Coercive control: Impacts on children and young people in the family environment: Literature Review (2018)

Published: 01/12/2018

Author: McLeod D, Flood S


Download citation file To use these files, you need citation software installed on your device that can read .ris file format. RIS is a standardized tag format designed to allow the exchange of citation data between computer programs. The format is used on platforms such as EBSCOhost, Scopus, ProQuest, OVID to export citations to various reference/citation managers such as RefWorks, Mendeley, Endnote or Zotero.

McLeod D & Flood S. (2018). Coercive control: Impacts on children and young people in the family environment: Literature Review (2018). Dartington: Research in Practice.


Section 1: About domestic abuse and coercive and controlling behaviour

‘The fact is that for most people, and especially for women and children, the family is the most violent group to which they are likely to belong.’ (Dobash and Dobash, 1979)

It is recognised that coercive control is a key feature of abusive relationships. It is therefore essential that the conceptualisation of coercive control and the impact it has on the whole family is recognised and understood. This review is concerned in particular with coercive and controlling behaviour within the context of public and private law proceedings. In this first section we will consider the definitions of key terms such as ‘domestic abuse’ and ‘coercive control’ and seek to contextualise them before discussing the impact they have on the family and legal proceedings. This section covers:

  • Defining domestic abuse
  • Defining coercive control
  • Prevalence of domestic abuse and coercive control
  • The nature of coercive and controlling behaviours.

1.1 Defining domestic abuse

The UK Government defines domestic abuse as:

‘Any incident or pattern of incidents of controlling, coercive, threatening behaviour, violence or abuse between those aged 16 or over who are, or have been, intimate partners or family members regardless of gender or sexuality. The abuse can encompass, but is not limited to: psychological, physical, sexual, financial and emotional.’ (Home Office, 2015)

The Family Procedure Rules Practice Direction 12J on child contact arrangements and contact orders adds to this definition:

‘Domestic abuse also includes culturally specific forms of abuse including, but not limited to, forced marriage, honour-based violence, dowry-related abuse and transnational marriage abandonment.’ (Ministry of Justice, 2017)

This definition, and an overall shift nationally towards a greater understanding that domestic abuse often involves aspects other than physical violence, has led to the creation of a new legislative framework to criminalise coercive and controlling behaviour.

1.2 Defining coercive and controlling behaviour

Coercive and controlling behaviour is described and defined in different ways in the literature. Hamberger et al (2017) highlight how widespread this inconsistency is. The term ‘coercive control’ is not always directly used within research. Other terms – such as ‘power and control’, ‘domination’ and ‘controlling behaviour’ – may also be used. Follingstad (2007) and Lammers et al (2005) note that a range of overlapping behaviours have significant similarities to the behaviours involved in coercive control. These include emotional abuse, psychological abuse, psychological maltreatment, emotional blackmail, psychological aggression, coercion and verbal abuse. Given its overlap with these other forms of abusive behaviour, it is useful to think of coercive control as an intention or goal and the types of abuse a perpetrator may use as tactics to develop or maintain that control.

This idea is the central point in one of the first conceptualisations of coercive control, as expressed in the Duluth Model in Minnesota (Pence and Paymar, 1993). The model outlined the ‘Power and Control Wheel’ (see Appendix C) to describe the myriad of behaviours that perpetrators use to control victims of domestic abuse. Power and control is central within the wheel. A number of themes/behaviours around the wheel – such as threats, intimidation, financial abuse and isolation – are the ways in which a perpetrator will seek to establish and sustain control and power over the victim. In the Duluth Model, physical violence is on the outer edge of the wheel; it is a tactic that a perpetrator will use when the abusive behaviours do not work or as a way to ensure the threat of violence maintains the victim’s compliance.

The origins of the term ‘coercive control’ are to be found in debates about the nature, extent and distribution of domestic abuse (Dobash and Dobash, 1992) and whether men and women experience it to similar levels (Straus, 1974). Michael Johnson (1995, 2008) suggests a dualistic typology in which women are disproportionally affected by particular types of abuse, namely coercive control, which he describes as ‘intimate terrorism’. In various studies control is described as being goal-oriented or intentional behaviour by a perpetrator (Hamberger et al, 2017). Beck et al (2009) outline how control is a pattern of behaviour that either partner may use to manipulate or influence the actions of the other.

In the UK the Government has developed a definition of coercive and controlling behaviour.

Overview of cross-governmental definitions (Home Office, 2015)1



‘Controlling behaviour is a range of acts designed to make a person subordinate and/or dependent by isolating them from sources of support, exploiting their resources and capacities for personal gain, depriving them of the means needed for independence, resistance and escape and regulating their everyday behaviour.’

‘Coercive behaviour is a continuing act or a pattern of acts of as­sault, threats, humiliation and intimidation or other abuse that is used to harm, punish, or frighten their victim.’


This is the definition also used within Practice Direction 12J on child contact arrangements and contact orders (MoJ, 2017) and Cafcass’s private law Domestic Abuse Practice Pathway, which slightly extends the definition of coercive behaviour.

1 In a consultation ahead of its draft Domestic Abuse Bill, in spring 2018 the Government proposed a slightly revised and statutory definition of domestic abuse, based on the cross-Government definition. The new definition would cover the concept of ‘economic abuse’ rather than simply financial abuse, which ‘can be restrictive in circumstances where victims may be denied access to basic resources such as food, clothing and trans­portation’. See:

Definition of coercive behaviour in Cafcass’s Domestic Abuse Practice Pathway

‘Coercive behaviour is an act or a pattern of acts of assault, threats, humiliation and intimidation or other abuse that is used to harm, punish, or frighten their victim. Coercive control involves repeated, ongoing, intentional tactics which are used to limit the liberty of the victim. Those tactics may or may not necessarily be physical. They can be sexual, economic, psychological, legal, institutional, or all of these. By deploying these tactics the abuser can create a world where the victim is constantly monitored or criticised and every move and action checked. Victims often describe coercive control as not being “allowed”, or having to ask permission, to do everyday things; and being in constant fear of not meeting the abuser’s expectations or complying with their demands. The term “walking on eggshells” is often used.’

(Cafcass, 2017)


The cross-government definition above was introduced in 2013 when, following consultation, the Government extended its definition of domestic violence to include coercive control (Home Office, 2013). It is not a legal definition, however. Coercive and controlling behaviour within relationships was subsequently criminalised in 2015 when the Serious Crime Act 2015 created a new legal definition. In this definition (see below), the effect of the coercive control on the victim is central: effects include causing a fear of violence and having an impact on the victim’s day-to-day life.

Definition of coercive control in the Serious Crime Act 2015

Section 76 of the Serious Crime Act 2015: ‘Controlling or coercive behaviour in an intimate or family relationship’

An offence is committed by ‘A’ if:

  • ‘A’ repeatedly or continuously engages in behaviour towards another person, ‘B’, that is controlling or coercive; and
  • At time of the behaviour, ‘A’ and ‘B’ are personally connected; and
  • The behaviour has a serious effect on ‘B’; and
  • ‘A’ knows or ought to know that the behaviour will have a serious effect on ‘B’.
  • ‘A’ and ‘B’ are ‘personally connected’ if:
  • They are in an intimate personal relationship; or
  • They live together and are either:
    • members of the same family; or
    • have previously been in an intimate personal relationship with each other.

There are two ways in which it can be proved that ‘A’s behaviour has a ‘serious effect’ on ‘B’:

  • If it causes ‘B’ to fear, on at least two occasions, that violence will be used against them: s.76 (4)(a); or
  • If it causes ‘B’ serious alarm or distress which has a substantial adverse effect on their day-to-day activities: s.76 (4) (b).

We explore the outcomes of criminal cases of coercive control in section 1.3.2. It is worth noting here that the threshold to demonstrate coercive control is lower in the family court, as a person no longer needs to be in a relationship or to live with the perpetrator for coercive and controlling behaviour to be considered. In the family court, in order to be considered as associated to the perpetrator, a victim must be or have at some time been:

  • Married, engaged or in a civil partnership with them
  • Living together (including as flatmates, partners, relations)
  • A relative, including: parents, children, grandparents, grandchildren, siblings, uncles, aunts, nieces, nephews or first cousins (whether by blood, marriage, civil partnership or cohabitation)
  • Had a child together or have or have had parental responsibility for the same child
  • In an intimate personal relationship of significant duration.

Under criminal law, coercion can no longer take place once a relationship has ended and there is no further cohabitation. However, as discussed in Section 3 of this review, we know that coercive and controlling behaviour commonly persists post-separation.

1.3 Prevalence of domestic abuse and coercive control

1.3.1 Prevalence of domestic abuse

International and national prevalence

Internationally, domestic abuse is recognised to be widespread. The World Health Organization (2013) estimates that 30 per cent of women worldwide will experience physical or sexual violence in the context of an intimate relationship within their lifetime. WHO regional estimates vary between 23.2 and 37.7 per cent, however. For low and middle-income European countries (eg, Lithuania, Romania, Serbia and Turkey), the WHO estimates 25.4 per cent of women will experience violence within an intimate relationship. Estimated prevalence for high-income countries, which includes the UK, is 23.2 per cent.

Data from the Crime Survey for England and Wales suggest around 1.2 million adult women (aged 16 to 59) experienced domestic abuse in the year ending March 2017 (ONS, 2017) – nearly 1 in 17 women (5.9 per cent). On average, every week in the UK two women are killed by a current or former partner.

Prevalence for children and young people

Research for the NSPCC involving a survey of more than 6,000 children and young people [2] in the UK (Radford et al, 2011a) provides us with insight into the numbers affected by domestic abuse. They found 12 per cent of children under the age of 11, 17.5 per cent of children aged 11 to 17, and 23.7 per cent of young people aged 18 to 24 had experienced domestic abuse [3] during their childhood (see Table 1.1). Overall, the study suggests that around one in five children in the UK are likely to have been ‘exposed’ to domestic abuse.

Table 1.1: NSPCC findings on prevalence of exposure to domestic abuse and family violence during childhood (Radford et al, 2011a)


Prevalence of domestic abuse in private and public law proceedings

Hunt and Macleod (2008) estimate that around one in four children in the UK are affected by parental separation. Government statistics show that during the calendar year 2016 over 256,000 new cases began in family courts in England and Wales (MoJ and ONS, 2017), an increase of 4 per cent on 2015. These included nearly 19,000 public law cases, an increase of 18 per cent from 2015 (the number increased by a further one per cent in 2017: MoJ and ONS, 2018). There were 43,327 children involved in public law orders made in 2016 (MoJ and ONS, 2017), while between April 2016 and March 2017 Cafcass received 14,599 care order applications, an increase of 14 per cent on the previous financial year.4

Private law cases starting in the family courts in 2016 increased by 11 per cent over the previous year from 43,347 to 48,244; that number increased by a further 5 per cent during 2017 (MoJ and ONS, 2018). Almost 165,000 children were involved in the private law orders made in 2016 (MoJ and ONS, 2017).

Approximately 10 to 15 per cent of parental separations result in court applications that involve allegations of domestic abuse (Aris and Harrison, 2007). Domestic abuse remedy orders can be applied for through the courts. Remedies include either a non-molestation order (which prohibits contact or activities or behaviour of one person against the other or children) or occupation order (which defines the rights and occupation of the home). In 2016 the courts completed around 19,000 non-molestation and 4,700 occupation applications (MoJ and ONS, 2017). While this gives us an idea of the prevalence of domestic abuse in cases before the courts, the true figure is likely to be higher as a remedy order will not be sought in all cases in which domestic abuse is referenced.

For example, research conducted by Cafcass and Women’s Aid (2017) found that 62 per cent of applications to the family court about where a child should either live or spend time featured allegations of domestic abuse. And in their study, Harding and Newnham (2015) found that 86 (49%) of 174 cases that had gone through the family court system involved an allegation of domestic abuse. Incidents ranged from attempted murder through to common assault and non-physical incidents, such as verbal abuse. The study does not identify how common coercive control was within those cases; however, we provide some estimates in the next section.

1.3.2 Prevalence of coercive and controlling behaviour

Prevalence of coercive control in cases of domestic abuse

While it is possible to estimate the prevalence of domestic abuse by utilising existing national data, understanding how prevalent coercive control is within those abusive relationships is more challenging. This is due to a number of factors, including the way in which agencies currently identify and flag cases that involve coercive and controlling behaviour. Now that the legal definition is in force (coercive or controlling behaviour in an intimate or family relationship became a criminal offence in December 2015 as part of the Serious Crime Act 2015), we will begin to see more robust data emerge with the tracking of the number of prosecutions. Figures were published for the first time in November 2017. These showed 4,246 offences of coercive control recorded for the year ending March 2017 across the 38 police forces that had available data (ONS, 2017).

