Intercountry Adoption Is a Child Protection Measure

Some Comments on “Investigating Historical Abuses” by Balk, Frerks and De Graaf (2022)

In: Journal of Applied History
Marinus H. van IJzendoorn Erasmus University Rotterdam The Netherlands Rotterdam
University of London UCL UK London

Search for other papers by Marinus H. van IJzendoorn in
Current site
Google Scholar
Marian J. Bakermans-Kranenburg Vrije Universiteit Amsterdam The Netherlands Amsterdam

Search for other papers by Marian J. Bakermans-Kranenburg in
Current site
Google Scholar
Open Access


In their article on ‘Investigating historical abuses’ Yannick Balk, Georg Frerks and Beatrice de Graaf (2022) present an applied history of intercountry adoption to the Netherlands over the past 70 years and conclude that a moratorium on intercountry adoption is necessary because of the many adoption abuses. In this paper we comment on their aims, methods, results, and conclusions. Applied historical analysis without considering the numerous empirical studies on the effects of (de-)institutionalization is problematic if the application is to impact policy. Furthermore, using inaccessible archival material and opaque triangulation hinders replication. An estimate of the overall frequency of adoption abuses is absent. Any adoption abuse is a serious violation of children’s rights and needs to be addressed. However, we argue that their findings do not necessitate the recommendation to (temporarily) stop intercountry adoption at the expense of children in institutions for whom intercountry adoption would be the last resort.

1 Introduction

A recent estimate of numbers of children left in institutions worldwide was estimated in 2020 to be 7.5 million.1 The number of children who became orphans during the COVID-19 pandemic is estimated to be around 5 million since 2020, and still counting.2 In many cases the wider social network will take care of these children, but many others run the risk of ending on the street or in institutional care. The recent war raging in Ukraine might add to these numbers.3 From our recent meta-analyses commissioned by The Lancet Psychiatry, covering more than 300 studies in more than 60 countries across 70 years, we had to conclude that institutional care has a devastating impact on children in all developmental domains, ranging from physical and brain growth to socio-emotional development.4

Institutional care is structural neglect because of fragmented 24/7 care by professionals working with large groups in shifts and in low-paid, high turn-over jobs.5 From an evolutionary perspective, institutional care falls outside the range of the environments of evolutionary adaptedness in which the human species evolved during millennia.6 Professional group-care is unique across evolutionary times and the institutions’ organization is atypical for any known environment created for the care of offspring.7 Structural neglect is not caused by bad intentions or lack of abilities in most of the staff or management but it is the implication of a total institution.8 Child protection and family-based care arrangements are urgently needed for institutionalized children, and a crucial question is whether international adoption should—for the time being—belong to the catalogue of child protection measures as a last resort when no other effective measures are available.9-10

In their paper on “Investigating historical abuses. An applied history perspective on intercountry adoption in the Netherlands, 1950s–present” Balk, Frerks and De Graaf argue that abuses related to international adoption were frequent, long-standing and persistent until today. They write that abuses are ‘endemic’ or ‘systemic’, inevitably part of an ‘adoption market’ and its ‘perverse financial stimuli’, against the background of global disparities between rich and poor countries, and the correlated disparity between the large demand for adoption children and the small supply thereof.11 It should be noted that the market metaphor, more often used in this context, is somewhat misleading. The millions of institutionalized children dwarf the demand, when we consider the estimated 400,000 children adopted to Western countries over the years 2004–2020, among them 9,114 adopted to the Netherlands.12 For example, Ukraine counted almost 800 orphanages housing more than 100,000 children in 2016.13 In that same year, 2016, a total of 399 Ukrainian children (less than 0.4 %) were adopted internationally.14 Despite the flawed market metaphor, however, the questions addressed by Balk and colleagues remain important: Did abuse of adoption to the Netherlands take place, how frequent were illegal adoptions and other adoption abuses, and are these abuses sufficient reason to call for a moratorium on international adoption to the Netherlands?

2 Estimating Adoption Abuses

What are the research findings reported by Balk, Frerks and De Graaf in their work for the Dutch Committee Investigating Intercountry Adoption (CIIA)? They analysed mainly archival material on adoptions to the Netherlands from five countries (Brazil, Sri Lanka, Bangladesh, Colombia, and Indonesia), covering about one-third of the international adoptions to the Netherlands (14,500 of 40,000 adoptions in total from 1950 to 2020). The five countries were mentioned in the research assignment to the CIIA and this selection was also inspired by the number of media reports on abuses in adoptions from those countries. Below we comment on the difference between media reports and scientific evidence. Adoption abuses from Brazil and Sri Lanka were elaborated in the paper and from the descriptions the following figures on ‘certified’ adoption abuses could be derived. From reports on Brazil, 45 out of 1800 adoptions were identifiable as abusing legal regulations or generally accepted norms. For Sri Lanka the authors could not estimate figures but on the basis of media coverage, oral history and local reports they concluded that adoption abuses must have been widespread. Three additional countries were studied more in depth to quantify adoption abuses. For Congo 30 abuses out of 140 adoptions were reported, from Haitian archival documentation 100 out of 1,100 adoptions could be characterized as abusive, and for China no figures on adoption abuses could be established due to the absence of relevant archives.

