THE DYNAMICS OF DOMESTIC ABUSE AND DRUG AND ALCOHOL DEPENDENCY

chester.ac.uk; Juliet Henderson, National Addiction Centre, Institute of Psychiatry, Psychology and Neuroscience, King’s College London, London, UK; Polly Radcliffe, National Addiction Centre, Institute of Psychiatry, Psychology and Neuroscience, King’s College, London, UK; Danielle Stephens-Lewis, Centre for Violence Prevention, University of Worcester, Worcester, UK; Amy Johnson, Centre for Violence Prevention, University of Worcester, Worcester, UK; Gail Gilchrist, National Addiction Centre, Institute of Psychiatry, Psychology and Neuroscience, King’s College London, London, UK.


Introduction
The 2019 Domestic Abuse Bill proposes to establish a statutory definition of domestic abuse that includes 'controlling, coercive, threatening behaviour, violence or abuse' encompassing 'psychological, physical, sexual, economic and emotional forms of abuse' (HM Government, 2019: 5). It proposes to widen the scope of Domestic Abuse Protection Orders so that suspected perpetrators of domestic abuse can be compelled to attend 'drug or alcohol treatment', as well as 'behavioural change' programmes by the family courts (if petitioned by victims or other relevant third parties, such as nongovernmental organizations) and magistrates courts (where the police would normally petition) (ibid. Explanatory Note Clause 3: 128). It is proposed that compliance with such orders will be secured in part through electronic monitoring. Breaches of such orders will be a criminal offence, punishable by up to 'five years' imprisonment, unlimited fine or both' (ibid. 30).
The Bill is informed by a prolonged consultation in which over 3,200 responses were received by government and expert opinion-primarily from organizations representing victims and survivors of domestic abuse and stalking-was submitted to two Home Affairs Committees (House of Commons, 2018). Cross-party support for the Bill was secured: as politicians registered the volume of domestic abuse cases raised with them by constituents; amidst news that the daughter of an MP had committed suicide following a relationship in which she suffered psychological-but not physical-torment that caused her to fear that she was mentally ill (Elgot, 2018); and during a campaign by David Challen to enable his mother to appeal her conviction for murdering his coercively controlling father (Moore, 2018). In strengthening the prohibition of 'coercive control' (Home Office, 2015: 2)-a concept advanced by Stark (2007: i) to explain 'how men entrap women in personal life' through 'intimate terrorism'-the Bill can be read as a logical extension of three decades of Conservative party policy that conceives the criminalization of a dangerous minority of men who abuse 'very vulnerable women and girls' to be a key part of the solution to domestic abuse (Heidensohn, 1995;Gadd, 2012). But this Bill was conceived within a more nuanced policy agenda than its predecessors. In the initial consultation document Transforming the Response to Domestic Abuse, which sought views on a raft of new measures, the then-Home Secretary, Amber Rudd, and Justice Secretary, David Gauke, called for policy that (1) recognizes that both 'women and men are victims of domestic abuse', though 'a disproportionate number of victims are women, especially in the most severe cases' (HM Government, 2018: 3); (2) 'actively empowers victims, communities and professionals to confront and challenge' domestic abuse; and (3) reduces regional variation in the quality of 'services to help victims' and 'punish and rehabilitate offenders' (ibid, our emphases). This receptivity to the rehabilitative ambitions of health and social care professionals derived principally from the findings of domestic homicide and serious case reviews (ibid, p21), which reveal the pertinence of a 'toxic trio' of domestic abuse, mental health issues and drug and alcohol problems in cases where women or children are killed (Brandon et al., 2010;Robinson et al., 2018), and how substance use features in around half of intimate partners homicides in the United Kingdom (Home Office, 2016). Transforming the Response to Domestic Abuse followed suit, highlighting the 'complex needs' of those living with 'drug and alcohol misuse, offending, mental illness and poverty' (HM Government, 2018: 10); domestic abuse 'victims' with 'problematic drug use' (p24); 'survivors… who have children on child protection plans' (p28); 'women at risk of having their children removed' (p28); 'female offenders' who have also 'experienced domestic abuse' (p31); and male 'perpetrators', who are too often depicted in terms of the 'stereotype' of a 'drunk… who… loses control and assaults their partner' (p11). Such 'simplistic' depictions were debunked for failing to 'reflect… the complex reality and lived experience of victims' and impervious to the 'dynamics of power and control which are present in many abusive relationships' (ibid. pp11 and 12). They had previously been challenged by official drugs policy that committed to supporting the disproportionate number of 'intimate partner violence' victims and perpetrators accessing substance misuse services (HM Government, 2017).
This article responds to this call to redress the dynamics of power that occur in relationships where substance use and domestic abuse co-occur. We contribute to such an understanding through the presentation of three couple dyads-each comprising a male perpetrator and his female partner-interviewed in-depth for the UK National Institute for Health Research funded Advancing theory and treatment approaches for males in substance use treatment who perpetrate intimate partner violence (ADVANCE) programme 1 . Our conclusion returns to the challenges the 2019 Domestic Abuse Bill poses to policy, practice and criminological theorizing.