Table 1.2: Outcomes data for coercive control offences (ONS, 2017)5

  Prosecution commenced at magistrates’ court Sentenced through court Convicted through court Police prosecution Police cautions
Coercive control cases 309 58 59 155 5


Another way we can begin to understand how prevalent coercive control is in relationships where domestic abuse is present is by understanding the nature of perpetrators’ behaviour. SafeLives collects the largest dataset in the UK on cases of domestic abuse. SafeLives Insights data6 show coercive control is prominent in the majority of cases receiving specialist support from an Independent Domestic Violence Advocate (IDVA). It shows 82 per cent of domestic abuse victims reported ‘jealous and controlling behaviours’ from the perpetrator and 69 per cent experienced ‘harassment and stalking’ in the three months preceding intake to IDVA services (SafeLives, 2017a). This suggests a high proportion of high-risk domestic abuse cases involve coercive control. Similar findings were found in other specialist support settings (see Table 1.3).

Table 1.3: Overview of coercive control in domestic abuse cases from SafeLives Insights 2014-20177

  % of domestic abuse cases involving jealous and controlling behaviours % of domestic abuse cases involving harassment and stalking
IDVA service* 82% 69%
Health-based services 83% 61%
Helpline services 85% 63%
Outreach services 72% 59%
Refuge services 89% 75%

*IDVA service data covers 2015-2017 only

Prevalence of coercive control in private and public law proceedings

Joint research by Cafcass and Women’s Aid (2017), which analysed domestic abuse allegations in 216 child contact cases, also found high levels of coercive control. In this research the records relating to each case were also considered. Researchers found that 32 per cent of local authority domestic abuse allegations included coercive control, as did 34 per cent of police allegations of domestic abuse.

Figure 1.1: Types of abuse noted in family court cases (Cafcass and Women’s Aid, 2017)


Challenges in estimating prevalence of coercive control

However, although the SafeLives and the Cafcass and Women’s Aid data do give an indication of the prevalence of coercive and controlling behaviours, the literature highlights some of the difficulties in establishing a more precise and accurate picture of their prevalence within relationships.

Firstly, differences in how coercive control is defined within the literature (as discussed in section 1.2) affect the way in which these behaviours are measured. For example, SafeLives’ Insights data include ‘jealous and controlling behaviour’ as well as ‘harassment and stalking’ as indicators of coercive control. Regan et al (2007) and Thiara (2010) also describe coercive control as including ‘jealous surveillance’. (We discuss the nature of coercive and controlling behaviour more fully in section 1.4.)

A second challenge is that the available data are limited. Walby and Towers (2018) recently concluded that the challenge of estimating the prevalence of coercive control from existing data has not yet been met satisfactorily. In an earlier study, Myhill also concluded that existing data were not robust enough to provide an accurate picture for researchers to measure coercive control (Myhill, 2015).

It is not clear how differences in the definition of coercive control have impacted practice and service delivery. While the Home Office definition has become (since it was introduced in 2013) the most commonly used definition, we can assume there was no consistency in practice nationally prior to this.

1.3.3 Numbers of children and young people affected by coercive control within the family

There are similar challenges and difficulties in estimating how many children live in households where coercive and controlling behaviour forms part of a perpetrator’s tactics. Radford et al’s large survey-based UK study mentioned above (see 1.3.1) gives some indication of the numbers of children likely to be living with domestic abuse. It also found 3.2 per cent of under-11s and 2.5 per cent of 11 to 17-year-olds reported exposure to domestic violence in the previous 12 months (Radford et al, 2011a).

By understanding how many children live in a household where domestic abuse is present, we can then begin also to understand their experiences of coercive control. SafeLives (2017b) data suggest that emotional abuse, trying to intervene and feeling as though they are to blame for the abuse are the most common experiences identified by children who have been living with domestic abuse (see Table 1.4).

Table 1.4: Children’s experiences of domestic abuse: Insights data (n=1,695)8 (SafeLives, 2017b)

  % of cases
Directly involved in abuse of parent 6%
Child or young person (CYP) tried to intervene to stop abuse 30%
CYP feels/felt to blame 23%
CYP emotionally abused as result of abuse 54%
CYP subject to neglect as a result of abuse 18%


1.4 The nature of coercive and controlling behaviour

1.4.1 Types of behaviour

In this section we discuss the types of behaviour that a perpetrator may use which can be classified as coercive control. (In Section 2, we will build on this discussion by considering the impacts these behaviours may have – for example, the impact on parenting.)

As discussed in Section 1.2 above, definitions of coercive control differ within the literature. In practice, the Home Office definition is most commonly used, including by Cafcass (as per Practice Direction 12J). For the purposes of bringing together research, however, it is useful to understand the different array of behaviours that are commonly associated with coercive control. The Duluth Power and Control Wheel (see Appendix C) outlines a range of tactics that the evidence shows perpetrators may use to coerce or control victims. In their review of 358 female homicides, Monckton-Smith et al (2017) found control had been present in 92 per cent of cases. Behaviours that demonstrated a perpetrator’s need to control the victim were also prominent. These included stalking behaviours, which were present in 94 per cent of the cases, obsession (94%), fixation (88%) and surveillance (63%).

Table 1.5: Examples of coercive and controlling behaviour

Coercive control can include:

  • Verbal, emotional and psychological abuse
  • Control of time, space and movement
  • Continual monitoring, including by stalking
  • Rape, sexual coerciveness and control of pregnancy
  • Financial abuse and the denial of resources
  • Isolation from sources of support
  • Hurting, distressing or manipulating others (including children) to upset the victim/survivor
  • Physical violence, intimidation and threats of violence against the victim/survivor, their loved ones (including pets) and property.

(Matheson et al, 2015; Sanders, 2015; Thomas et al, 2014; Stark, 2007, 2009, 2012; Lehmann et al, 2012; Miller et al, 2010)


In a recent study Hamberger et al (2017) conducted a literature review of the evidence in relation to coercive control. They found there were three consistent elements that defined coercive control within relationships: intentionality or goal orientation in the abuser; a negative perception of the controlling behaviour by the victim; and deployment of a credible threat (see Figure 1.2). These are discussed below.

Figure 1.2: Synthesis of elements present in coercive control (Hamberger et al, 2017)

1 Perpetrators’ use of ‘credible threat’

The notion of ‘credible threat’ (Hamberger et al, 2017) as part of the array of tactics that define coercive control is referred to by a number of studies. For example Dutton et al (2005) found that coercive control is established through punishment or negative reinforcement, which also underpins the ongoing control throughout the relationship. The type of credible threat perpetrators use will often be physical violence. We can see this through models such as the Duluth Power and Control Wheel in which violence is used as an instrument to enforce regimes of power and control. Hamberger et al (2017) highlight that the threats need to be credible; threats alone do not appear to be directly related to coercive control. There has been significant research that demonstrates how perpetrators will ensure victims are aware of their willingness to deliver negative consequences, in addition to their ability to do so (Day and Bowen, 2015; Krause et al, 2006; Stark, 2007). This means there is some overlap between violence and coercive control, making their separation challenging. (This is discussed further in Section 1.4.2.)

One focus for perpetrators’ threats is use of the home – for example, threats of homelessness or making clear that they will find a victim’s new location if the victim attempts to flee. We know that families who experience domestic abuse often move location multiple times. Radford et al (2011b) found in their interviews with 37 mothers in London that the majority had moved at least once. The researchers also interviewed children and young people who had lived with domestic violence. Two siblings under nine years old had moved home eight times and changed school seven times to try to escape the perpetrator. The credible threat perpetrators pose in relation to housing is particularly pertinent in relation to child contact because many professionals (eg, social workers and police, as well as housing officers) encourage victims to move home – and preferably some distance away – to ensure perpetrators will not be aware of their location (Radford et al, 2011b). However, child contact was often the last form of control perpetrators could use to locate victims, as family courts would often grant contact.

The victim’s role as a parent is often threatened by a perpetrator in order to control the victim. For example, perpetrators may use the threat of statutory services or having the children removed if the victim does not comply. Lapierre (2010a) found some women would avoid disclosing to agencies due to fear of the potential adverse consequences of having their children removed. Stark (2007) states that threats to take or harm the children and the practice of mother-blaming are part of a perpetrator’s coercive and controlling behaviour, and can make women fearful about disclosing the abuse. Women fear being perceived as a ‘bad mother’, a fear often reinforced by the perpetrator.

2 A perpetrator’s intention or goal in the context of coercive control

The idea that coercive control is routinely driven by the intent of a perpetrator to achieve a particular goal is consistent with other research (Stark, 2007; Day and Bowen, 2015). While some studies suggest that particular behaviours may be coercive without the perpetrator recognising them as such (Dutton et al, 2005; Ehrensaft et al, 1999), most research indicates perpetrators are aware that their behaviour is coercive and have a clear goal in mind, which is to exert power over the victim. Stark (2007) explains that coercive control is used to suppress potential conflicts or challenges to a perpetrator’s authority, and not as a result of conflict or stress. In order to exert power and control over victims, perpetrators may use a range of goal-oriented behaviours as described below.


Williamson (2010) suggests behaviours are used to isolate women from their network of support. This has the benefit of ensuring perpetrators can continue to control victims with little challenge from others. Monckton-Smith et al (2017), reviewing 358 cases of homicide committed by men against women between 2012 and 2014, found isolation by the perpetrator in 78 per cent of these cases. We therefore know that isolation is a key tactic used to control victims. In their triennial review of almost 300 serious case reviews, Sidebotham et al (2016) identified a pattern – victims in a relationship with ‘aggressively controlling men who would isolate those women, impose restrictions on them, and control many aspects of their lives’ – that often had fatal consequences for children. (We explore this more in section 1.4.3.)

Isolating women from their network of support often extended to the experiences of children also. In her interviews with 15 mothers and 15 children who had lived with coercive control, Katz (2016a) found that when a perpetrator controlled the mother’s movements, this also severely restricted the children’s ability to form social lives and peer networks. It would prevent them from engaging with wider family, peers and extra-curricular activities, for example.

They [the kids] couldn’t have any friends round because he’d kick off or something. ‘Kids’ parties were another problem because he’d be accusing me of trying to “get off” with one of the dads, so parties were out the question. We couldn’t do any after school clubs because I had to be back by a certain time. Me and the kids weren’t allowed to go round to see their grandparents.’ Victim of abuse (Katz, 2016b)

In Coy et al’s (2012) study, the methods perpetrators would use to isolate victims included unrealistic expectations of chores to be completed in the home. This meant some women were too anxious to leave the house in case they had forgotten to complete a task. Katz (2016b) found women would avoid going to the supermarket or having children’s parties, as the perpetrator would accuse them of having an affair. Such isolating behaviours often meant children’s opportunities to create resilience-building relationships with non-abusive people outside their immediate family were limited or denied, as were opportunities to experience the multiple benefits that engaging positively with grandparents, friends or after-school clubs can have on children’s social skills, confidence and development (Katz, 2016b).

Perpetrators’ tactics to isolate women from support networks can affect the way victims engage with services. Women may be fearful of speaking to agencies (eg, police or social workers) in case the perpetrator finds out and it makes things worse. This is often also the case for children, who may see speaking out as ‘risky and dangerous’ (Callaghan et al, 2017). Radford et al (2011b) found some children thought speaking to teachers or friends at school about what is happening at home could make things worse. Perpetrators trap mothers and children in ‘unrealities’ shaped by manipulations, distortions, excuses, minimisations and denials which are designed to keep them confused and compliant (Williamson, 2010).

Distorting reality

Researchers have identified other common behaviours that perpetrators use to exert power and control over victims. These include making women question their own reality or, as some have described it, making them feel that they are ‘going mad’. This is referred to in some literature as ‘gaslighting’ (Tracy, 2016). In Enander’s (2011) research, victims spoke about the way in which perpetrators would present as a Jekyll and Hyde character. This meant they would behave positively in public by showing affection and charm, whilst at home behaving negatively through dominance and abuse.

Table 1.6: Coercive control tactics used by perpetrators (Coy et al, 2012)

  • Refusing to tell women their shift patterns so they were unable to plan seeing friends or childcare
  • Turning women’s alarm clocks off so that they were late for work
  • Threatening to plant drugs on women and report them to the police if they tried to leave
  • Undoing women’s housework then telling them they had not even started it.

