Background and sociocultural context

It is generally understood that prior to colonisation of Aotearoa New Zealand Māori children were raised in an extended rather than nuclear family. Cooper and Wharewera-Mika (Reference Cooper, Wharewera-Mika, McIntosh and Mulholland2011) note Māori parenting was thought to be indulgent and caring with customary beliefs and practices valuing family and children being regarded as treasures. Taonui (Reference Taonui2010: 192) notes that ‘traditional whakataukī’ (aphorisms) expressing values and codes of conduct speak of the importance of children and the bond between parents and children. It is strongly held by many that contemporary child welfare statistics reflect the severe downstream and ongoing impacts of land and language loss, cultural and political alienation, racism and deeply entrenched inequalities that accompanied colonisation (Taonui, Reference Taonui2010; Ware et al., Reference Ware, Breheny and Forster2017). While Māori comprise 28 per cent of the child population, they account for 40 per cent of those notified to the statutory social work agency and made up 60 per cent of the children in care in 2015 (Ministry of Social Development, 2015). These disparities mean that many Māori families are deeply suspicious of the statutory agency, especially given a history of racism and abuse in state care (Ministerial Advisory Committee, 1984; Stanley, Reference Stanley2016). Ware et al. (Reference Ware, Breheny and Forster2017: 515) note that further differences in health and education outcomes, colonialism and negative beliefs about young Māori parents in particular go on to ‘produce the “problems” of social exclusion, such as intergenerational welfare dependency and negative outcomes’ for families.

In Aotearoa New Zealand one particular child abuse tragedy (the death of twin babies in a Māori family) led to a series of political events (Beddoe, Reference Beddoe2014; Hackell, Reference Hackell2016) in which welfare austerity policies came together with a demand for a more authoritarian and surveillant approach to child protection. This demand resulted in shifts to a more individualised approach which privileges personal responsibility and psychological dysfunction over structural accounts of cycles of poverty and historical trauma (Pihama et al., Reference Pihama, Reynolds, Smith, Reid, Smith and Nana2014). Hyslop and Keddell (Reference Hyslop and Keddell2018: 5) argue that the disproportionate number of Māori children who figure in these events, and in all parts of the child protection system, ‘has been a significant driver in shifting the narrative away from family empowerment and towards a child rescue paradigm’. The aforementioned case brought together racism, conservative concerns about welfare dependency and gendered violence (men) and moral failing (women), with Māori men and women a particular target (Beddoe, Reference Beddoe2014; Hackell, Reference Hackell2016).

The politics of child protection: common sense

The main focus of this article, the ACEs checklist, is not yet widely used as a diagnostic tool within Aotearoa New Zealand. Anecdotally we do know that social workers are starting to be given the checklist as part of training on ‘trauma informed practice’ which is being delivered across both statutory and NGO sector child welfare services.

Child protection does not happen in a politically neutral vacuum. Politicians with responsibility for children’s services face a challenging task as each child death tests the very idea that child protection can work. It is increasingly argued that it is perhaps an unattainable mission:

Adverse childhood experiences: the ACEs checklist

The ACEs checklist was first conceptualised as the result of an attempt to understand links between childhood experiences and long-term health outcomes (Felitti et al., Reference Felitti, Anda, Nordenberg, Williamson, Spitz, Edwards and Marks1998). It is important to understand that this initial research was undertaken with a US based population sample consisting of, mostly, white, average to well off, insured patients within a medical setting (Wade Jr et al., Reference Wade, Cronholm, Fein, Forke, Davis, Harkins-Schwarz, Pachter and Bair-Merritt2016), and that the questions developed for the checklist were not the result of a rigorous review to select those variables most likely to predict health outcomes (Finkelhor, Reference Finkelhor2017). Using such populations, with limited samples, as a ‘norm’ from which to judge global populations is a noted problem (Henrich et al., Reference Henrich, Heine and Norenzayan2010), and despite a growing body of research showing some global generalisability for the ACEs checklist (Stoltenborgh et al., Reference Stoltenborgh, Bakermans‐Kranenburg, Alink and van Ijzendoorn2015; Hughes et al., Reference Hughes, Bellis, Hardcastle, Sethi, Butchart, Mikton, Jones and Dunne2017), we would add our voices to the calls from others in this themed section that the ACEs checklist often fails to consider wider societal pressures. Of particular concern for us in Aotearoa New Zealand is that the ACEs checklist fails to adequately account for the (often) entangled effects of poverty, racism, and colonisation.