Correlations and typologies
Evidence for the relationship between domestic abuse and drug and alcohol intoxication is plentiful in crime surveys but tends to focus, peculiarly, on the behaviour of victims more often than offenders. For example, the 2016 Crime Survey for England and Wales revealed that 'adults aged between 16 and 59 who had taken illicit drugs in the last year' were three times more likely to report 'being a victim of partner abuse' than those who had not done so (ONS, 2016: 25). However, using illicit drugs does not invite assault and the identification of such 'risk factors' in the absence of explanation of their relevance accentuates the victim-blaming some perpetrators deploy to control their victims (Gadd et al., 2014). The international evidence reveals that men, but not women, tend to perpetrate more severe assaults when they have been drinking (Graham et al., 2011;Reno et al., 2010). Women are more vulnerable to assault when they too are intoxicated, but this is at least partly because those living with abusers are less diligent at pursuing safety strategies when they have been drinking (Iverson et al., 2013). Substance use features in around half of all UK domestic homicides. Since 2011, substance use has been detected among domestic homicide perpetrators more than four times as often as it has among those killed by them (Home Office, 2016) In sum, the relationship between substance use and domestic abuse is not straightforward. Moreover, Different substances have different pharmacological properties. They are used in variable quantities and combinations fostering a range of effects-including docility as much as aggression-that are contingent upon the user's experience of them, prehistory of use, mood and the context in which the consumption takes place (Zinberg, 1984;Gilchrist et al., 2019). Laboratory research reveals that those with low levels of inhibition, empathy and self-regulation and elevated levels of sensitivity to threats and insults ('instigative cues') are more prone to violence when they have consumed alcohol up to four hours ahead of a perceived threat or 'provocation' (Leonard and Quiqley, 2017). Cocaine consumption can induce similar reactions. Like cannabinoids and opiates-the effects of which are rarely studied in the context of aggression or violence-cocaine can also alleviate anxiety and exacerbate underlying problems with depression, paranoia and hallucinations (Sacks et al., 2009). Consequently, regular use of such drugs, like the consumption of excessive alcohol, can impinge upon mental well-being and intimacy, generating indirect and belated relationships between victimization and substance use that extend far beyond periods of intoxication.
Feminist scholarship on domestic abuse has tended not to engage with the pharmacological impacts of substance use and has focussed instead on how some abusive men retain power over women by attributing their violence to intoxication, by insisting that their drinking caused them to act out of character, or by denying any memory of assaults perpetrated when intoxicated (Hearn, 1998;MacKay, 1996) interventions for perpetrators have thus needed to be alert to the ways in which substance use is invoked to minimize violence. Women's accounts of victimization have had to take precedence over men's self-reported offending as measure of changes given the potential for such minimization (Dobash et al., 1999). But as Stark's (2007) review of officially reported 'intimate terrorism' cases illustrates, substance can also be implicated in the perpetration of 'coercive control' and victims' responses to it. His analyses reveal that some victims do self-medicate to manage the depression the daily anticipation of violence engenders and that some perpetrators control victims by increasing their dependence on substances before restricting their access to them. Finally, Stark highlights that some women who have been terrorized over many years take matters into their own hands after the law has failed to protect them, mounting grievous attacks on perpetrators when they are too intoxicated to retaliate.
Typological research on men's domestic abuse perpetration has also addressed the role of drugs and alcohol anecdotally. For example, Holtzworth-Munroe and Stuart (1994) suggest that there are three distinct types of male domestic abuse perpetrator, one of which-the 'antisocial batterer'-is defined by their dependence on drugs and alcohol, engagement in crime and paternalistic values. The other two groups, they propose, include 'family only batterers' who are seemingly 'normal' men who are violent at home and conventional in their sexism; and 'emotionally dysphoric batterers' with clinically diagnosable 'borderline personality disorders' who tend to be overtly misogynistic, especially when their relationships are failing, or they distrust the fidelity of their partners. Yet, what the relationship is between substance use and violence for the antisocial batterer remains untheorized in Holzworth and colleagues' tests of their typology (Holtzworth-Munroe et al., 2000). This is despite clinical evidence suggesting that drug use and violence co-occur most among men with diagnosed mental health issues, poor concentration and problems understanding and remembering their pasts (Sacks et al., 2009). In relation to 'family only' perpetrators, Johnson et al. (2014: 65) suggest that this group is more likely to be involved in 'situational couple' and 'separation-instigated' violence that is more 'gender-symmetrical', and derivative of arguments over domestic matters, finances, childcare or 'objections to the other partner's excessive drinking' that evolve into 'fights'. For this subgroup of 'family only perpetrators', the link between alcohol consumption and domestic abuse may have more to do with everyday conflicts than personality traits, though the difference between them and intimate terrorists can be overdrawn (Gadd and Corr, 2017). Sociologically speaking, control 'is a continuum. Everyone controls their partner to some extent' (Johnson, 2008: 87), begging the question as to when and why the desire to control becomes pertinent.

Complex interdependencies
Answers to this question can be found in the few qualitative studies that explore how drugs and alcohol feature in the relationships of couples living with domestic abuse. These reveal that some perpetrators pose greater risks to their partners, not when they are high, but when they are irritable, withdrawing or are struggling to finance alcohol or drug purchases (Gilchrist et al., 2019). One exemplar is Hydén's (1994) study of middle-class Swedish couples reported to the police for domestic abuse. Follow-up interviews with 20 couples where alcohol consumption was noted by the police revealed that, although drunkenness and its expense were the source of many arguments that led to violence, social drinking, especially at parties, was also what held some relationships together. Afterwards, some couples reconciled on the basis that it was the alcohol that caused the conflict. They asserted that the perpetrator was normally a 'good person' who could be helped. Men who had caused injuries when intoxicated often claimed they could only recall feeling hurt-sometimes in ways that reminded them of painful experiences in their pasts-by female partners who criticized them or acted aggressively towards them and not the assaults they themselves had perpetrated.