Some women who have experienced perpetrators creating a public persona different to that displayed at home have discussed the barrier this creates in terms of their ability to disclose abuse for fear they will not be believed (Coy et al, 2012).

Deprivation of resources

There is much research exploring the intent of perpetrators’ behaviour in relation to the goal they seek to achieve. Most notably, Kelly et al (2014) coined the term ‘space for action’ in order to describe the intent of perpetrators in limiting victims’ ability and capacity to make choices. Perpetrators use an array of behaviours to do this; these were included in the development of a number of new measurement tools created through Kelly et al’s research. In limiting a victim’s space for action, perpetrators will often limit or deprive them of resources such as money, food, transport or heating.

Thiara (2010) describes perpetrators’ ‘micro-management’ in relation to their partners, which often involves limiting access to money. Based on the survey responses of 49 women who had experienced economic abuse and individual interviews with 20 of the women, Sharp (2008) identified four different themes or ‘types’ of economic abuse: interfering with women’s employment; preventing women from having money; refusing to contribute to household bills; and creating debt for which women are liable. In their more recent research, Sharp-Jeffs and Learmouth (2017) highlight the way in which perpetrators may interfere with a victim’s ability to acquire, use or maintain economic resources.

Table 1.7: Overview of economic abuse (Sharp-Jeffs and Learmouth, 2017)




Interfering with/sabotaging partner’s edu­cation, training and employment; prevent­ing partner from claiming welfare benefits.

Demanding receipts, checking bank statements; keeping financial information secret; making partner ask to use car/ phone/utilities; threatening to throw partner out of home.

Refusing to contribute towards household bills and the cost of bringing up children; spending money set aside for bills; generating costs, such as destroying property that then needs replacing; using coercion/fraud to build up debt in victim’s name.


These techniques reduce the victim’s capacity to be able to parent by limiting their access to the resources they require. Through economic abuse, perpetrators can ensure a victim uses all her resources on bringing up the children, leaving nothing to spend on other areas of her life, or ensure she relies solely on the perpetrator to be able to provide what the children need, thereby limiting the victim’s ability to leave the relationship.

Katz (2016b) describes a number of ways in which a perpetrator may deprive a mother and her children of basic resources. One mother and son reported how the perpetrator would:

  • Tell them they couldn’t touch the food in the fridge, limiting their ability to eat.
  • Unplug communications technology (eg, phones and internet) to limit their ability to contact the outside world – this also restricted the son’s ability to do his homework and the mother’s access to money through online banking.
  • Lock them in the house when he went out, preventing them from leaving the home.
  • Disconnect the electricity, limiting their ability to keep warm or wash, turn on lights or watch TV.

These examples demonstrate how a perpetrator’s coercive and controlling behaviour impacts the whole family, and not only the primary adult victim. Katz (2016a) poses the question as to whether coercive control should in fact be seen as a form of child abuse in its own right. (We explore the links to other forms of child abuse in section 1.4.3.)

Surveillance and monitoring

Monitoring is another common tactic used to limit victims’ space for action. Controlling or coercive behaviour is not confined to the home; the victim can be monitored from a distance by phone or social media (Home Office, 2015). Research demonstrates that this type of behaviour, which can be described as ‘jealous surveillance’ (Regan et al, 2007; see also Thiara, 2010), can result in accusations of unfaithfulness, resulting in women having to stay at home to avoid further conflict.

3 Victims’ negative perceptions of coercive and controlling behaviour

The third facet of coercive control identified by Hamberger et al’s (2017) literature review is that the coercive control is perceived negatively by victims. This is demonstrated by a number of studies. For example, through their interviews with over 50 women who had experienced domestic abuse, Coy et al (2008) found that all described the abuse in similar terms, which included emotional abuse, psychological tactics, mental bullying, mind manipulation, being belittled and demeaned. These descriptions highlight that victims are clear that the perpetrators’ behaviours are negative.

1.4.2 Links to violent behaviour

Coercive control and violence are inextricably linked. Perpetrators may use physical violence to create and sustain control over victims (Hamberger et al, 2017). Perpetrators may also create a fear of violence to enforce their regime, which means violence can often become a consequence of control (Johnson, 2008; Lammers et al, 2005). As discussed in Section 1.2, the Duluth Model places violence as the outer shell of perpetrators’ controlling and coercive behaviour. It suggests that victims become susceptible to these forms of control out of a fear of violence if they do not comply. Although some perpetrators may use physical violence frequently, we know others will use little or none. Rather, some perpetrators will prefer to maintain dominance over their partner through psychological abuse and the control of time, movement and activities (Westmarland and Kelly, 2013; Johnson, 2008). Perpetrators may see physical violence as a last resort because it is more likely to draw attention to their abuse (Coy et al, 2012).

A number of studies place violence, intimidation and threat as key factors in how perpetrators will establish coercive control (Beck et al, 2009; Cook and Goodman, 2006; Miller and White, 2003; Stark, 2007). This suggests that coercive and controlling behaviours alone are not violent, but rather that violence is used to enable coercive control.

Not all academics agree with this approach, however. For example, Walby and Towers (2018) describe how coercive control is now being interpreted in public debate as focused on non-physical abuse, such as emotional abuse, rather than on physical violence. They place coercive and controlling behaviour as a form of violence itself. Other researchers have also highlighted how coercive control is violent behaviour. Cook and Goodman (2006) describe how perpetrators can use violence unpredictably as a means of creating terror, which is a form of coercive control. This makes the separation of coercive control from violence a challenge.

1.4.3 Links between domestic abuse and other forms of child abuse

Domestic abuse can itself be classified as causing direct harm to children. The Adoption and Children Act 2002 (s.120) extended the legal definition of ‘harm’ (as stated in s.31 of the Children Act 1989) to include the ‘impairment suffered from seeing or hearing the ill-treatment of another’; this definition came into force in 2005. As well as harm resulting directly from the domestic abuse itself, children may experience direct harm through wider forms of abuse that are often linked to domestic abuse. For example, Humphreys and Thiara (2002) and Mullender et al (2002) both found co-occurrence of children experiencing physical or sexual abuse as well as domestic abuse. Devaney (2015) suggests three reasons why domestic abuse perpetrators may also engage in child abuse:

  1. Violent adults often may not discriminate between different family members.
  2. Adult victims may not be able to meet the physical, emotional or supervisory needs of their children as a result of physical injury and/or poor mental health.
  3. Children may be injured while trying to intervene or while being carried by the adult victim at the time of assault.

Sidebotham et al (2016) emphasise that ‘living with domestic abuse is always harmful to children’ and that it is rightly seen as a form of child maltreatment in its own right (Humphreys and Bradbury-Jones, 2015). Citing research by McGee (2000) and Mullender et al (2002), Harne states:

‘Children and young people [describe] a catalogue of fathers’ cruel and emotionally abusive behaviour towards them, such as destroying school work, school reports and toys, harming pets, not allowing children out of the house, not allowing them to speak to their mothers and not allowing friends to phone or come to the house. Some fathers are shown to deliberately emotionally abuse children and young people, insulting them and humiliating them in a similar way to their mothers.’ (Harne, 2011)

From this we can begin to piece together the ways in which a perpetrator’s coercive and controlling regime will impact on children and young people, and in some instances link to wider forms of abuse. Earlier research by Kelly (1994) described perpetrators’ behaviour having ‘double intentionality’, which often results in children being directly abused. This double intentionality means perpetrators:

  1. Abuse the mother by abusing and mistreating the children
  2. Abuse the children by exposing them to, and involving them in, the abuse of the mother.

We deal with the second of these intentions in Section 2 where we discuss the impact on the family. In this section, we will now consider Kelly’s first point by exploring the links between domestic abuse and forms of child maltreatment, namely: neglect, emotional abuse, and physical harm and homicide.

1 Domestic abuse and neglect

In their national UK study, Radford et al (2011a) found that children who had lived with domestic abuse were between 2.9 and 4.4 times more likely to experience physical violence and neglect than young people who had not lived with domestic abuse. Domestic abuse can itself result in forms of child abuse such as neglect and/or emotional abuse. Neglect, which is the most common form of child abuse in the UK, is defined in statutory guidance as the persistent failure to meet a child’s basic needs (HM Government, 2018). It can involve physical neglect, educational neglect, emotional neglect or medical neglect (NSPCC, 2017). As described by the NSPCC, this may include a child being ‘put in danger or not protected from physical or emotional harm’.9 When a perpetrator is being abusive or violent within the home, it can be argued they are failing to protect their child from emotional harm.

Despite its age, Christine Cooper’s (1985) checklist of a child’s seven basic needs is still commonly used by practitioners across the country. Security is one of those basic needs. It is defined as ‘continuity of care, the expectation of continuing in the stable family unit, a predictable environment, consistent patterns of care and daily routine, simple rules and consistent controls and a harmonious family group’ (Cooper, 1985). It is clear from the definition of domestic abuse that patterns of coercion and control do not provide a stable, predictable or harmonious family unit. Perpetrators’ behaviour can therefore be seen as neglectful.

2 Domestic abuse and emotional abuse

Definition of emotional abuse from Working Together to safeguard children: A guide to inter-agency working to safe­guard and protect the welfare of children (HM Government. 2018)

‘The persistent emotional maltreatment of a child such as to cause severe and persistent adverse effects on the child’s emotional development. It may involve conveying to a child that they are worthless or unloved, inadequate or valued only insofar as they meet the needs of another person. It may include not giving the child opportunities to express their views, deliberately silencing them or ‘making fun’ of what they say or how they communicate. It may feature age or developmentally inappropriate expectations being imposed on children. These may include interactions that are beyond a child’s developmental capability, as well as overprotection and limitation of exploration and learning, or preventing the child participating in normal social interaction. It may involve seeing or hearing the ill-treatment of another. It may involve serious bullying (including cyber bullying), causing children frequently to feel frightened or in danger, or the exploitation or corruption of children. Some level of emotional abuse is involved in all types of maltreatment of a child, though it may occur alone.’


The definition of emotional abuse is set out in the Government’s statutory guidance, Working Together to Safeguard Children (HM Government, 2018). As we have seen, s.120 of the Adoption and Children Act 2002 (introduced in 2005) also extended the concept of ‘significant harm’ to include ‘impairment suffered from seeing or hearing the ill-treatment of another’. Referrals to Children’s Services increased steadily following this amendment (Radford et al, 2011b).

SafeLives Insights data demonstrate the varying ways in which coercive control may overlap with child abuse, including emotional abuse. In addition to seeing their non-abusive parent subjected to the perpetrator’s ongoing control, we know a high proportion of children live in fear of being harmed themselves. Analysis of the Insights data collated by SafeLives (2017b) reveal that four out of ten (41%) children who had lived with domestic abuse were afraid of being harmed themselves and six out of ten (59%) feared their parent may be harmed, a finding which demonstrates emotional abuse (see Table 1.8).

Table 1.8: SafeLives Insights data of children’s experiences of domestic abuse (n=1,695) (SafeLives, 2017b)

Experience of abuse % of cases

Often at home when abuse took place


Child or young person is/was a direct witness to the abuse


CYP injured as a result of abuse of a parent


Non-abusing parent fearful of harm to child


CYP fearful of harm to self


CYP fearful of harm to parent


CYP tried to intervene to stop abuse


CYP emotionally abused as result of abuse


CYP subject to neglect as a result of abuse



Although the Insights data (SafeLives, 2017b) highlight that around half (54%) of children living with domestic abuse will be emotionally abused, Harne (2011) found few perpetrators understood that domestic abuse would be emotionally abusive to children. Similarly, in their survey of 3,234 perpetrators who had been convicted of domestic abuse, Salisbury et al (2009) found the majority had not considered the impact of their behaviour on their children. Interestingly, the desire to change behaviour to reduce the impact on children has been found to be a key motivator for behavioural change (Stanley et al, 2009), which suggests interventions should focus on raising this awareness.

Victims can also underestimate the impact a perpetrator’s non-physical tactics can have on children. For example, Radford et al (2011b) found some mothers believed their children were not at risk if they were not physically harmed. They suggest mothers are often preoccupied with trying to keep themselves and their children safe, and that the impact coercive control and abuse may have on the children can be overlooked.