A critique of ACEs through considering incarceration rates

There have been calls to consider expanding the ACEs checklist to incorporate other, more societally based factors, such as bullying, neighbourhood, schooling and socioeconomic status (Finkelhor et al., Reference Finkelhor, Shattuck, Turner and Hamby2013); however, to our knowledge, no one has considered the list with regard to colonisation and indigenous peoples. It is important to note that parenting assessment frameworks are often another tool of colonisation reifying Western notions of family and children (Choate and Lindstrom, Reference Choate and Lindstrom2017), and whilst the ACEs checklist is not specifically a parenting assessment, it does ask questions that assess parenting behaviours and assumes certain family structures. It is not our intention here to provide a comprehensive overview of what a critique of ACEs from a colonisation or even decolonisation perspective might look like (and arguably as Pākehā (New Zealand European) women our views on this are limited). However, we would like to sketch out some concerns we have by using one of the questions in the original checklist by way of example.

Within the original ACEs checklist, respondents are asked ‘Did a household member go to prison?’ (Felitti et al., Reference Felitti, Anda, Nordenberg, Williamson, Spitz, Edwards and Marks1998). We note there are substantial associations between parental incarceration and childhood outcomes (Bell et al., Reference Bell, Bayliss, Glauert and Ohan2018). However, we want to trouble the notion that analysis of the relationship between parental incarceration and childhood outcomes should stop there. Incarceration rates are differentially impacted by ethnicity, a point that Allen and Abresch (Reference Allen and Abresch2018: 286) note in cautioning that ‘social determinants often lie upstream from ACE variables’. Thus, we suggest that a concentration on incarceration as a question obscures the impact of racism in considering what might be adverse childhood experiences; and, critically, it individualises a larger societal problem.

In Aotearoa New Zealand, much like many other colonised nations, our indigenous population, Māori, experience differential incarceration rates (Morrison, Reference Morrison2009). These differential rates are not limited to convictions: rather, we know that, in comparison to Pākehā, Māori are four to five times more likely to be apprehended, prosecuted and convicted, seven and a half times more likely to receive a custodial sentence, and eleven times more likely to be remanded in custody (Morrison, Reference Morrison2009). For Māori women, in comparison to Pākehā women, the figures are more stark, with them being five and a half times more likely to be apprehended and ten times more likely to be given a custodial sentence (Morrison, Reference Morrison2009).

Whilst the ACEs question, as it is framed in the checklist, asks about ‘prison’, the sub-heading states this is a measure of ‘criminal behaviour in the household’: a small but not insignificant distinction that demonstrates lack of rigour. Whilst it could be argued that prison is a measure of criminal behaviour, as a question this is limiting. This can be illustrated through examining the differential rates for less serious sentences: Māori are more likely than Pākehā to be given community service and/or periodic detention than a monetary fine, something Triggs (Reference Triggs1999) notes is likely due to financial resources. Thus poverty (and presumably poverty exacerbated by racism) impacts on whether a person is able to avoid certain sentences despite the initial crime being similar. The inevitable result, in terms of the ACEs checklist, is that the criminal activities of Pākehā parents are significantly less likely to be captured than those of Māori parents, thus creating ‘false negatives’ in addition to failing to capture the impact of institutional racism. In both examples, what is done in response to the information gathered through surveillant data collection is a crucial question – but one beyond the limits of this brief article. Therefore, the question itself, set up to measure the impact of parental criminal behaviour (through parental incarceration) on children, often measures the impact of racism (in the case of Māori children) and relative privilege (in the case of Pākehā children).

We would also note this is not a problem that is likely to be confined to those countries with indigenous populations and a history of colonisation, as differential incarceration rates can be seen across other minority populations in other countries (Mitchell, Reference Mitchell2005). For example, we hypothesise that the increased surveillance of Muslim families for radicalisation, something that social workers are being drawn into (Stanley and Guru, Reference Stanley and Guru2015), could well result in differential incarceration rates across Western nations, again with the ACEs question set up to measure the impact of parental incarceration on children, instead demonstrating the impact of racism, thus creating similar challenges to those we have discussed. These approaches while superficially appearing to be ‘common sense’ mask deep inequalities and ongoing harm.

By way of example, recent attempts to introduce a data driven risk checklist in Aotearoa New Zealand have seen the development of a predictive risk model that uses existing government data to generate ‘scores’ of risk for any given individual child (Keddell, Reference Keddell2015). An analysis of the performance of the model found that Māori families were over-represented in the results, with the researchers unsure of the reasons for this (Rea and Erasmus, Reference Rea and Erasmus2017). However, the recommendation from the researchers were that the model should still be deployed (Rea and Erasmus, Reference Rea and Erasmus2017). To date the model has not been implemented, with the new Minister stating she has no immediate plans to do so (Martin, Reference Martin2018). However, this has not stopped derivations of the Aotearoa New Zealand model being used in other areasFootnote ² .

Finally, whilst we note that there is some acknowledgement by the police in Aotearoa New Zealand that racism, at least of the implicit kind, is a problem (Bush, Reference Bush2015) and there is therefore a move to change attitudes, change is likely to be slow and does not mitigate against harms already done. Further, given the ACEs framework is one based on retrospective experiences, it is likely to be years before one could conceivably, and even ethically, ask a question about parental incarceration without considering the impact of racism (for either Māori or Pākehā children).