Evidence of the relevance of emotional pain can also be found in Motz's (2014) case analyses of couples in therapy. This reveals how some women who had been abused or neglected as children attempted to cope with feelings of vulnerability 'through the creation of highly dependent relationships with men who… offer… protection, and through getting into states of mind where these feelings can be pushed away… through drugs or alcohol' (p69). Motz depicts the emotionally impoverished lives of abusive men with whom some drug-using women cohabit, many of whom feared abandonment because of experiences of abuse, neglect or institutional care. Some of these men had 'little capacity to tolerate emotional intimacy' (p93) and thus found it 'impossible to relate' to their families or sexual partners unless 'high on drugs' (p93). Over time, Motz suggests, these couples became 'doubly dangerous', leaving their children uncared for when intoxicated, withdrawing or fighting, and unable to 'come together safely' in an emotionally connected way to 'manage and contain distress' afterwards (p158). 'Toxic couples', Motz argues, deny their own dependencies and instead project them onto each other, leading them to view their partners as more out of control than they are. For some men such projection amounted to 'a fantasized attempt at creating a state of invulnerability and absolute control' (ibid) when their own lives are in disarray.
Evidence of this kind of 'splitting'-where good and bad, safe and dangerous, vulnerable and invulnerable, qualities in the self or other are imagined as irreconcilably polarized-upon which such projective processes rely, can also be found in Gilbert and colleagues' (2001) focus-group study of women enrolled in North American methadone programmes. Participants described how altercations materialized rapidly when high on crack cocaine or when drunk, as intoxication induced paranoid sexual jealousy that led to hostile accusations by men who became like 'Jekyll and Hyde'. When withdrawing from heroin, some men attacked their partners for failing to provide money for drugs, some women cited 'irritability' as explanations for their own use of violence towards their partners when intoxicated or withdrawing, meanwhile others emphasized that drunkenness intensified their male partners' criticism of them for failing to fulfill household tasks. Some women described engaging in prostitution to raise money for drugs as evidence of their love and care for male partners. When the women subsequently refused to raise funds in this way or sought support from professionals to reduce their own drug use, some male perpetrators threatened further violence whereas others encouraged them to relapse back onto heroin or crack, thus entrapping stigmatized and socially isolated women in relationships with them.

Method
In what follows, we expand the argument for a more relationally sensitive analysis of the dynamics of power that pertain in the lives of couples where domestic abuse towards a partner occurs alongside substance use. Such analyses, we argue, need to be attuned to the gendered power dynamics of drug use and domestic abuse: dynamics that may be reciprocal even while unequal; financial, emotional and pharmacological; involve violence that is perceived as 'situational' by one partner and 'coercively controlling' by the other; and recalled as involving movements between intimacy and distance among the exchange of insults and assaults, craving for drink and drugs, intoxication and withdrawal. We seek to illustrate these points by drawing on dyad interviews-with male perpetrators in treatment for substance use problems together with their current and former female partners-undertaken for the ADVANCE programme.
The ADVANCE programme seeks to develop and test an integrated intimate partner violence and substance use group intervention that will reduce intimate partner abuse perpetrated by men receiving substance misuse treatment. We report here on the programme's preparatory workstream. This involved interviewing male domestic abuse perpetrators receiving treatment for substance use and their current or former partners about their relationships and support needs. Adult men were recruited from six community-based substance use treatment services in London and the West Midlands. The treatment services were for people who regarded themselves as 'substance dependent', typically because they regarded their drug and alcohol usage as 'compulsive', necessary to deal with problems, taking up a lot of time and energy, costing more than they could legitimately afford, and/or very difficult to stop 2 .
Seventy men were screened for lifetime domestic abuse against a partner. Men who currently had court orders preventing contact with their (ex)partners were excluded. Forty-seven of the 70 men screened were eligible, and 37 of these 47 men were then interviewed. Male interviewees were asked to provide contact details of their current or former female partners, and in 14 cases these women were interviewed. All participants were advised that there were limits to the confidentiality that could be afforded where unaddressed risks of harm and safeguarding issues were disclosed. Women and men were always interviewed by different researchers to ensure no information was inadvertently shared between participants. Participants were paid £20 to compensate for their time.
Interviews were undertaken using reflective techniques derived from the Free Association Narrative Interview Method (Hollway and Jefferson, 2000), with participants being supported through active listening to tell the stories of their drug use, relationships, domestic abuse and intervention experiences. Digital recordings of the interviews were transcribed verbatim and transcriptions were checked twice for errors. Timelines were created to track the sequence of events through the life of each participant. Case studies were then written-up as 'pen portraits', which sought to capture the complexity revealed in each interview, including apparent contradictions, avoidances and implausible claims. In the 14 cases where both partners were interviewed, men's and women's accounts were compared with each other. Although all of the perpetrators interviewed could have been coded as 'antisocial' in Holtzworth- Munroe's (2000) terms, given their drug use and criminal histories, such categorization would oversimplify matters. All but two of the men depicted their violence as situational and/or a product of some form of mutual combat, whereas all but one of their partners depicted coercively controlling abuse, to which around half the women responded with some degree of violent resistance. In terms of their drug use, the 14 men who were interviewed with their partners appeared to be broadly comparable with the other 23 whose partners were not interviewed (Table 1). The majority used heroin with other illicit substances, notably crack cocaine and/or powder cocaine, though some also mixed benzodiazepines with alcohol. Nine out of the 14 were also heavy drinkers. Five of the 14 men also described medical or psychological diagnosis consistent with emotional dysphoria. Eight males disclosed perpetrating violence that was extra-familial in addition to their abuse of partners. Contact with children had, at some point or other, been restricted for all the men in the study.
Given the high degree of similarity among the men on key variables, we explored if more meaningful distinctions could be drawn by distinguishing the dyads in terms of whether victims had ever used drugs and, if they had, whether they were desisting from substance use or still using. Only four of the women described themselves as substance dependent at the time of the interviews. Five had never been substance dependent, and another five were desisting from substance use, either having become completely abstinent from using or having only had temporary relapses. A three-fold distinction could thus be drawn across the dyads that revealed some important variations in terms of how domestic abuse and substance use manifested themselves.