‘I don’t think anyone could tell me that the children were at risk and I would believe them, because I just didn’t understand what the risk meant. For me it was like – well they are not here when he is throwing the blows so they are not physically at risk, and that’s how I understood it. When the police came because they wanted to check physically that the children were okay, and no one asked about their mental state.’ Victim of abuse (Radford et al, 2011b)

Harne (2011) also found that the way in which perpetrators interact with children differs from non-abusive fathers, often resulting in inconsistent care, which in itself could be considered neglectful or harmful. Fox and Benson (2004), for example, found that violent men often used more punitive behaviours and less positive parenting behaviour than non-violent men. This does not tell us much in terms of men who use coercive control specifically, but as we know from research findings described in earlier sections, men who use violence are likely to do so to enforce control. Therefore we can make some assumptions that violence will be used to express control over children also. In addition, behaviour that may appear to be positive parenting can in fact be a demonstration of power and control. Bancroft and Silverman (2002) discuss how abusive fathers’ ‘declarations of love’ would often reflect the father’s opinion that children are ‘emotional property’, rather than being a true expression of care for the child’s wellbeing.10 More recently, Bancroft et al (2012) found that fathers who are perpetrators of abuse have common parenting styles (see Table 1.9); a characteristic of all three styles identified by the authors is that the perpetrator had a poor understanding of children’s development and needs. Adding to the unpredictability of living with coercive control, a perpetrator’s parenting style will often alternate on a day-to-day basis.

Table 1.9: Parenting styles of perpetrators of domestic abuse (Bancroft et al, 2012)

  1. Authoritarian: inflexible and rigid, demanding obedience, controlling, allowing no critiques.
  2. Uninvolved, neglectful and irresponsible: unwilling to put aside their own desires to do the work of childcare, spending little time with children, not knowing key facts about their child’s life.
  3. 3. Self-centred: intolerant of baby’s crying or children’s noisy play, expects the children to drop everything when he wants their attention, expects the children to be caring towards him and cater to his needs without reciprocating



The Insights data (SafeLives, 2017b) highlight that 95 per cent of children living in households experiencing abuse are at home when abuse is taking place and 74 per cent witness abusive incidents directly. However, coercive control is not a series of isolated instances, but rather an ongoing and sustained pattern of behaviour. We know that victims experience coercive control as something that is ongoing and has a cumulative impact (Morris, 2009; Stark, 2007, 2009); we can assume the experience for children will not be dissimilar. A number of longitudinal studies of children who have experienced domestic abuse (and other forms of maltreatment) suggest adversity may accumulate over time, with effects becoming more entrenched. Hester and Pearson (1998), Cawson (2002) and Humphreys (2000) all found that both the severity of the violence and abuse, and the period of time over which it continues, increase the risk to children and worsen the impact and outcomes for children.

In addition, the different types of abuse a child may face can increase the negative impact for a child. Rutter (1985, 1987), for example, found six factors that were associated with poor adaptive outcomes for children, one of which was domestic abuse. The more factors a child faced, the more likely they were to experience adverse outcomes. Hughes et al (1989) considered the impact of domestic abuse based on how children experience it. They found children who experienced direct abuse, as well as witnessing abuse between parents, were more likely to display problematic behaviours compared to those who had witnessed abuse but not been directly abused. Based on this research, we can assume that children who are exposed frequently to coercive control will accumulate higher levels of harm over time. This is likely to be exacerbated for young people who also witness the abuse of parents and are harmed directly.

3 Domestic abuse, physical abuse and child homicide

It is worth noting that children will often be abused directly as part of a perpetrator’s abuse. The Insights data (SafeLives, 2017b) found that 10 per cent of children sustained injuries as a direct result of domestic abuse. Coy et al (2012) heard accounts of perpetrators using physical violence towards children:

‘When [son] turned three he started attacking him – humiliating him, verbally, psychologically, when he was crying he would slap him just because he was crying.’ Victim of abuse (Coy et al, 2012)

Dallos and Vetere (2012) suggest violence and intimidation are often directed at children as well as the adult victim, making the separation of domestic abuse and child abuse challenging. Hester (2000) adds that children can often be abused in the context of domestic abuse as a perpetrator’s strategy to further intimidate and control their partner. This is demonstrated in a quote from Coy et al (2012):

‘When I was at work, he used to hit her [daughter]. That’s why she doesn’t want to see him … he said he would kill her if she told anyone.’  Victim of abuse (Coy et al, 2012)

There is also significant overlap with child homicide (SafeLives, 2014; Hester, 2000; Humphreys, 2007; Jaffe et al, 2012; Radford et al, 2013). An overview of serious case reviews (SCRs) in England identified high levels of domestic violence in the cases studied (Brandon et al, 2009). In her study of child homicides by 13 fathers (29 children, 13 families), Saunders (2004) found that in five cases the killings appeared to have been conceptualised by the father as a form of revenge on the children’s mother. More recently, in their review of 293 SCRs Sidebotham et al (2016) found that domestic abuse was a feature in almost all instances of overt filicide. This ranged from cases with a history of overt and severe physical violence to a high number that involved aggressive coercive control. These often included cases that seemingly went ‘under the radar’ and did not raise professionals’ level of concern or include a disclosure from the victim:

‘The women would behave in ways that highlighted the fear within which they lived, but often would not disclose the abuse they were experiencing from their partners.’ (Sidebotham et al, 2016)

Section 2: The impact of coercive and controlling behaviour on the family

There is a substantial evidence base describing the impact coercive and controlling behaviour has on victims. Coercive control reduces a victim’s power to make decisions, which limits the ability to exercise independence (Robertson and Murachver, 2011), and it is widely recognised throughout the literature that coercive control impacts a victim’s whole life, including their interpersonal relationships, education and employment opportunities, and use of economic resources (Kelly et al, 2014; Bair-Merritt et al, 2010; Beck et al, 2009; Robertson and Murachver, 2011; Stark, 2007). We can assume that coercive control will also have a significant impact on children who have experienced this form of abuse, and on the relationship between children and their parents. This section aims to bring together what we know about the impact on the family as a unit. We will focus primarily on:

  • The impact on parenting
  • Children’s role within the family in the context of coercive control
  • Children’s experiences of coercive control
  • The impact of coercive control on children.

2.1 Impact on parenting

There has been significant research exploring the impacts of domestic abuse and coercive control on a victim’s ability to parent. In the instance of coercion and control, there are clear examples of perpetrators using children to abuse their mother. These include alienating the mother from the family by forming an alliance with the children and undermining the mother’s role as a parent. This section will explore how these tactics feature as part of a perpetrator’s overall coercive and controlling regime.

2.1.1 Maternal alienation (as a tactic of coercive control)

Morris (2009) highlighted the term ‘maternal alienation’ which is used in much of the literature to describe the way in which perpetrators will systematically attempt to alienate children from their mothers to prevent them forming an alliance. It is important to note here that we are not discussing parental alienation in the context of the High Conflict Practice Pathway (currently being developed by Cafcass), which focuses on alienation after separation11. Rather, we are discussing the ways in which perpetrators of domestic abuse will systematically alienate the victim from the family unit by recruiting children into the abuse. Unlike parental alienation in relation to the high conflict pathway, parental alienation within the context of domestic abuse has a clear intent to exert power or control over a victim and will often be used within the relationship as well as during and after separation. We will explore three ways in which perpetrators alienate mothers within the context of domestic abuse: forming an alliance with children to recruit them to directly abuse their mother, demeaning the mother and recruiting wider family members.

Perpetrators can disrupt the relationship between the victim and their children by using the children as part of their coercive control tactics. Bancroft and Silverman (2002) explain that perpetrators achieve power and control within the household by emotionally manipulating children into forming an alliance with them that undermines the mother-child relationship and isolates the mother within the family. The study highlights how a perpetrator may ‘joke and play, spend money on them [the children], or take them out to do things’ in order to form an alliance, which can result in children seeing the abusive parent as ‘fun’ and blaming the non-abusive parent for the abuse. In their research with victims of domestic abuse who were going through court to arrange contact, Coy et al (2012) found that perpetrators would seek to form an alliance with children by involving them in the abuse of the victim.

‘He involved my eldest son in the emotional abuse against me.’ Victim of abuse (Coy et al, 2012)

SafeLives Insights (2017b) data support the notion that in some cases perpetrators alienate the victim by involving children in the abuse, finding that six per cent of children were involved directly in abusing the victimised parent. Similarly, Radford and Hester (2006), Thiara and Gill (2012) and Katz (2016a) all found that a child’s relationship with their mother is often directly targeted by the perpetrator, with children being manipulated to act abusively towards their mother, particularly (but not exclusively) in the context of child contact. This manipulation may include the perpetrator making threats or offering bribes to ensure the child complies.

Hardesty et al (2016) describe how children can be enrolled in coercive behaviours, used as tools to exert control and be direct victims of controlling and coercive acts. In their study involving mothers who had recently separated, the researchers categorised the mothers into three groups: no violence (n=74), situational couple violence (n=46) or coercive controlling violence (n=34). They found the latter group had the poorest quality of co-parenting. That group also reported considerable levels of fear that the perpetrator would continue harassment, which was cited as the biggest barrier to co-parenting.

Many studies include women’s accounts of how perpetrators would demean and abuse them in front of the children (see, for example, Coy et al, 2012). Perpetrators deliberately disrupt the trust and emotional relationship between the victim and their children as a strategy to isolate the mother and maintain control (Humphreys et al, 2006). This isolation is exacerbated by the fact that many perpetrators will also isolate victims from external influences such as peers, wider family or support services, so children may be the only form of contact a victim has. Mullender et al (2002) found perpetrators would use a number of strategies to alienate victims from their children. Tactics can include ‘demeaning women in front of the children, recruiting children in the abuse of their mothers, and diminishing women’s abilities as mothers’ (Coy et a, 2002).

‘My little boy is supposed to be looking up to a woman who is being shot down the whole time. My little boy said I am a weak woman the other day.’ Victim of abuse (Coy et al, 2012)

Parental alienation could also include the wider family network. For example, Thiara’s (2010) research of coercive control within South Asian families in the UK found that perpetrators would deny victims a relationship with their children and often reinforce this in collusion with their own parents and siblings. The perpetrator’s family can also be involved in controlling and denigrating the victim.

2.1.2 Undermining the victim’s role as a parent

Perpetrators can undermine a victim’s parenting ability, making them feel they are not a good enough parent. Lapierre (2010a) states that ‘men’s attacks on mothering and mother-child relationships are central in their exercise of control and domination’. Radford and Hester (2006) explain that women experiencing domestic abuse will often lose confidence in their parenting ability and capacity. The tactics a perpetrator will use to undermine the victim’s parental role will often leave them feeling emotionally drained and distant and as though they have little left to give as a parent.

As explained by Bancroft and Silverman (2002), one way in which perpetrators may seek to undermine a victim’s role as a parent is by encouraging children to question their mother’s authority. This same strategy was identified by Coy et al (2012) in their interviews with more than 30 victims of abuse.

‘He would say all the emotional and psychological abuse in front of [son], undermine my authority. He’d criticise me to him.’ Victim of abuse (Coy et al, 2012)

A number of studies describe how children can be directly involved in coercive and controlling activities, which may serve the purpose of undermining the non-abusive parent’s parental role. These include isolation, blackmailing, monitoring activities and stalking; these can also be used in other ways by abusers to minimise, legitimise and justify violent behaviour (Johnson, 2008; Stark, 2007). Perpetrators will often attempt to damage children’s respect for their mother, prevent the mother from being able to provide consistent routines and attempt to turn the children against their mother.

Stanley (2011) describes how this type of behaviour can impact on children by teaching them that their mother is not deserving of respect. Children act accordingly, treating their non-abusive parent with no respect. This creates a cycle in which the mother begins to feel her relationships with her children are stressful, unhappy and beyond her control. With her authority undermined, it then becomes difficult for mothers to develop and enforce boundaries within the home. This deepens the victim’s need in relation to the perpetrator, as children may not listen to them when the perpetrator is not around.

Another way in which perpetrators of coercive control undermine a victim’s role as a parent is by constraining their ability to parent. Katz (2016a) found perpetrators would demand a high level of attention from mothers, which would be at the expense of her children. In her study she offers examples such as a victim attempting to brush her daughter’s hair and the perpetrator saying ‘You’ve spent enough attention on her, what about my attention?’ and a daughter describing how whenever her mother would sit down to play with her, the perpetrator would call the mother into another room. Coercive control limits the amount of maternal attention children are able to enjoy and reduces opportunities for fun and affection (Katz, 2016a).

Anderson and Saunders (2003) explain that the tactics a perpetrator will use to undermine a victim’s ability to parent will often include an element of unpredictability whereby the perpetrator alternates between periods of abusive and loving behaviour, blaming their partner for the violence and/or claiming that they can change. Overall the impact of maternal alienation creates significant barriers in the mother-child relationship, alienating one from the other. This is something to be considered during both private and public law proceedings, as relationships between the child and perpetrator and the child and mother may be complex to unpick as a direct result of children being systematically alienated from their mother.

