extremes: high levels of aggressiveness are tolerated for a prolonged period before measures are taken (Harbin and Madden, 1979).

The maintenance of the secret is characterised by the refusal of confrontation or open discussion about the violent behaviour, the minimisation of that behaviour and its effects, the refusal to take consistent measures (“since nothing is happening”, it would not be consistent to do so) and the refusal of outside help.

As it becomes more and more difficult to keep what is happening secret, contact with the outside progressively diminishes, which leads to an isolation that in turn favours the increase in violent behaviour. This isolation may also result from a demand of the aggressor child, who sees in it a greater ease in achieving his or her objectives. The isolation then favours the maintenance of the secret, creating a vicious circle which, at the same time, aggravates the problem.

Among the reasons that make this behaviour understandable, we can cite the myth of family harmony and peace mentioned above, the preservation of the image the subjects have of themselves, the sense of failure in the parental task and, above all, the shame that all this should be made public and that the failure of the family project should be exposed: few things dismantle our ideal of the family more than the reversal of the natural sequence represented by a child striking a parent.

The effort to maintain the secret and the supposed family harmony can lead to types of communication and relationship typical of psychotic families: communication deviance (Wynne and Singer, 1977) and pseudo-mutuality (Wynne et al., 1971).

3. Separation and fusion

Cyrulnik (2005), in his description of young aggressors, points out that “all the maltreating adolescents we have met had been deprived of the opportunity to experience the effect of separation”. He also notes that they had very often been conformist and anxious children before becoming tyrants.

In almost all the families seen at Euskarri, the emotional fusion between the aggressor and the assaulted parent appears as a stage preceding the appearance of the violent behaviour. This intense relationship may seem strange, especially if we come to know the case at an already very advanced stage, with a very damaged parent-child relationship; but, on exploring it carefully, it appears in the great majority of cases.

This relationship is more obvious in a single-parent family or one going through a phase of single parenthood, but it also occurs in families where both parents are present, owing to distance or conflict between them.

It is easier to observe in single-parent families, where the child often occupies the place of the parent's partner, sometimes producing a pseudo-incestuous relationship: confidences are exchanged, even the most intimate ones, mutual support is sought, they go out together, they share the bedroom or the bed.

This very close relationship, which suited both of them for a time, enters into crisis when growth, circumstances or the hormonal outburst of adolescence make it very dangerous. In such a close, fusional bond, the onset of violence can be understood as a primitive attempt at distancing, at escaping from a relationship now experienced as oppressive, limiting or dangerous. The secondary gains of the violent behaviour – control, power – then appear, contributing to its maintenance.

This excessive closeness between parent and child moreover entails the impossibility of creating a hierarchical relationship. If my son is my friend, my confidant, my support, I cannot place myself in a role of authority with him; this leads him to take on an independent role before his time. The same is true in families where both parents are present, but where the exercise of authority is blocked by the parental conflict or by the abdication of the paternal role. But, as Harbin and Madden (1979) explain, this autonomy is not real, since the young people continue to depend economically and, above all, emotionally on their parents. It is therefore a pseudo-independence, which makes separation even more difficult. In this situation, the use of violence can also be understood as a failed way of seeking that distance.

Onset of the violent behaviour

As we noted in the previous part, we can distinguish processes of onset and of maintenance of the violence. These differ according to the family model – multi-violent families, single-parent families, families with both parents present –, while retaining certain common aspects.

– In multi-violent families, violence appears as an almost natural consequence of the family relationship: it is necessary to exercise it in order to survive, or else it is what one has learned one can do to discharge the tension created by conflicts and disagreements. Once this mechanism is established, no other will be used until one has learned to do so and has found advantages in it over resorting to violence.

– In families where both parents are present, the emotional fusion between the violent child and the assaulted parent also appears, as a stage preceding the appearance of the violence.

In an earlier phase there is usually an unresolved conflict between the parents, generally of great intensity and long duration, which leads to triangulation, that is, to the incorporation of the child into the marital difficulties.

This triangulation resembles the one described by Selvini et al. in their book Family Games (1990). After the conflict and the marital imbroglio that leads to a situation of impasse, the child intrudes into the game, allying with one of the partners. But, in this case, the betrayal of the child that leads to psychosis does not occur: the relationship of alliance is maintained, producing a real – separation – or emotional distancing of the other partner. With time, it is the child who tries to move away, and the parent who blocks this attempt, which triggers the outbreak of the violent behaviour.

– In single-parent families, we can understand this process as a phase of what is described in the previous paragraph, even if the triangulation and the conflict may appear distant in time and already forgotten. Here, the emotional fusion is more obvious because, in the absence of the other parent, the bond can tighten more easily.

– In the last two cases, it is the difficulty of separation and autonomy that, as time goes by, leads to conflict and produces the explosion of violence. If these were families with psychotic transaction, it is likely that individuation and the growth of personal identity would be sacrificed in favour of maintaining the relationship and family harmony, to the point of pseudo-mutuality. Fortunately, in most cases, the desire for autonomy and independence, together with the aspiration to build an identity of one's own, prevail over the wish to maintain the relationship and satisfy parental needs, so that movements towards separation begin. However, far from being encouraged, these are blocked, and violence appears as that desperate attempt to achieve separation.