Group 1. Victim had never been substance dependent (n = 5)
Within the sample, there were five couples where the female partner had never been substance dependent, though all the women interviewed drank alcohol socially, and one smoked cannabis occasionally. Women in this group had almost no involvement in crime. Four of these women had never been separated from their children, but one woman had children who had been required to live with their grandfather as she would not leave her abusive partner. These non-substance dependent women were typically confused as to why relationships that had started out well had suddenly deteriorated; why their partners engaged in unexplained and peculiar behaviours; and why they had accused them of unfounded infidelities while lying about their own substance use and/ or the criminal activity that generally supported it. Substance use among the 14 male perpetrators whose partners were also interviewed 13 11 13 9 Substance use among the 14 female partners 10 8 10 6 Group 2. Victim was desisting from substance use (n = 5) Within the sample, there were five couples where the female partner had abstained from using drugs or alcohol, having previously been substance dependent. None of these women had criminal convictions. The stories these desisting women told tended to be of intimacy lost. Sharing feelings and traumas that motivated drug use, and about what made it difficult to give up, had generated understanding and closeness when they had first met their partners. Conflicts had then developed when the men resumed drug use or drinking whereas the women were trying to reduce their own or abstain. Only two of the women in this group had children of their own. In both cases, these women had raised their own children, but with some intermittent professional oversight.

Group 3. Victim was substance dependent (n = 4)
Within the sample, there were four couples where both the male perpetrator and the female victim were both currently substance dependent. All the women in this group used crack cocaine and heroin to varying degrees. Though they sometimes mentioned love, they often explained their persistence with relationships that had become abusive in terms of daily needs for protection, somewhere to live and the sharing of drugs. The women in this group had much more frequent and entrenched patterns of criminal involvement than the other 10 in the sample. Their criminal involvement activities included shoplifting, petty frauds and prostitution to finance their drug use, typically with encouragement from male partners who relied, to some extent, upon the income the women generated. All four women in this group had been separated from their children when these children were young, though two women had re-established relationships with their children in adulthood.
In what follows, we present one couple from each of these groups to further illustrate the different power dynamics that can pertain in relationships where domestic abuse and substance use co-occur. Italics are used to highlight points where the participants emphasized a relationship between substance use and domestic abuse.

Group 1 exemplar. Victim was never substance dependent
Wayne (early thirties) and Rhian (late twenties) met while she was managing a pub where he drank, sometimes 'heavily', during the 'daytime'. She had never cohabited with a partner before, but he already had a child with another woman and had served at least six prison sentences, two for 'kidnap' of his own child. By Rhian's account, when they first met three years before, the relationship 'progressed really quickly': 'within a couple of weeks' Wayne was staying in her flat. By Wayne's account, he and Rhian visited the grave of his grandfather before spending the night together. After that, Wayne said, he 'couldn't get rid' of Rhian: he 'loved her to bits' and their relationship was 'proper good' for 18 months until he began to 'blag' money from her to buy heroin. Rhian's account, by contrast, was recollected more as an unfolding nightmare, in which she did not know why Wayne was being so controlling until after their baby was born when he revealed he was getting treatment for heroin dependency.
Rhian recalled that Wayne first assaulted her within a couple of months of moving in. After a drink with friends and not knowing that she was already pregnant, Rhian felt ill and went to bed. When Wayne returned home, he became 'very argumentative', 'coming right' in her 'face', accusing her, without foundation, of sleeping with someone else. Rhian wanted to end the relationship then, but Wayne was profusely apologetic, convincing her to keep the baby and get their 'own place', explaining his life was 'empty' before they met. Wayne provided a detailed account of the emptiness he felt. His mother, who was separated from his father, had worked nights in a pub, leaving her children 'home alone' to run 'wild'. In her absence, Wayne began smoking 'weed' regularly. He had 'weird thoughts' and would pick fights with anybody, feeling no 'remorse' afterwards. Wayne's mother sought help but did not receive any. Instead, Wayne became abusive to her, 'calling her a slag', threatening to hit her when she took his brother's side, and constantly 'smashing her house up'. Invited by his cousin to 'smoke' heroin to 'forget' his grief' after his grandmother's funeral, Wayne claimed his clandestine usage escalated from there, Rhian erroneously assuming he was cheating on her, doing nothing to 'help' him come to terms with his loss, and causing him to cheat on her. As a 'druggie', Wayne said, he became unable to show Rhian 'affection' or 'love anybody', including himself.
Wayne's accusations of infidelity caused Rhian much distress while his behaviour became more erratic and threatening, 'his jaw… going and his eyes' being 'wired'. He smashed Rhian's phone because he was 'convinced' that he had 'seen someone's name' on the screen. Though their relationship was 'over' during most of her pregnancy, Rhian 'literally couldn't go anywhere' without Wayne constantly 'phoning' and 'texting' her, questioning her about what money she had spent, and sometimes barricading her in the house until she asked his mother to come and get him. After a nurse overheard Wayne discussing drugs on his phone during one of the few antenatal appointments he attended, Rhian took the opportunity to ask him what it was about because she was concerned that social services would see a child protection risk for her baby. Wayne responded by calling Rhian a 'sick' 'liar' and insisted that it was she, not drugs, that was 'driving' him 'crazy'.