2.1.3 Victims as protective parents

There is a general consensus within the literature that supporting the non-abusive parent to protect their children is often the most effective form of prevention and protection for children (Humphreys and Absler, 2011). However, it is important to understand the context of coercive control and how this can impact a mother’s ability to be protective. Stark comments:

‘A growing consensus favours intervention in relationships where there is extreme violence, stalking, or an injury to a child. But by the time abuse reaches this point, coercive control is likely to have severely eroded a woman’s personhood from the inside out, the way carpenter ants devour a house.’ Stark (2007)

When understanding the protective behaviours of mothers, we must view their actions and behaviour in the context of a restricted ‘space for action’, rather than by the standards of societal expectations. Wendt et al’s (2015) study found two main ways in which mothers will commonly seek to protect their children:

  1. Protection as an act to stop physical violence being perpetrated on their children by their partner.
  2. Protection as a constant process they engage in to create an environment that is free of violence and provides some form of stability or normality for their children.

In particular, coercive control will mean women face significant barriers in disclosing their own or their child’s suffering; rather, they may try to manage the harm and risk themselves. From their analysis of almost 300 SCRs, Sidebotham et al (2016) highlight the finding that repeated opportunities, in a safe and trusting environment, are necessary to encourage disclosure. It is important, therefore, that mothers who do not disclose domestic abuse are not seen as not being protective; rather, a mother may feel that by avoiding professionals’ input she is reducing the risk of harm the perpetrator poses to herself and her children.

Lapierre (2010b) interviewed 26 victims of domestic abuse and identified a number of ways in which they would attempt to manage the perpetrator’s behaviour in order to protect their children. For example, women tried to monitor their partners’ moods and prevent violent incidents by behaving in ways that would not upset their partners. They also tried to shield the children during violent incidents by ensuring they were in another room or away from the house. Some had challenged their partners’ violent behaviour and asked them to leave.

‘I just remember just trying to get the kids out of the way whenever he comes in. I would try and make sure that they were in bed.’ Victim of abuse (Lapierre, 2010b)

Sidebotham et al (2016) emphasise the need for professionals to understand a mother’s behaviour in the context of her experience of coercive control, which may not always make sense to professionals. Sidebotham et al’s case example (see box) of a pregnant mother discharging herself from hospital highlights the perception professionals may have. However, in this case the mother was balancing managing the risk to all her children, something medical staff were not aware of at the time. It is an example of the sorts of explanation it is important for professionals to consider if a victim of coercive and controlling behaviour takes actions that do not seem protective.

Case example from triennial review of SCRs (Sidebotham et al, 2016)

When … a pregnant mother discharged herself from hospital against medical advice and appeared to be acting against the best interests of her unborn baby, she may actually have been doing her best to protect her older children who were at home in the care of her controlling partner.

Sidebotham et al also found that in response to threats and coercion from the perpetrator, some women may play down the impact or harm resulting from the abuse, both to themselves and their children. This can be a direct strategy to avoid contact or assessments from services, which may provoke a negative reaction from the perpetrator.

Haight et al (2007) interviewed 17 mothers who had experienced domestic abuse and identified a number of strategies mothers use to protect children from the impact of the abuse (see Table 2.1). Interestingly, mothers did not feel able to develop strategies that directly challenged or sought to change perpetrators’ behaviour; rather, the strategies focused on limiting children’s exposure to the behaviour and providing a buffer to its impact by discussing the behaviour.

Table 2.1: Protective strategies described by mothers (Haight et al, 2007)

Strategy Description
Reassuring and supporting

Mothers described the importance of providing their children with emotional sup­port, including reassuring them that they are loved, they will be taken care of, they are safe now, the fighting was not their fault, and leaving was a good decision.

Limited truth-telling

Mothers emphasised the importance of providing children with factual informa­tion, but doing so in a way that does not further traumatise them. Mothers stressed, ‘don’t lie to them about it’ and, ‘answer their questions’. The challenge is to provide enough information to honestly address the child’s concerns without causing ad­ditional distress.

Instilling hope

Mothers also discussed the importance of instilling hope in their children by direct­ing their attention to the future or, if the abuse had ended, the present. Mothers spoke of the importance of helping children to ‘move on’ and not ‘dwell’ on the trauma, and of letting the child know that ‘things will get better’.

Prevention education

Mothers stressed to their children that violence is wrong, taught alternative respons­es to interpersonal conflict, and provided substance abuse education.



There is some evidence that suggests victims’ protective strategies identified by Haight et al are in line with children’s wishes. For example, when Mullender conducted a focus group with children who had experienced domestic abuse, one child said:

‘Grown-ups think they should hide it and shouldn’t tell us but we want to know, we want to be involved and we want our mums to talk with us about what they are going to do, we could help make decisions.’ (Mullender, 2005)

This is useful to consider when seeking a child’s views and wishes as part of private and public law proceedings.

In their interviews with 23 children, Radford et al (2011b) found the majority had confidence in their mother’s ability to protect them from abuse by the perpetrator. Some recognised this would be difficult, however. In one example the mother would provide her children with guidance on how to keep safe.

‘She told me to … she told me to … when he comes somewhere, hide in the other place and take your walkie-talkie and your radio with you, and then when he walks in, you turn your radio on and you tell me everything that he says on your walkie-talkie, so that’s what I did.’ Child (Radford et al, 2011b)

Children’s examples of how their mother tried to protect them included moving house or going to a friend or family member’s house until Dad was ‘in a good mood’. Interestingly, most children’s perception of what a non-abusive parent could do to protect them in the context of domestic abuse was to create physical space between them and the perpetrator (Radford et al, 2011b). This is often not possible when contact arrangements have been made through the court, however. In their interviews with children, Mullender et al (2002) also found mothers were seen as the children’s most important source of help and main source of support in coping. This suggests interventions that support positive reinforcement in this role could be crucial for the improved wellbeing of children living with coercive control.

When discussing victims of domestic abuse as protective parents, it is vital professionals avoid victim blaming and that perpetrators are held solely responsible for harm to children that has resulted from the abuse. This was a finding from Women’s Aid’s (2016) review of 19 child homicides. Within the SCRs, researchers found a number of instances where professionals criticised mothers’ choices rather than focus on limiting, disrupting or challenging perpetrators’ use of coercive and controlling tactics, which no doubt would have influenced her ability to take action.

2.2 Children’s role in families where there is coercive control

2.2.1 Triangulation

Descriptions of the role of children living in a context of coercive control differ in the literature. In some studies children are referred to as ‘witnessing’, in others as ‘exposed to’ and in some as ‘victims’. A review of 177 articles relating to domestic violence and abuse found that 85 per cent described children as being ‘exposed to’ domestic violence and 67 per cent used the term ‘witness’ (Callaghan et al, 2015). The authors discuss how much of the literature around children’s role within domestic abuse involves them being ‘collateral damage’ to adult domestic abuse. They argue that by positioning domestic abuse as a dyadic relationship we ignore the direct role children play within a context of coercive control and abuse. While there is no statutory definition, the description used most often is ‘children who experience domestic abuse’, which does take into account the varying forms the experience may take. The statutory definition of domestic abuse clearly states that victims are aged 16 and over, therefore children would not be considered victims of domestic abuse.

Dallos and Vetere (2012) suggest children are often drawn into the dynamics of the parental dyad and that we should understand the dynamic as ‘triangulation’. Within the context of coercive control, triangulation can often involve conflict, distress and force children to establish coalitions and alliances against a parent. As discussed earlier in section 2.1, perpetrators will often form alliances with children against the victim (Coy et al, 2012; Mullender et al, 2002; Thiara, 2010) so understanding how children experience this role is important.

2.2.2 Children as agents (rather than passive) in the context of coercive control

It is important to understand the role that children play in settings where there is coercive control. A number of studies have noted that in order to support children effectively, it is important they are not viewed as having been ‘exposed to’ or ‘witnesses’ to domestic abuse; rather, children should be seen as ‘human beings who live with, experience and make sense of’ (Callaghan and Alexander, 2015) domestic abuse (Mullender et al, 2002; Øverlien, 2011). If we are to consider this explanation of the role children play within domestic abuse, then we must consider also how their role as agents within the family home impacts their experience of coercive control. In their research, in which they interviewed 21 children about their experiences of domestic abuse, Callaghan et al (2015) echo these findings. Children recounted the disruption and distress they experienced as a consequence of coercive control within the family setting. The authors argue this demonstrates that children are not passive, but immediately involved and affected by coercive and controlling behaviour. Perpetrators’ behaviours are not limited to targeting the adult victim, they argue; rather, the entire family is targeted.

Figure 2.1: Overview of children’s involvement in domestic abuse (Mullender et al, 2002)


Mullender et al (2002) categorised children’s involvement in domestic abuse in three ways (see Figure 2.1). This conceptualisation supports Callaghan et al’s (2015) idea that children should not be seen as passive in the context of coercive control, and suggests children may become directly involved in abuse by seeking help or intervening. In some examples, Mullender and her colleagues found children would climb through windows to get to their parent or siblings to intervene or seek help. Similarly, SafeLives’ Insights data (2017b) found that 30 per cent of children intervened directly and 8 per cent sought help by calling the emergency services. As discussed in previous sections, whether a child directly intervenes in the abuse, seeks help for the non-abusive parent or overhears abuse, they will be affected. It is vital that assessments consider the risk attached to each of these roles, particularly in light of the example provided by Mullender et al (2002) in which children are engaging in dangerous behaviours (climbing through windows) to protect their non-abusive parent. Working with children to develop safety plans has been shown to make children feel more empowered and may a useful consideration in the context of contact.

2.3 Children and young people’s experiences of coercive and controlling behaviour

The literature is limited in terms of how children with lived experience of domestic abuse actually experience the abuse. Those limitations are even more apparent in relation to how children experience coercive control (Callaghan et al, 2015). In this section, we explore what literature there is to understand how coercive control within the family is experienced by children and young people.

2.3.1 Awareness of the perpetrator’s use of coercive control

There is some evidence that children are aware of the coercive and controlling behaviours perpetrators use against the non-abusing parent. The Understanding Agency and Resistance Strategies (UNARS) study explored the views and experiences of young people (Callaghan and Alexander, 2015). The project involved interviews with 110 children across four European countries: Greece, Spain, Italy and the UK. Children demonstrated an understanding of the patterns of abuse and control that existed within their home. One young person in the UK described his awareness of his father’s use of control over his mother, stating that the father would not allow the mother to go out of the house to see friends (Callaghan et al, 2015). He recalled how the perpetrator would either say he needed the mother’s help (eg, to clean or cook a meal) or would resort to violence by throwing things at her. This demonstrates children’s capacity to understand the nuances of coercive control. In this example the young person was aware of the tactics the perpetrator would use to control, as well as the resulting effects (eg, his mother not going out).

Callaghan et al (2015) found children have a sophisticated understanding of control dynamics, as well as an understanding of the impact of such behaviours. For example, children recognised that subtle controlling behaviours, such as a perpetrator’s desire to know all aspects of family activity, were used to prohibit both the non-abusing parent’s and their own space for action. In one example, a young person explained how his father would ‘do things to scare us’ if he did not feel he was aware of what everyone in the family was doing. The young person explained this would be to prevent him, his siblings and his mother from going out. This example highlights how coercive control from perpetrators will also often directly impact on children within the family, and that children are often aware of this and try to operate within the unreasonable boundaries set by perpetrators.

In Callaghan et al’s (2015) study, children are also aware of and explain how perpetrators continue to control the family following separation. In one instance, a child explained his father’s use of gifts to manipulate him into sharing information about his non-abusive parent. He described how the perpetrator would either ignore him or ‘be like … oh come on, I’ll get you something’ when he wanted the child to tell him about what the non-abusive parent had been doing. In another example, a child explained how the perpetrator would wait or be present in locations where he knew the victim and the children would be, such as the shop. The child in this case recognised that this was not a coincidence but a deliberate tactic of the perpetrator to reinforce his presence. Both of these examples highlight the awareness children have of coercive control in the family, and the impact it has on themselves and family life.

2.3.2 Being used as a pawn

A common feature in terms of children’s experience of coercive control within the family is their experience of being used as a pawn by the perpetrator. As Bancroft and Silverman (2002) explained in their study, even seemingly positive behaviour from the perpetrator may in fact arise from the perpetrator’s view of the child as a ‘trophy’ or instrument in the control of the victim. It is therefore understandable within this context that children will be used as pawns. Callaghan et al (2015) found examples of abusive partners trying to involve children in hurting the adult victim, either emotionally or physically, or encouraging children to act as an informant about the non-abusive parent. Thiara and Gill (2012) also report examples of perpetrators encouraging children to hit their mother.