Here, in detail, is a typical sequence of the outbreak of violent behaviour:

A few examples drawn from different cases illustrate this process:

Parents' reactions to CPV

Haim Omer (2004) has studied in detail the reaction of parents to their children's violent behaviour and proposes to classify the responses into two types, which he calls hard and soft:

– Hard reaction or symmetrical escalation: the one in which hostility is answered with hostility, and previous violence with greater violence. The parent tries to overcome the child's assaults with the same weapons, prompting a new violent response from the child; a symmetrical escalation thus occurs, with a progressive increase in hostility. Violent resources of every kind are used: threats, shouting, humiliation, physical assault, punishment, etc. Each of the two parties believes it is acting in self-defence: the other is the aggressor. The more trapped they feel in this behaviour, the higher the levels of aggressiveness.

– Soft reaction or complementary escalation: the one in which an attempt is made to use verbal persuasion to convince the child that his or her behaviour is wrong and must change. Logical arguments, pleading and displays of affection are resorted to, in an attempt to establish an empathic connection so that the child will abandon his or her attitude. However, all that is obtained is an increase in the child's demands, emboldened by the parent's submissive attitude. The increase in violence generates more submission, producing what Omer calls complementary escalation, in which submission increases the demands.

Complementary escalation is asymmetrical and is characterised by a dynamic of blackmail. The message conveyed is one of weakness, which reinforces the use of violence to achieve one's objectives.

According to the same author, a mixture of reactions sometimes occurs, in a pendular fashion, swinging between hard and soft, since the two escalations feed each other.

The child's violence provokes, for example, submissive behaviour, which brings about an increase in his or her demands. But this attitude of responding to attempts at empathy and persuasion with ever greater demands arouses in the parent a progressive frustration and hostility, which end up erupting in an outburst of aggressiveness and violence. This reaction, instead of managing to calm the child, generates an increase in his or her aggressiveness; a symmetrical escalation then begins, which can reach such levels that the parent becomes frightened of the possible consequences, or ashamed of losing control, and returns to temporising and submissive behaviour in the face of the child's demands. It is a circular interaction, which inevitably leads to the aggravation of the problem.

Maintenance of the violent behaviour

The maintenance of violent behaviour rests on the secondary gains derived from its use. All aggressive behaviour within the family seeks to increase power and domination. In the case of CPV, this search concerns not only control, but also the attainment of objectives: coming home at whatever time one wants, getting more pocket money, being able to decide what one eats and when; in short, total freedom of action. Attempts are also made to reduce competition – if any exists – to a minimum, to restrict the parents' movements and communications in order to avoid outside interference that would jeopardise the power acquired, and to frighten the parents into helplessness.

With time, the violent child progressively orients towards power; parent-child relations become increasingly rare and of poorer quality, and those established with siblings – if there are any – deteriorate in the same way. The parents learn to ignore the child's negative behaviours in order to avoid confrontation, so that the child needs to shore up his or her power with increasingly extreme behaviours (Omer, 2004).

The aims of the aggressor child end up being the same as those of social and political violence: domination through the fear it produces, by repeatedly using violent actions with a progressive increase in the level of threat, which, according to Sluzki's (2002) schema of social and political violence, ends up producing in its victims a reaction of numbness and submission.

The effects of this behaviour within the family setting can be summarised as follows:

• The focus of the relationship centres on the violence, which drives the family members apart from one another.
• The family progressively isolates itself from friends and relatives, from the social environment.
• It is the adolescent who receives the label of problematic or conflictual, or else a psychiatric diagnosis.
• Biased perceptions of family members towards one another are recorded, generating ever more relational difficulties.
• Communication channels close, or messages become distorted, so that the conflict is denied or minimised.

Conclusion

Having defined the phenomenon of CPV and established a differentiation between the various existing types, we have pointed out, from an ecological understanding, the factors that influence both the emergence and the maintenance of this problem: a) social – with particular emphasis on their relationship with parenting styles –, b) individual – linking personality characteristics to a possible presence of psychopathology – and c) family factors, which constitute the central interest of the present work.

In this last area, although the family models in which this type of behaviour presents are made explicit, the analysis centres on the family dynamics – history of violence, parental style, relationship between the partners, dysfunctional areas, etc. – that favour both the emergence and the maintenance of the violent behaviour.

We also point out the pattern of functioning that tends to be repeated frequently in the new CPV:

• A marital conflict, current or past, which generates a significant educational deficit, particularly affecting the hierarchical area, the imposition of rules and their maintenance.
• The marital conflict includes one of the children (triangulation), who forms a close alliance with one of the parents. A very close, fusional relationship then develops between the child and the allied parent, frequently associated with a single-parent phase of the family life cycle.
• Appearance of violence as a primitive attempt, on the child's part, to distance himself or herself from this very close relationship. Its secondary gain is soon perceived, and a circuit of functioning sets in that tends towards its repetition and entrenchment.

For all these reasons, we believe it indispensable to include the relational approach in this type of problem, considering moreover, among its main objectives, the modification of the family dynamics; for if treatment is limited to the disappearance of the violent behaviour – an objective which is, of course, paramount –, its continuation and subsequent aggravation are more likely.

This will involve simultaneous work on the three areas described as dysfunctional – hierarchical organisation and rule-setting, protection of the family image, separation and fusion – without losing sight of the rebuilding of the bond between parents and children which, in these cases, is severely affected by the violent behaviours, which tend to foster a narrowed and biased perception in each of the members.