Wayne apologized after their son was born and claimed that holding the baby inspired him to 'change'. He and Rhian then took, what she recalled was, a 'perfect' family holiday together in which he explained that he had been prescribed Subutex (buprenorphine), a semi-synthetic opioid used to treat heroin dependence. Wanting her son to be raised by two parents, Rhian agreed to let Wayne move back in, but when his drug use resumed, he became 'physically aggressive again'. Rhian recounted three occasions when Wayne throttled her, once asking her to put the baby in another room so the child would not see, and on another occasion putting a knife to Rhian's throat and rationalizing, 'You're killing me… So, I should … make it look like you killed yourself'. Sometimes Wayne would speed off, with their baby in the back of the car, in a hurry to buy drugs. Other times, he locked Rhian in the flat for days because he suspected she was 'cheating' and feared she would leave him. The violence only ceased, Rhian said, when Wayne called the police on himself after pinning her down and grabbing her by the neck. Rhian said she was pressurized by the police to make a statement against Wayne, but that the case was withdrawn when he made a fraudulent counter-accusation and a friend of his posted content on Rhian's Facebook page, as if by her, purportedly confessing. Wayne, by contrast, only recalled one assault explaining they 'didn't argue a lot' partly because his 'mind wasn't there'. He admitted driving off with the baby because he was in a 'rush' to get drugs but insisted the argument occurred only on his return when Rhian, who thought 'she was more powerful than everybody else', punched him 'in the chest' to stop him leaving again. In response, Wayne claimed, he had 'moved' Rhian by the 'face' because she 'wouldn't let' him leave and was 'kicking' his 'legs' and because he knew he 'would have ended up battering' her as he did not always know what he was doing when 'on heroin'. Wayne was thus surprised to later be awoken by 'two police officers' who arrested him, but relieved when the courts concluded that Rhian had lied, and pleased that, post the break-up, he had been able to access some support for the mental health problems that had been troubling him since his childhood.

Group 2 exemplar. Victim was desisting from substance use
Mitchel (early fifties) and June (mid-forties) were in a relationship for over 15 years. They met while in residential rehabilitation for heroin dependence when they both were 'emotionally raw'. Mitchel felt the 'deepest love' for June when they met. June thought she had 'met her soulmate': an 'affectionate' man with whom she had much 'empathy' given his 'horrendous' experiences of 'child abuse'; a man who helped her overcome the death of her son's (also heroin-using) father. As a teenager, Mitchel too had found 'comfort' in heroin use after he was sexually abused by his brother and his brother's friends while their mother, was out 'trying to find somebody to love her'. As a child, June was repeatedly coerced into having sex by a man who threatened to report her to social services for caring for her siblings while her mother received hospital treatment. June said that as a university student, she lacked the social skills to say 'no' when she was introduced to heroin. When she became pregnant, she weaned herself off it but relapsed when her mother accused her of inviting the sexual abuse she was subjected to as a child. Having taken heroin to 'bury' the 'hurt' this accusation inflicted, June self-referred into drug rehabilitation for two years so she could raise her son in an environment in which her desistance from drugs was effectively managed. From Mitchel's perspective, problems in their relationship emerged after they left the residential rehabilitation. June, he said, was 'damaged goods': 'although the love was there' she was 'frigid'. He pursued sex with a woman called Rose and hoped that 'the three of' them 'could love each other', introducing both women to crack cocaine to facilitate this. But, according to Mitchel, the 'resentments grew', until June became pregnant and began 'hounding' him to commit to her, even though she knew he 'loved' Rose 'more', trapping him, paradoxically, in a sexless relationship by becoming pregnant. Owing money to a crack cocaine dealer, Mitchel said, he and June fled to his mother's house for a 'fresh start', during which they could cease using drugs and get their own place.
June, by contrast, made no mention of a polyamorous relationship and said that she had remained 'clean' of drugs for a decade after leaving the rehabilitation centre while Mitchel's drug use resumed. Thereafter, she said, Mitchel would constantly 'put' her 'down and compare' her to another woman. She said that while the heroin would 'subdue' Mitchel, when drunk he became 'aggressive and arrogant, looking for a problem'. This abusiveness heightened during her pregnancy; a time when she felt increasingly 'isolated' and 'insecure'. The 'fresh start' she had been promised never materialized though this was partly because she started drinking heavily, blurring the line between his 'put downs' and her responses to them. Drinking, June said, helped her tolerate Mitchel's 'screaming', but sometimes he was determined to escalate arguments, once pouring a bucket of water over her while in bed. When Mitchell returned from university, where his relationship with Rose had resumed, June said he 'was drinking pints of vodka' as well as using heroin while undertaking odd jobs for cash, and that she would come home from work to find him 'asleep' in front of 'a plate full of heroin and needles' while the children played unsupervised. June said that when she 'confronted' Mitchel, he 'cleared' out her bank account, leaving her reliant on money borrowed from her mother to provide food for the family. Feeling 'depressed', 'trapped' and defeated', June began using heroin again.
Mitchel made no mention of these incidents but said June had become domineering about domestic matters when he returned from university. ' Violence' became their 'means' of 'communication' at this time with him threatening to hit her 'back' when she 'lashed out'. June explained that she had once hit Mitchel in retaliation, whacking him 'with a folder' when he lent her car to an unqualified driver and the police questioned her about it. Mitchel responded, she said, by 'kicking' her 'from the head up', breaking her jaw, causing her unforgettable pain. June said she lay on top of her son to protect him when Mitchel went 'ballistic' because the boy had failed to clear up the kitchen after making his own lunch. June conceded that she 'hit' Mitchel 'right back'. Mitchel said he regretted hitting June 'like a man', clarifying that normally he would ' just' hit her back with an 'open hand slap', but that on this occasion she 'came at' him, creating an 'explosion' before having a 'breakdown', perplexingly 'terrified' of him.