One child, who was encouraged to lie about an argument to undermine the non-abusive parent, discussed how this changed his view of the perpetrator from being a ‘nice guy’ to a ‘really bad person’ (Callaghan et al, 2015). This shows how aware children are of being used as a pawn to further control their non-abusive parent. In a separate example, one boy discussed how the perpetrator would use information and knowledge he provided to control the victim. This included information about who the victim had spoken to or where she had been.

Children’s perception of being used as pawns provides some useful insight. In the previously mentioned example, the child discussed how aware he was of the power he possessed as the one with knowledge and information to share, while the perpetrator was in a weaker position of wanting the information. Although children are sometimes used as pawns, this demonstrates that some are aware of this and a small number may even resist such coercive control. We explore how young people may resist coercive control in Section 2.5. An important point in relation to young people’s perceptions, however, is the level of risk perpetrators may pose to them. As discussed earlier (see Section 1.4), there are clear links between coercive control and domestic abuse and various forms of child abuse, including physical violence. Therefore professionals should be aware of the potential for violence if children resist control.

2.3.3 Managing coercive control

As SafeLives Insights data (2017b) demonstrates, a number of children become involved in trying to manage the ways in which the perpetrator will seek to control the non-abusing parent and family unit, with 30 per cent intervening and 8 per cent calling emergency services. Callaghan et al (2015) also found children reported how they would disclose to others and seek help in an attempt to manage coercive control. One child discussed calling her grandmother for assistance when she was concerned. In that example the authors describe the child’s central role in securing her family’s safety.

There are also more subtle ways in which children will seek to manage a perpetrator’s behaviour. One child recounted refusing money and gifts from the perpetrator because they knew there would be strings attached that would involve their support in controlling their non-abusive parent: ‘I’m not going to like try to be buyed’ (Callaghan et al, 2015). Recognising the behaviour as intended to control the non-abusive parent and actively resisting it demonstrates the protective role some children play within families where there is coercive control. Other children describe managing their own safety by removing themselves from the room or home when physical violence was taking place.

2.4 The impact on children and young people

This section will explore the literature base with regard to the impact of coercive control. We know that living in a climate of ongoing fear and controlling behaviour is likely to have a negative impact on children’s health and development (Radford, 2011b). Overall, however, in the current literature the voices of children who have lived in households where there is coercive control are limited. But there is research in relation to the wider impacts of domestic abuse and the types of behaviours a perpetrator may exhibit, some of which overlap with coercive control.

Practice Direction 12J on the impact of domestic abuse on children

‘Domestic abuse is harmful to children, and/or puts children at risk of harm, whether they are subjected to domestic
abuse, or witness one of their parents being violent or abusive to the other parent, or live in a home in which domestic abuse is perpetrated (even if the child is too young to be conscious of the behaviour). Children may suffer direct
physical, psychological and/or emotional harm from living with domestic abuse, and may also suffer harm indirectly
where the domestic abuse impairs the parenting capacity of either or both of their parents.’

Practice Direction 12J – Child contact arrangements and contact orders: domestic abuse and harm (Ministry of Justice,

We can use this research to understand the ways in which coercive control affects the outcomes for children. Callaghan et al (2015) highlight that research tends to focus on children’s experiences of coercive control, but their actual voices are under-represented through qualitative studies. Callaghan and Alexander (2015) argue that qualitative studies are needed that give voice to children’s own experiences ‘if we are to avoid the risk of oversimplifying their responses, and of reducing their experience to pathology and damage’.

2.4.1 Psychosocial impact on children

For children who experience coercive control, the impact is likely to be similar to adult victims; children are likely to suffer from limited opportunities to choose, feel free and develop a sense of independence and competence (Katz, 2016b). A large proportion of the literature indicates that coercive and controlling behaviour within the household will have a negative psychosocial impact. For example, several studies discuss the increased risk of children experiencing poor mental health (Helweg-Larsen et al, 2011; Fergusson et al, 2005; Meltzer et al, 2009); others focus on the impact on physical health (Bair- Merritt et al, 2006). Mullender et al (2002) describe how children who have lived with the emotional abuse that forms part of a perpetrator’s tactics of domestic abuse express higher levels of anxiety, fear and a reluctance to trust others. Buckley et al (2007) confirmed these findings, adding that children feel a lack of security and safety. This is echoed by the children in Callaghan et al’s (2015) study who describe how their interpersonal relationships, both inside and outside the home, were ‘managed in relation to the controlling and abusive relationships that characterised their home life’.

Coercive control can also impact a child’s ability to function within an educational setting, with some studies discussing challenges experienced in school, such as learning difficulties and disengagement (Byrne and Taylor, 2007; Willis et al, 2010). Interestingly Lepistö et al (2011) and Baldry (2003) also found that children living with coercive control are at higher risk of bullying, both as a victim and perpetrator. This is not surprising as the concept of polyvictimisation presented by Finkelhor et al (2005, 2009) suggests children exposed to abuse at home are subsequently vulnerable to other forms of victimisation.

In their interviews with 21 children in the UK (12 girls and 9 boys), Callaghan et al (2015) found that children are significantly affected by coercive control. The authors argue that children can and should be described as direct victims of coercive and controlling behaviour. We can therefore also assume some overlap between the impact that domestic abuse has on adult victims and children.

2.4.2 Coercive and controlling behaviour constraining children

As discussed above, Callaghan et al (2015) suggest we should understand the impact of coercive control through seeing children as direct victims. As such, it is important to understand the ways in which coercive control within the household constrains children. Katz (2016b) conducted a study interviewing 15 mothers and 15 children to understand the impact of coercive control on children. She found that the most common impacts were:

  • Control of time and movement within the home
  • Deprivation of resources and imprisonment
  • Isolation from the outside world.

Callaghan et al (2015) presented the effects described by the children they interviewed, which included constrained use of space within the home and constrained self-expression. Self-constraint defined children’s experience of coercive control; in order to protect themselves, they learnt to manage what they said and did to prevent themselves from being ‘too visible, too loud, too noticeable’ to the perpetrator.

Swanston et al (2014) describe children living with domestic abuse as ‘miniature radar devices’ who would attempt to predict the perpetrator’s behaviour and response, despite this often being unpredictable. This was mirrored in Callaghan et al’s (2015) research. Children discussed how they would attempt to read the situation at home. One young person talked of attempting to read the perpetrator’s mood and how he was likely to react to whatever she might say, however innocuous. She described herself as ‘always thinking ahead’. Another child described how she would manage her day in preparation for the perpetrator coming home from work. She would keep an eye on the clock, watching for ‘coming home time’ when she knew she would have to limit her self-expression.

Constraining their own behaviour is a clear coping strategy employed by many children as a way of creating a sense of safety. It is therefore important that this hypervigilance is noted as a clear impact of coercive control at home, and professionals consider how this increased vigilance and constraint may affect a child’s wellbeing.

2.4.3 Feeling safe at home

By moving away from the traditional conceptualisation of domestic abuse and recognising the wider impacts of ongoing coercive and controlling behaviour, we can learn something of the daily impact that children experience. Swanston et al (2014) highlight children’s attempts to create spaces of safety within the home. One child described how she would remove herself and her sibling to a safe place within the home when she anticipated an escalation that would result in violence. If we focus only on incidents of violence and their impact, we will miss these wider behaviours used by children.

Callaghan et al (2015) explore how children often attempt to manage space within the house to avoid the perpetrator. In their interviews with 21 children, all were able to identify where the safe and ‘risky’ spaces in the home were. Communal areas were generally seen as unsafe. Feeling that spaces were unsafe often extended outside the home. One child discussed the outside world as feeling unsafe as she always needed to consider the risk the perpetrator posed to her; at home she clearly understood the spaces that were safe.

2.4.4 Impact of domestic abuse and coercive control at various ages

There is no significant research to suggest at what age or developmental stage a child’s experiences of domestic abuse are likely to have a greater or lesser impact. However, there has been substantial research that suggests the ways in which abuse is likely to impact on children will vary according to their age.

Due to their developmental stage and inability to communicate verbally, infants are only able to indicate distress through health or development outcomes. Compared to those who have not experienced domestic abuse, children under the age of one who witness such abuse demonstrate poor health, poor sleeping habits and excessive screaming, for example.

Children can begin to show symptoms of trauma from a young age; in Bogat et al’s (2006) research, 44 per cent of mothers of one-year-old infants reported signs. These were more likely to be present if the mother also had symptoms of trauma herself.

Pre-school age children are particularly vulnerable to the impact of domestic abuse. This may be due both to cumulative exposure to domestic abuse and very young children’s limited capacity either to escape or manage it because of their limited intellectual or verbal ability (Stanley, 2011).

In their UK-based study Meltzer et al (2009) considered the impact of domestic abuse on school-age children and found that experiencing domestic abuse was correlated with the onset of conduct-related disorders. Other studies have confirmed the anxiety school-age children feel at having to keep domestic abuse a secret for fear of the consequences from the perpetrator if they tell people (see, for example, Callaghan et al, 2017). Coy et al (2012) discuss how this secrecy can impact on private and public law proceedings, as children are conditioned to maintain secrecy about their home lives and will often maintain this by not speaking to professionals or lying to disguise the perpetrator’s behaviour.

2.5 Children’s resilience and resistance of coercive control

As discussed above, children employ a range of coping strategies when living with coercive control. And just as adult victims use an array of strategies to cope with coercive control both within and after the relationship, this is true for children also (Crossman et al, 2016). Many of those strategies are intended to resist control. As such, they demonstrate the resilience of children. As much of the literature is focused on the negative impact of coercive control on children, it is important also to recognise children as ‘creative agents’ who respond, adapt and change to manage the effects on them (Callaghan et al, 2015). Katz (2016a) found that resistance often took the form of children finding ways to maintain as many elements of ‘normal life’ as possible and to maintain a close mother-child relationship when possible. Devaney (2015) differentiates between two models of ‘coping’ that children use within the context of domestic abuse (see Table 2.2).

Table 2.2: Types of coping exhibited by children (taken from Devaney, 2015)

Emotional-focused coping Problem-focused coping

Managing and reducing stress, such as a child withdrawing from violent episodes and distracting themselves by listening to music or playing with toys.

Changing the problematic situation – for example, by attempting to intervene physically, by distracting the violent parent or summoning help.


The strategies children use will differ based on the context and situation as well as the individual young person’s own internal and external resources. For example, younger children are more likely to use emotional-focused coping, such as disengagement, rather than problem-focused coping.

Callaghan et al (2015) found that in resisting control, some children extend their resistance as a means of protecting the non-abusive parent. For example, one child discussed how he would lie to the perpetrator about his mother’s access to money. He was seeking to protect his mother because he recognised the financial control the perpetrator would regularly exercise over her. Another child discussed the way in which the perpetrator would attempt to use financial incentives, such as gifts, in exchange for information about the non-abusive parent once the relationship had ended. In recounting this example, the child clearly understood the perpetrator’s intent to control and took active steps to resist.

Many children went beyond resisting coercive control and also found ways in which to build a positive sense of self (Callaghan et al, 2015). For example, one child describes how she saw the possibility of achieving independence in her adult life. Having lived with the effects of coercive control, she sees for herself the possibility of a life of independence, particularly financial independence, which she sees as protecting her from dependency and control (Callaghan et al, 2015). This sophisticated processing demonstrates that children are capable of building resilience after coercive control. Interestingly they found that through expressive strategies such as art, music, sport and play, children were able to feel safer and a sense of more control of their environment (Callaghan and Alexander, 2015). Some children also discussed the notion of telling people about the abuse at home, which they often felt would be too risky. Callaghan and Alexander (2015) identified a number of ways in which children and young people would resist coercive control within the home – see Figure 2.2 on the following page.

Figure 2.2: Ways in which children and young people resist of children coercive and controlling behaviour at home (Callaghan and Alexander, 2015)

Section 3: Coercive and controlling behaviour, child contact and court

We know that domestic abuse and coercive control do not stop at the point of separation. In fact some research has shown abuse can increase. Stanley et al (2010) found that post-separation abuse was a factor in approximately half of domestic abuse related police calls, whereas Hester (2009) found they made up around 30 per cent. There is also significant research that highlights separation as being a point of heightened risk for homicide (Jeffries, 2016; Humphreys and Thiara, 2003). Only a minority (one in ten) of parental separations reach the family court system as a means of settling disputes over residence and contact with children; however, domestic violence is commonly raised in proceedings (Coy et al, 2012; ONS, 2008).

In their analysis of more than 300 applications for contact orders after separation, Hunt and Macleod (2008) found domestic violence was the most common welfare issue raised; allegations of domestic violence perpetrated by the non-resident parent had been made at some point in exactly half the sample cases. They highlight that coercive and controlling behaviours in particular are rarely considered credible, and even less so if there is no physical violence. There is considerably less research exploring the role of ongoing coercive control at the point of separation and during private and public law proceedings. In this section we will explore the available literature in relation to the ways in which coercive control can feature after separation through the court systems and during child contact. We will cover:

  • Coercive control after separation
  • An understanding of coercive control and child contact
  • Child contact as a tactic of coercive control.

3.1 Coercive control after separation

The largest predictors of intimate partner homicide are emotionally abusive and controlling behaviours and victim-instigated relationship separation (Jeffries, 2016). It is widely recognised that separation is a point of high risk in the context of domestic abuse. Humphreys and Thiara (2003) highlight the difficult situation many victims are in, often being required or advised to leave abusive relationships in order to safeguard children, despite separation being a trigger for escalation. In their triennial review of SCRs Sidebotham et al (2016) found that coercive control almost always continued following parental separation and was often extended by child contact arrangements. In some cases where child contact restrictions were placed on the perpetrator, these appeared to be the trigger for the fatal incident.

Separation can result in an increase of physical violence and potential lethality (Ansara and Hindin, 2010; Johnson, 2008; Campbell et al, 2003). Research shows that domestic violence frequently continues post-separation and often increases in severity (Jeffries, 2016). Toews et al (2003) also found that perpetrators who had not previously been violent were at increased risk of using physical violence as a way to maintain control and dominance. In fact, coercive control was found to be the defining element as to whether ongoing abuse and harassment would occur (Hardesty and Ganong, 2006).

The risk can increase when victims remain in contact with the perpetrator in order for contact with children. A review by Women’s Aid (2016) of 19 child homicides (12 families) found many agencies continue to mistake parental separation as an indicator that abuse will end and the risk to the mother and children will reduce. In reality, the researchers found that in most cases the perpetrator would remain in contact with the mother for the purpose of child contact and the risk would continue or increase after separation. In 9 of the 12 cases analysed, abuse continued after separation. In one instance it continued in the form of abusive and threatening telephone calls made by the perpetrator from prison; on release the perpetrator was granted a residence order through the family court for his children’s care. In another case, the victim and her children had escaped the perpetrator and settled in a refuge, but ongoing contact was imposed following the perpetrator’s application to the family court. It is essential, therefore, that child contact arrangements following separation are recognised as a point of high risk (Hester, 2011). In Section 3.3 below we will explore how child contact is used by perpetrators as a tactic to further control the victim.

3.2 Understanding coercive control and child contact

3.2.1 Contact and risk

There has been significant debate about the relationship between perpetrators and their children. Specialist women’s services have highlighted that children will always be affected by coercive control and abuse within the household. There have also been concerns about the risks associated with child contact for both the victims and their children. In particular, a number of studies have highlighted that the adult victim and their children are at risk of control, threats, harassment and violence at the point of handover during child contact (Hester and Radford, 1996; Radford et al, 1997; Thiara, 2010).

Saunders (2004) reviewed the deaths of 29 children in 13 families who were killed as a consequence of contact arrangements between 1994 and 2004 in England and Wales. She found domestic violence was a feature in 11 of the families. Following this review, Practice Direction 12J was developed to ensure the prioritisation of children’s safety in contact decisions through the family courts. More recently Women’s Aid (2016) conducted a similar study that explored 12 cases involving the deaths of 19 children, each killed by a perpetrator who had formal or informal contact with children after separation. A key finding was that professionals did not understand that separation was often a key risk point in domestic abuse.

3.2.2 Perpetrators’ motivation and intent

Child contact can often be used as a means to further control the adult victim, so it is essential that professionals understand the intent of perpetrators throughout the process. Thiara and Gill (2012) highlight how the initiation of contact proceedings can be used to continue to control the family. In one example from Callaghan et al (2015) the perpetrator would continuously move contact dates to coincide with days when the children were due to visit extended family members. For example, the perpetrator would pick Sundays for contact, as that was the day the children were due to see their maternal grandmother. The daughter described the perpetrator’s intent as being to disrupt contact with her wider family and inconvenience the non-abusive parent. She described the frequent formal and informal changes to contact as the perpetrator ‘messing my mum about’. Her non-abusive parent ‘didn’t have a choice’ because the perpetrator would go back to court if she disagreed. This type of behaviour prevents children’s ability to maintain other important relationships also. Coy et al (2012) argue that courts should determine the intent of the parent seeking contact and consider whether it is to continue a pattern of coercive control.

Thiara (2010) discusses how perpetrators will ‘rigorously pursue’ contact with their children through the family court system, only to fail to attend contact visits once they have been agreed. This demonstrates the challenge in trying to understand or determine a perpetrator’s intention and if contact or control is the end goal. Anderson (1997) found that perpetrators’ intentions were often triggered by external factors beyond their desire for contact, at points where they may be experiencing a loss of control. Similarly, in Coy et al’s (2012) research interviewing victims of coercive control who had been though the family court system, applications for contact in six cases were clearly a response to the victim’s application for an injunction. In one example, the victim sought to protect herself from her ex-partner’s coercion and threatening behaviour through an application for a non-molestation order. In challenging this, the perpetrator additionally initiated an application for child contact, which had not previously been a desire.

It is not only victims of coercive control who recognise perpetrators’ intentions to use child contact as a means to continue to control. Coy et al’s (2012) study included an online survey of 113 legal professionals (53 solicitors, two trainee solicitors, 56 barristers and two legal executives) who had supported both victims and perpetrators through the court system. They found that 79 per cent of legal professionals thought that ‘regaining power and control’ was the intention underpinning the perpetrator’s application for child contact.

‘Some perpetrators get the message that the relationship is over and are genuinely interested in their children. Others use it as a mechanism to keep back into the relationship, others to undermine the victim’s confidence and self-esteem, others to make the victim feel trapped and some to “get at them” further by undermining them as a parent.’ Barrister (Coy et al, 2012)
Furthermore around 30 per cent of legal professionals thought revenge was the intention and 18 per cent suggested the perpetrator had financial motivations. When asked how much they agreed with the statement ‘some perpetrators of domestic violence use contact proceedings to exert power and control over the victim’, half (50%) of the participants ‘agreed somewhat’ and two-fifths (44%) ‘completely agreed’ (Coy et al, 2012).
‘Often, and particularly when the victim has had “the nerve” to leave the abuser, the abuser utilises the proceedings as a means to control the victim and simply to see them.’ Solicitor (Coy et al, 2012)

3.3 Child contact as a tactic of coercive control

Building on the research findings in relation to perpetrators’ intentions when pursuing child contact, it is also useful to understand the ways in which the contact itself can be used as an additional tool to exert power and control over victims. In explaining how her father would routinely change arrangements, one child described the perpetrator’s use of contact as ‘messing my mum about’ (Callaghan et al, 2015). By regularly challenging child contact arrangements, both formally and informally, perpetrators are able to exert power over the adult victim and children. The child in the example above was able to highlight how these continuous changes and the perpetrator’s behaviour impacted on the family through constant disruption, interfering with the children’s ability to maintain wider family relationships.

Another child was aware of how the perpetrator would continuously breech contact arrangements and use financial means to try and entice him into unscheduled contact (Callaghan et al, 2015). This type of behaviour creates uncertainty for the family who are always awaiting changes in contact and are unsure when the perpetrator may ‘just turn up’. Harrison (2008) found perpetrators would often fail to adhere to contact arrangements deliberately in order to disrupt the victim’s ability to plan and continue routines, thereby controlling what they can and cannot do.

Several studies, including Thiara (2010) and Thiara and Gill (2012), demonstrate how children are often used within child contact to further control the victim. Holt (2011) explains how perpetrators will manipulate children, make negative comments about the non-abusive parent and ask them to repeat abusive messages. Edleson et al (2003) also highlights how perpetrators can use children as a source of information. They will probe the children for details about the non-abusive parent’s whereabouts and daily life. Crossman (2016) found that a high number of victims experienced ongoing stalking, harassment and threats during contact arrangements.

In addition to the more overt forms of coercive control, Callaghan et al (2015) also highlight how perpetrators’ attempts to maintain ongoing control can be more insidious by simply making the victim aware of their ongoing presence.

‘Once I broke down and he was driving round and round my car. I was waiting for my [new] partner to come and help but I couldn’t get out the car as he kept driving past me and I felt very vulnerable. There have been other times he would turn up outside school when I was picking up my son and follow me back to my house. It’s just pure intimidation to let me know he is still around.’ Victim (Coy et al, 2012)

Some victims and children describe a sense of ‘being watched’, with the perpetrator’s unwanted contact with victims in public places producing social embarrassment, a tactic employed to secure the victim’s compliance (Callaghan et al, 2015).

3.4 Using the evidence of coercive control in planning contact through public or private law proceedings

From the evidence set out above we can conclude the following key messages, all of which are important to consider when planning contact between children and their abusive parent.

Key messages

  • Separation is likely to be a time of increased coercive and controlling behaviour from perpetrators and a point of heightened risk for children and the non-abusive parent.
  • Perpetrators who have never been physically violent before are likely to feel that violence is now necessary to maintain and sustain control.
  • From the perpetrator’s perspective, the purpose of child contact may be to further control the victim – their tactics can include regularly changing the contact arrangements, using contact to disrupt the child or victim’s routine, failing to attend contact once it’s secured, or using the children to pass messages between perpetrator and victim.
  • The trigger for initiating child contact arrangements may be related to the perpetrator experienc­ing a loss of control – this might follow reduced contact from the victim or the victim starting a new relationship, for example.
  • Perpetrators may display coercive control with children during contact visits – this can range from undermining their mother, calling them names or offering financial bribes to provide information or ‘spy’ on their mother.
  • Perpetrators may use contact to demonstrate their power – for example, by contacting children directly, arranging informal contact with children or turning up outside agreed contact times.

Baynes (2015) has developed a tool for practitioners (based on work by Elsbeth Neil) to use to support contact planning in public law cases. It sets out a five-step approach informed by available research (see Figure 3.1) and outlines key questions to consider in relation to contact. In this review, we have considered a number of points practitioners may wish to consider based on what we know about coercive control (see Table 3.1).

Figure 3.1: Planning contact arrangements in public law cases (Baynes, 2015)


Table 3.1: Considerations to identify coercive control through public law arrangements

Purpose Risks and strengths Contact plan Provisional support plan Review

Consider the intention of the perpetrator in seeking contact. For example, do they continue to have contact with the non-abusive parent? Have there been any relevant changes in either their or the non-abusive parent’s life (eg, a new partner; the non-abusive parent has moved)?


Child has not witnessed/ does not imitate violence.

Child has positive memories – perpetrator may have formed an alliance with children through bribes to abuse the non-abusive parent. Practitioners should consider this when discussing ‘positive’ memories. Consider whether most positive memories relate to when one parent was not present?

Child freely wants contact, is not afraid – practitioners should consider the lasting impact of coercive control; is the child clear they will not be in trouble or at risk if they do not want contact?

Relates to child in positive, non-abusive way – consider manipulation of children.

Accepts harm caused to child – consider questions relating to control. Do they blame the non-abusive parent for harm? Do they discuss non-physical violence and the harm that causes?

Does not use contact to undermine/threaten or cause conflict with carers – consider use to control non-abusive parent too – eg, through abduction or harm to child.


Consider the role perpetrators may have in coercing or manipulating children into too much contact too soon.

Supervised contact can mean perpetrators adopt a non-abusive public persona. A plan should be in place to identify disguised compliance – eg, does this person reflect how the child describes them, how they remember them?

This should include safety planning with the child (if age appropriate) or with key people (eg, placement family) to ensure children will be safe.

Recognition that children may have been manipulated or coerced – asking questions to identify whether the perpetrator discussed the non-abusive parent? Details about their whereabouts or whether the child has contact with them?

Consider inconsistencies or hesitations in the young person’s account – is there fear of ‘saying the wrong thing’, which could indicate the perpetrator encouraging the child to lie to professionals, placement family or others?