The police attended but arrested neither of them as they had both been drinking. So, June said, she tried to leave for a friend's house with the children and eyes that were too swollen to open, but Mitchel kicked and beat her again. After a period in hospital, June said June contacted a drugs and alcohol dependency team who put her on a methadone programme, but Mitchel started taking the methadone because he feared he would lose the house and his children if June recovered. June's version was that she only succeeded in leaving Mitchel after she awoke to find him 'forcing' tablets down her 'throat', to make it look as if she had killed herself by overdose. Mitchel made no mention of this attempted murder but explained how bitter he was that June secured a court order that prohibited him from seeing the children merely because he had made the 'mistake' of buying a very large 'bag' of heroin and despite always having done the hoovering and cooking 'for them'.

Group 3 exemplar. Victim was substance dependent
Joe (mid-thirties) and Kate (late twenties) had been together for six years. A week after having met in the streets and gone out for a drink, Kate arrived at Joe's house with just a suitcase and never left. Kate had been sexually assaulted both as a child and as a teenager and was estranged from her family. Joe, whose parents were both deceased, was sexually abused while in care and was estranged from his siblings. Kate's children lived exclusively with her previous partner, their father, because of Kate's alcoholism. Joe had been a heavy drinker since his molestation and had served multiple prison sentences: two for attacking men he had seen 'touch up' women without their consent and one for assaulting Kate. All but one of Joe's many previous relationships had involved violence, some grievous and directed at him, but for which he had often been arrested, leading him to the conclusion that 'it is really sexist out there': 'there's one rule for blokes and one rule for women'.
Despite being 'frightened of men', Kate initially found Joe 'really nice'. She said he 'spoilt' her and did 'sweet' things, taking her to restaurants and bringing her flowers. They both emphasized that they had loved each other, though Kate said she struggled to 'handle' Joe's attention and was sometimes 'mouthy ' and 'hateful' towards him when drunk, merely to elicit a different 'reaction'. From Joe's perspective, however, 'every argument' was 'about drugs and money'. He understood that Kate was using drugs-something she barely disclosed in her interview-'to block out the pain' of her past, but the drug use had affected their sex life, while chronic pancreatitis had left her with 'only… a few years left to live'. Joe did not like Kate 'clipping'-'robbing' men she deceived into believing they would have sex with her-to fund their drug use and wanted her to steal from supermarkets instead. He said he worried that Kate would be raped or killed by men she had clipped and that he had lost teeth defending her from men she had tricked. Joe admitted being ' jealous' and afraid that Kate was 'cheating on' on him, though he knew she was not 'a slag' despite 'acting' like 'one'. Joe considered himself to be no longer 'alcohol reliant', having given up spirits, but claimed that he became 'addicted' to heroin a year ago, trying it to show Kate he could 'understand' what it was like for her. Heroin withdrawal had been the real 'devil' for Joe, leaving him unable to 'walk' at times, 'depressed' and vulnerable to a descending 'red mist' that he claimed rendered his temper uncontrollable. Joe commenced a Subutex prescription during his most recent prison sentence which, since his release, he had shared with Kate to 'make sure that she ain't sick' (i.e., suffering withdrawal symptoms), sometimes also using heroin or crack cocaine in addition to his prescription.
From Kate's perspective, however, Joe's protectiveness could be 'suffocating'. She explained that although Joe initially 'understood' how her childhood experiences of the sexual violence affected her, his capacity for understanding was now contingent on whether she had sex with him. He now treated her like a 'child' and 'as his' property and feared he 'could kill' her in an 'accidental angry' moment. When coming down from being high or drunk, Joe was often 'controlling' and could 'switch very easily with anyone'. Kate explained that previously, when Joe had been smoking crack, he assaulted a 'pervert' who had touched Kate 'in an inappropriate way'. After he had finished assaulting the 'pervert' Joe proceeded to strangle and batter Kate, breaking some of her ribs. Hence, Kate avoided doing anything 'sudden' that would make Joe 'paranoid', despite having invited him to ' just fucking kill' her rather than keep 'terrifying' her. After 'days' of 'not sleeping and just drinking', Joe tried to provoke an argument. When Kate walked away, he mimed 'putting bullets' in her head, so she 'pushed him away' and he 'punched' her. While Joe was in prison for this assault, Kate twice attempted suicide. She continued to blame herself for his violence and drank alcohol 'to feel happy' while questioning whether it was 'really' her 'fault' that Joe was so 'messed up', as he has claimed. Joe, by contrast, claimed Kate had hit him 'over the head with a hammer' because he 'wouldn't buy her drugs' and explained that the assault on her, for which he went to prison, occurred after she 'slapped' him 'round the head' because he did not 'have… money for drugs'. It was unfortunate, he said, that the police drove past just as he was hitting 'her back' in 'self-defence'. Though Kate said she 'loves' Joe 'to death' she doubted whether the 'damage' to their relationship could be 'mended'. He, by contrast, was desperate 'to get her clean', as he imagined this would enable him to get his own 'life back'. Joe assumed that if Kate became sober enough to see her children again, it would save his relationship with her from 'ruin'.