Appendix A: List of coercive and controlling behaviours (Home Office Statutory Guidance Framework, 2015)

Types of behaviour

The types of behaviour associated with coercion or control may or may not constitute a criminal offence in their own right. It is important to remember that the presence of controlling or coercive behaviour does not mean that no other offence has been committed or cannot be charged. However, the perpetrator may limit space for action and exhibit a story of ownership and entitlement over the victim. Such behaviours might include:

  • Isolating a person from their friends and family
  • Depriving them of their basic needs
  • Monitoring their time
  • Monitoring a person via online communication tools or using spyware
  • Taking control over aspects of their everyday life, such as where they can go, who they can see, what to wear and when they can sleep
  • Depriving them of access to support services, such as specialist support or medical services
  • Repeatedly putting them down such as telling them they are worthless
  • Enforcing rules and activity which humiliate, degrade or dehumanise the victim
  • Forcing the victim to take part in criminal activity such as shoplifting, neglect or abuse of children to encourage self-blame and prevent disclosure to authorities
  • Financial abuse including control of finances, such as only allowing a person a punitive allowance
  • Threats to hurt or kill
  • Threats to a child
  • Threats to reveal or publish private information (eg, threatening to ‘out’ someone)
  • Assault
  • Criminal damage (such as destruction of household goods)
  • Rape
  • Preventing a person from having access to transport or from working.

This is not an exhaustive list.

Appendix B: Quotes about coercive control

From children

From (Callaghan et al, 2015)

  • ‘I think it was because my mum wanted to go out with her friends, and he didn’t want her to go out and all that ((.)) and started like throwing stuff and saying “You’re not going to go ((.)) and you need to help” and I dunno, ”help clean and make the food”.’
  • ‘Things would just get escalated ((.)) like if he knew what she was doing all the time, he could control like, everything, he would try to like, do stuff to scare us and I, I dunno, but I dunno what he would do, it’s just he wants to know like what’s going on so he just knows.’
  • ‘We didn’t actually know that he was a really bad person, so my mum ((.)) for some reason my mum got into this massive argument with him and then ((.)) he was, when I was there he started telling me that if I, if we went to court I was meant to tell the judge that mum, my mum was being a bad person.’
  • ‘If you touched the newspaper before he read it you were grounded.’
  • ‘Oh no, I’ve got to keep my mouth shut and I can’t say anything.’
  • ‘Cause he’ll be at the shop when mum wants to go in, and she wouldn’t wanna go in, so she’ll have to wait and get her bits ((.)) and then they’ll be snide comments.’
  • ‘You didn’t really wanna go outside ‘cause like, every time you did you were like, is that him? Is that him? And you just, even like now, when I go in the car park and it’s dark ‘cause I’m taking the rubbish out, it’s still like, is he still there? Or is someone there watching us or something?’
  • ‘Yeah, he spent three years, and then we went to this court thing and then, he got this thing to say that he can see us kids, but ((.)), he’s been messing my mum about, first he goes like “yeah it’ll be on a Thursday after school for a couple of hours”, so we could still go to our Nan’s for Sunday dinner, so now we hardly see my Nan, and then, like he’s changed it to wanting the whole of Sunday ((.)) ‘cause he was busy on a Saturday. Mum’s like “No”, but she had to do ‘cause he, he went to court again.’
  • ‘I’d sort of like sneak downstairs and check that no one was arguing or anything and if it was all OK, I’d come downstairs and sit down.’
  • ‘Sometimes he would like, ignore, ‘cause like I said he don’t like to take information, and then ((erm)) sometimes he would say “Oh come on, I’ll get you something” ((laughs)) and I was little so I was like “Okay”, ‘cause I can’t say no to when he says “Let’s get one of your favourite things” or something ((smiles)) I can’t say no to that so I come with him and then I have to talk to him so, yeah.’
  • ‘The first thing was he drove past … and then I looked on the road and I actually saw him, I was like “What?!” And then I kept walking and he was saying my name, and then he went down, then went to the zebra crossing, turned around and then it was alright ‘cause I knew some older people that were behind me, but anyway, he just like, he just like, put his like, two five pounds like that to me ((demonstrates how his dad held out money for him)) and then he didn’t say anything, and then I just walked on but took it and walked on, and then he just turned around and went back ((.)) he went.’
  • ‘I’d always hesitate of what I would say ... even if I said “Hello”, I’d always think before like, is he just going to shut me out? Is he going to respond in a nice way, or be angry or anything like that? I’d always think ahead of what I was saying.’
  • ‘When my mum gets money he takes it off her, so I say, I don’t say anything, she don’t get no money…. I mean like when she gets money out of the bank my dad takes it off her. So I have to lie to him.’
  • ‘I don’t know ((erm)) I can’t really explain it, I didn’t really have a feeling ((.)) like I knew he wanted like information for exchange, but ((.)) at the end of the day, I have the information, he doesn’t so I could technically control it so ((.)) it’s easier for me to just ((.)) get gifts ((laughs)) and it’s harder for him to get the information.’
  • ‘And then the next time he came … it was just, we were just walking and then we just saw him again, and then he was like “Do you want a lift?” and I ignored him and then he said “Do you want a lift?” I said “No” and then he said “Why?” and I said “I’ll call the Police because you know you’re not allowed to come near me” or something, and then ((.)) and then he said “but are you sure you don’t want a lift?”, I said “No I’m fine, I’ll call the Police”, then he went “Okay sorry” and then he went off.’
  • ‘I think the last year or so it’s made me think, “I’m not going to answer my phone if you’re going to ask about mum. I’m not going to answer my phone if you’re going to ask me questions. I will answer my phone if you say hi Jess how’s your day? And I will answer my phone if you’re going to give me money.’

From victims and survivors

From Coy et al (2008)

  • ‘I was walking on eggshells and could never predict if he would be a monster or be nice.’
  • ‘Although he was a doting husband in public, in private he would change.’
  • ‘He was a street angel and house devil.’
  • ‘I had a certain role to play – the typical role of a mother to do everything in the house.’
  • ‘Simple arguments would trigger it. When we were living together it was things like, you are not a proper woman you don’t clean the house up properly. You don’t cook enough.’
  • ‘I was cooking for six hours a day, he wanted fresh meals, he wanted different items. I was cooking to keep the peace. If I was to go out with my friends, he would make sure that I had his meal ready in the oven or all mixed together so all he had to do was put the oven on. It so scared me that sometimes I would forget to do that and I’d run back from the station just to do that so I wouldn’t get in trouble.’
  • ‘He was always trying to change me, saying ‘you will be a good woman’ ... he treated me like a servant.
  • ‘He kept me up until all night accusing me of having an affair, he had no idea who I was having an affair with (because I wasn’t). I explained that this was ridiculous, that there was no time, working and looking after the child. He became obsessed that I was having an affair ... he’d call my manager and question her on whether I was having an affair.’
  • ‘I have so much debt – loans, credit cards – all because of his financial abuse. He did not allow me to work.’
  • ‘He always had my bank card, and I wasn’t allowed to have any money. All the time I only had £2, and he used to count that money every morning and every evening. He used to put money onto my Oyster card, and after two or three days or after a week, he used to take the statements out of my Oyster card about where I was going.’
  • ‘My whole life is affected. My social life. My sexual life. I can’t even have a boyfriend. He isolates me away from people. I don’t go to work because he’ll turn up to harass me.’
  • ‘I didn’t feel safe. The intimidation was still there – he was able to drive up and smirk and intimidate me. He’d park a few cars up from the police station. Knowing he was there I didn’t feel safe and the fact that his mum was the mediator made it more difficult. I had to have a friend come with me.’
  • ‘He said ‘what I can’t f**king see my son? I’m going to f**king kill you, I’m going to shoot you, I’m going to stab you up’. He’s said that in the past and I thought that was heat of the moment, but I actually took this very seriously and I called the police. They just said not to let him in.’
  • ‘I kept trying different ways, but he told me a few times that he was going to take him away and never return him, so I stopped contact, and didn’t hear from him, then he went to a solicitor and applied for contact.’

Appendix C: Duluth Model Wheels

Power and Control Wheel
Post Separation Wheel


Appendix D: Coercive Control Change Checklist (Bancroft, 2002b)

(From: Bancroft (2002b) Why does he do that? Inside the minds of angry and controlling men. New York, NY: Berkley.)

There are two main principles to keep in mind when deciding how much potential an abuser has to become a kind, respectful partner in the long run:

  1. He cannot change unless he deals deeply with his entitled and superior attitudes. No superficial changes that he may make offer any real hope for the future.
  2. It makes no difference how nice he is being to you, since almost all abusers have nice periods.

What matters is how respectful and non-coercive he chooses to become.

Holding on to these fundamental points, you can use the following guide to help you identify changes that show promise of being genuine. We are looking for “yes” answers to these questions:

Has he learned to treat your opinions with respect, even when they differ strongly from his?

Yes No

Is he accepting your right to express anger to him, especially when it involves his history of mistreating you?

Yes No

Is he respecting your right to freedom and independence? Does that include stopping all his interference with your friendships and giving up the demand to always know where you are and whom you are with?

Yes No

Has he stopped making excuses for his treatment of you, including not using your behaviour as an excuse for his?

Yes No

Is he being respectful about sex, applying no pressure and engaging in no guilt trips?

Yes No

Has he stopped cheating or flirting with other women, or using other behaviours that keep you anxious that he will stray?

Yes No

Does he listen to your side in arguments without interrupting, and then make a serious effort to respond thoughtfully to your points, even if he doesn’t like them?

Yes No

Have you been free to raise your grievances, new or old, without retaliation from him?

Yes No

Has he stopped talking about his abuse as if it were an accident and begun to acknowledge that he used it to control you?

Yes No

Is he actually responding to your grievances and doing something about them (for example, changing the way he behaves toward your children)?

Yes No

Has he greatly reduced or stopped his use of controlling behaviours (such as sarcasm, rolling his eyes, loud disgusted sighs, talking over you, using the voice of ultimate authority, and other demonstrations of disrespect or superiority) during conversations and arguments?

Yes No

When he does slip back into controlling behaviour, does he take you seriously when you complain about it and keep working on improving?

Yes No

Is he being consistent and responsible in his behaviour, taking into account how his actions affect you without having to be constantly reminded?

Yes No

Is he behaving in a noticeably less demanding, selfish, and self-centred way?

Yes No

Is he being fair and responsible about money, including allowing you to keep your own assets in your own name?

Yes No

Has he stopped any behaviours that you find threatening or intimidating?

Yes No

Has he significantly expanded his contribution to household and child-rearing responsibilities and stopped taking your domestic work for granted or treating you like a servant?

Yes No

Has he begun supporting your strengths rather than striving to undermine them?

Yes No

Have you had any major angry arguments with him in which he has shown a new willingness to behave non-abusively?

Yes No

Appendix E: Universal Seven Step Response Framework (Breckenridge and Ralfs, 2006)

Professionals in frontline universal services need to be equipped with the knowledge and skills to identify and respond to children living with domestic violence in a respectful and appropriate way that considers the impact of coercive control. This framework is a useful process for professionals to follow to understand and consider the families ‘space for action’. The framework aims to acknowledge and strengthen the mother-child relationship.

  1. Developing awareness of domestic violence and its impact on children by obtaining knowledge from research on domestic violence, on the dynamics of abuse and on child development.
  2. Considering and noticing the needs of children – thinking about how the abuse is likely to affect the children, and hearing and understanding what children say.
  3. Attending to safety – this covers the child’s physical and psychological safety, and the safety of the mother and the professional.
  4. Deciding how to engage with children and/or caregivers – not ‘rushing in’, but considering what approach to engagement is appropriate, and the consequences of action or inaction, and being respectful of the child’s wishes and feelings.
  5. Responding specifically and appropriately to the individual child’s family context – to do this, professionals need to know what information and support is available for the individual child, who are the key people in other agencies and where are the gaps in services.
  6. Deciding on whether and how to follow up.
  7. Reflecting on practice through monitoring and looking for improvement.

Footnotes and References

See PDF download for full footnote information (if applicable) and references.


Digital download

Your price:


Please note: If applicable, VAT charges applied will be detailed on your invoice(s). This sale is not subject to VAT.

Digital download

Your price:


Please note: If applicable, VAT charges applied will be detailed on your invoice(s). This sale is not subject to VAT.

Professional Standards

PQS:KSS - Child and family assessment | Adult mental ill health, substance misuse, domestic abuse, physical ill health and disability | Developing excellent practitioners | Purposeful and effective social work | Developing excellent practitioners

PCF - Professionalism | Values and ethics | Rights, justice and economic wellbeing | Knowledge | Intervention and skills