Analysis
In this article, we have presented three relationship scenarios where domestic abuse pertained alongside drug or alcohol dependency. These relationships diverged primarily in terms of the female partners' histories of drug and alcohol consumption as all men were in treatment for substance dependence. All three men-like the majority of those interviewed in the ADVANCE project-considered 'drugs', their own and/or their partner's use of them to have damaged or ruined their relationships. Their depictions of violence as 'isolated incidents' in which they were only partially culpable were consistent with perpetrators' accounts more generally (Stark, 2007;Women's Aid, 2018;Gilchrist et al., 2019). Wayne, Mitchel and Joe all described discrete, regrettable and unplanned assaults that derived from everyday conflicts over alcohol and drug use, financial pressures, sexual jealousies and domestic chores: conflicts that were sometimes accentuated by being intoxicated. Nevertheless, the stories these men told suggested that their need to control became increasingly acute when their relationships were in crisis, when they had secrets to keep, when they felt dependent on drugs or alcohol, were afraid of losing their minds, their partners and their children, when money was scarce, and when homelessness and criminalization were distinct possibilities. As these men projected this sense of being in disarray onto their partners, the women began to feel like they were being driven crazy, in part because they did not have full knowledge of the drug and alcohol use that was consuming the men's time and minds. As the women began to question what was happening, the men's attempts to coercively control became more dangerous and desperate, e.g., in the refusal to let partners leave their homes or in their efforts to tempt or coerce the women into consuming drugs. Despite their unhappiness, these men, like their partners, often lacked the emotional strength and economic resources required to separate (Walby and Towers, 2018). Instead, the men often blamed discrete incidents of violence, as they construed them, on drugs and/or money-related issues that could be fixed if they entered treatment and their partners were prepared to fight for the relationship, for the sake of children whose well-being had not been paramount (to the men) previously.
By contrast, Rhian, June and Kate, described steadily accumulating patterns of abuse, forgiven initially as promises of fresh starts, either in new places or after drug treatment, were made. The women's reasons for enduring domestic abuse or for giving the men another chance began with this hope for change but often mutated as they encountered the financial and emotional difficulties of leaving homes, the prospect of losing their children (forever in Kate's case) and the concomitant risk of criminalization when the men threatened to report them for hitting back or for using drugs. Hence, the reasons these women stayed were complexly configured around drug and alcohol use. Wayne's abusive behaviour had proved confusing to Rhian, who knew only that he was a heavy drinker until his heroin use was confirmed after their baby was born. Then, as someone with little experience of either drugs or relationships, Rhian was persuaded to give Wayne another chance while he sought drug treatment, assuming mistakenly that this would redress his violence. June, by contrast, had some empathy with Mitchel, having relapsed with heroin herself and recognizing that her own drinking contributed to their arguments. June had been persuaded that moving might facilitate a fresh start, without drug use. However, when June sought opioid substitution treatment for herself, Mitchel found a new way of controlling her, diminishing her capacity to leave by controlling her access to her prescription and then trying to administer an overdose. The challenges for Kate were different again. She had a long history of heavy alcohol consumption and illicit drug use, the latter of which Joe had joined in with, compounding their mutual dependence on shoplifting and pseudo-sex work to maintain their supplies. Joe construed his heroin use as an attempt to empathize with Kate, though it appeared that he persisted with drug treatment partly because it legitimized his management of her drug use. Joe hoped he would get his 'life back' if he could facilitate a reconciliation between Kate and her children. In the interim, Kate suffered grievous violence, while living in Joe's home: violence that was construed as part of the protection he afforded her against men she had clipped.
For the women in these relationships, criminal justice intervention was often greeted with trepidation, for it rarely provided the protection it promised. Instead, they had often concluded that it was simpler to suffer difficulties within their relationships, attribute violence to drugs use and attribute drug use to earlier traumas, of which there were many in our participants' lives. For June and Kate, the pains of child abuse, mental health problems and bereavement were partly responsible for the solace they had sought in alcohol and heroin consumption, as well as in their relationships with men. However, as their drug and alcohol usage became complicated by domestic abuse, a range of different strategies were pursued by each couple, typically to avoid attracting the attention of social services or the police. These strategies included taking prescribed medications to minimize their need to commit crime to fund illicit drug use (Joe), moving away while also severing ties with friends and family (Mitchel, June), switching substances (Mitchel, Joe), pursuing relationships with others who use illicit drugs to avoid feeling 'trapped' (Mitchel, Kate), consuming drugs or alcohol to cope with the aftermath of conflict (Wayne, Mitchel and June), engaging in crime together (Joe, Kate) and tacitly encouraging partners to participate in drug use (Mitchel, Joe), compounding the risks faced by women who wished to abstain or keep their use moderate. Although drug and alcohol use could increase sociability and enhance feelings of closeness between partners, the fear of dependency also induced feelings of worthlessness-evidenced most vociferously in Wayne's belief that he could not love anyone and the paranoid accusations this engendered, but also hinted at in Joe's jealousy and Mitchel's infidelity.

Discussion and conclusion
These cases reveal how the projective dynamics that impart blame, often through men's claims that their female partners are 'driving' them 'mad', are easily facilitated by the nuances of sexism and reinforced by the perennial threat of violence. These dynamics were compounded as drinking and drug use generated financial pressures, which intensified conflicts that left the women, as well as some of the men, feeling that their partners regarded sustaining their substance use as more important than their relationship, avoiding criminalization and social services intervention, and the threats posed by those from whom money and drugs had been borrowed or defrauded.
For time-pressured police officers, social workers and magistrates faced with partial evidence and counterclaims, discerning the 'truth' of who had done what to whom in which circumstances would have been particularly difficult. Evidently, some abusive men tell highly convoluted stories to exonerate themselves. But some women who are the primary victims in such relationships do not and cannot always tell the whole truth either, not only because they fear further violence and abuse but also because of the stigma of their own drinking and drug use, the fear of child protection proceedings being instigated and the risk of being incriminated by perpetrators they have hit in selfdefence or retaliated against (Wolf et al., 2003;Felson and Paul-Phillipe, 2005). What is under-acknowledged in many serious cases of domestic abuse is that both perpetrators and victims often share in the shame associated with being abused as adults and children, of failing to protect their own children, anticipate their partner's needs, having hit back, gotten drunk or engaged in illicit drug use.
Like many of the men in the ADVANCE programme study, the perpetrators we have depicted here dealt with feelings of trauma and grief from their pasts through drug use and by scaring their partners in ways that the women experienced as acutely controlling. While frequently terrifying, such behaviour was not only instrumental and controlling but also expressive of how painful some aspects of their pasts were and how unwilling they were to concede their dependency on both substances and partners who provided care, funds, a place to live and the support needed to maintain precarious relationships with children. Similar experiences of child abuse, mental health problems and drug dependency were sometimes part of the story of intimacy that held these couples together despite grievous domestic abuse. Then, when the risk of criminalization or estrangement presented, men who were coercively controlling sometimes used such prehistories against their partners by threatening to expose them for raising children in contexts that were unsafe. Hence, the 'madness' that the women in these relationships often felt was not simply symptomatic of their own mental health problems but projected onto them by men who had become desperate to impose their own versions of reality.
This imposition of the perpetrator's reality sometimes became more forceful when the criminal justice system intervened. The risk of 'legal systems abuse' occurs when perpetrators adept at coercive control harness the powers of the police or the courts to further intimidate their partners (Douglas, 2018). It has, to some extent, been be amplified by the advent of gender-neutral policy, which recognizes that men can be victims too, alongside incident-focussed approaches to policing that direct attention to what has just happened-such as a man being hit-rather than the history of the relationship-such as a woman being terrified or controlled by the same man over a prolonged period (Walklate et al., 2018). The 2019 Domestic Abuse Bill attempts to counter this risk by prohibiting perpetrators from cross-examining victims in the family courts and providing greater recognition of the impact of the ways in which economic abuse makes it harder for many victims to leave. But compelling alcohol and drug-using perpetrators to receive treatment may introduce unforeseen possibilities for coercive use of the law. Some women will consider themselves too culpable to seek support and will ultimately be let down within a criminal justice process calibrated to identify the perpetrator of assaults at the scene and/or whether they were intoxicated, and hence be easily blindsided by the mutualizing discourses some serial offenders offer in their defence (Tolmie, 2018). Others will stay under the misapprehension that the domestic abuse will cease once treatment for substance use begins. This is an unlikely outcome, though intervention is nonetheless worthwhile. There is tentative evidence to suggest that reducing drinking among perpetrators can diminish resort to violence (Wilson et al., 2014) and that opiate substitution treatment can help alleviate dependence on illicitly purchased drugs and acquisitive crime and improve mental, physical and sexual health among heroin-dependent polydrug users (Gossop et al., 2000;Strang et al., 2010;MacArthur et al., 2014). But, although treatment interventions can reduce the harms of substance use, where drug and alcohol use and domestic abuse co-occur, treatment needs to be part of a range of measures that include support in changes in thinking and modes of relating, securing the housing and economic resources couples need to be able to contemplate living apart, the support and empowerment of survivors, the safeguarding of children and professional help with mental health problems. These skilled forms of intervention are critical to deescalating the dynamics that sustain substance use in the lives of people enduring the worst forms of domestic abuse but are often in short supply.
By contrast, the evidence that domestic abuse perpetrator programmes-as currently commissioned by the UK government-'work' remains mixed (Vigurs et al., 2017). Although the best interventions risk encouraging men who have been physically violent to adopt more emotionally abusive tactics (Kelly and Westmarland, 2016), the UK's Probation Inspectorate is doubtful as to whether the private Community Rehabilitation Companies currently delivering such interventions provide adequate practice in terms of safeguarding victims and their children (House of Commons, 2018). Both the domestic abuse and substance use treatment sectors in the United Kingdom have suffered sustained funding cuts over the last 10 years (Women's Aid, 2016; ACMD, 2017), often secured through the non-renewal of local procurement contracts via competitive tendering processes that favour cheaper and less specialist provision. One danger with compelling drug or alcohol treatment is that it will place clinicians and health practitioners in the ethically compromising position of having to report those who relapse, together with those whose prescriptions have proved insufficient, or who have decided that they would be better trying to reduce their substance use gradually, to the courts where they may face further criminalization (Seddon, 2007;Werb et al., 2016).
More generally, models of treatment for alcohol and drug use that acknowledge that 'relapse' is common are hard to reconcile with domestic abuse policy founded on compliance with court orders that insist upon 'zero tolerance' of reoffending (Benitez et al., 2010). Criminalizing responses are rarely challenged in domestic abuse policy, where academic research has tended to extol the benefits of naming 'perpetrators' as such and victims, though sometimes recognized as 'survivors', are usually cast as their opposites. Such an approach runs contrary to academic conventions in substance use research where a concerted effort has been used to avoid stigmatizing terminology that reduces individuals' identities to their drug consumption (Broyles et al., 2014).
Hence, acknowledgement of complexities in the power dynamics of domestic abuse that co-occurs with drug, alcohol and mental health problems raises acute challenges, not only for the delivery of policy that attempts to reconcile safety, justice and rehabilitation but also for academics who have framed the problem of domestic abuse primarily as one of either gender or psychology. Not only do criminologists need to reconceptualize domestic abuse more dynamically but they must also ask why some men choose to secure control in coercive ways when so many other aspects of their lives appear out of control. There is a need to recognize how the interdependencies-including the prospect of economic abuse-involved in intimate relationships are intensified by poverty, stigma, co-dependency, child abuse and neglect, poor mental health and the fear of police and social services intervention. In theory and in practice, we must ensure that shorthand explanations derivative of personality disorders do not obscure what can be learnt from the more complex descriptions both survivors and perpetrators can offer of their relationships. Policymakers need also to ensure that evaluations of treatment options for substance-using perpetrators extend beyond the longstanding fixation with acquisitive crime to include measures that take stock of their impact on children and partners, whether current and former, and to recognize that establishing effective practice will require the reestablishment of expertize and service provision that is increasingly scarce.