In their paper Balk and colleagues do not present an overall (lower or higher boundary) quantitative estimate of adoption abuses, nor an educated guess based on their privileged access to archival and other documents. But the answer to the question whether any abuse of adoption to the Netherlands did take place is clearly and definitively in the affirmative: Yes, illegal adoption took place, also in the Netherlands. That is disturbing as each and every abuse committed during adoption of vulnerable children is a serious legal and ethical violation of the rights of the children involved and of their biological or social parents if they are still alive and available. These adoption abuses should be sanctioned, apologized for, future abuses should be prevented, and damage inflicted on the victims (children and parents) should be repaired as much as possible. In agreement with the CIIA and based on their findings, the authors support recommendations for apologies by the Dutch government, the establishment of an independent National Centre of Expertise, and a moratorium on intercountry adoption. All three recommendations were part of the final CIIA report and have been implemented almost immediately after its publication in 2021. The authors therefore conclude that their study “resulted in an immediate policy impact” (p. 28), which is not surprising as the third author was one of the three members of the CIIA that was formally responsible for the report and its recommendations, and the other two authors conducted the research underlying the CIIA report.

3 From Is to Ought

But are the evidence base and the ethical argumentations sufficient for all three recommendations, in particular for a moratorium? In the short run a moratorium certainly stops adoption abuses but how is the collateral damage of such a stop weighed in this recommendation and subsequent implementation of a moratorium? Are the research findings of Balk et al. sufficient evidence base to interpret adoption abuses as endemic or systemic and lasting up until recent times? The 1993 Hague Adoption Convention, meant to regulate international adoptions, was signed by the Netherlands in 1998. The authors state that the Hague Convention did not prevent abuses from occurring with high frequency, although they do not quantify this frequency. Does it mean that a moratorium is the only sound recommendation to stop adoption abuse with the least damage to children? We should take into account that these children in the meantime might be waiting for adoption as a last resort because their basic rights are seriously violated on the streets or in orphanages when no other local solution is available.

To address these questions, we have to go back to the historical approach, central goal, and methods chosen by the authors. They argue to conduct ‘applied historical analysis’ with the exclusion of developmental research methods or findings. They combined archival analysis with oral history, analysis of parliamentary documents, and secondary literature. How this combination did corroborate, verify or ‘triangulate’ their findings is not explained. Their goal is to provide a ‘deeper, more thorough’ analysis than the developmental focus on the ‘contingent question’ of the best interest of the child and its legal protection. Their aim is to go beyond the polarized and politicized debate about the desirability of adoption raging in the developmental research community, with a historicizing approach that contextualizes abuses in history and provides ‘a narrative and genealogy’ of adoption abuses avoiding ‘hindsight bias’. This is a laudable but utterly strong ambition, certainly when the ultimate goal of their applied history is that their findings “need to be translated into revised governmental policies” (Abstract of the article). In our paper on ‘Replication crisis lost in translation?’ we argued that translating research evidence into policy or practice is jumping from is to ought, which requires the use of replicated findings as well as sound ethical reasoning.15 Both the replicability of the findings and the ethical foundation of the recommendation for a moratorium leave some room for doubt.

4 A Bridge over Troubled Water

Replicability is the first pillar of a bridge from is to ought. According to the authors replicability of their research is problematic because “our governmentally-sanctioned project was granted unrestricted access to all relevant archives, non-public ones included. That means that other researchers cannot replicate some of our findings.” (p. 6). What data are inaccessible for other researchers remains in the dark, and inter-rater reliability within the research team has not been reported. The biomedical, neuroscientific and behavioral sciences have learned the hard way how crucial replicability is; open and transparent access of data is tantamount to replicability. Core difference between scientific research and investigative journalism is replicability of the findings, which in the scientific forum is open to scrutiny by independent colleagues. In journalism credibility and accountability are important, but confidential sources without the possibility of replication are allowed to play a role in the journalist’s narrative (e.g., Statement of Principles for Investigative Journalism, Canadian Association of Journalists).16 This makes media reports of adoption abuse problematic sources for research findings. At the same time, scientific research with historical or empirical methods that are dependent on inaccessible sources or data are problematic, in particular if no public access is expected to be warranted in the future. Permanently privileged access to data is incompatible with replicability. This may also point to the need to make archival governmental data accessible in a way that does not violate individuals’ privacy.

A sound ethical argument is the second pillar of the bridge. But a firm ethical basis for translation of research findings to policy recommendations (from is to ought) is absent in a study when it leaves out findings from developmental research on the impact of international adoption on the triangle of birth parents, adoptive parents, and the adoptees. An exclusive focus on the historical context of adoption abuses is of course legitimate. Specialized research must limit its domain of inquiry to be able to dive deep into the socio-historical processes leading to adoption abuses, and such a specialization is a welcome complement to other disciplinary approaches. But it is indeed only one approach. Problems may emerge when such a one-sided perspective is used for prescriptions to change governmental policies beyond the specific domain of research, in this case the (inadequate) monitoring of legal and procedural frameworks for intercountry adoption. In previous work we proposed to fill the gap between is and ought using an ethical argument based on the ‘original position’ developed by Rawls in his theory of justice (1971/1980).17-18 The idea is to imagine that before birth we do not know our position in the adoption triangle when deciding about the ethical justification of adoption. In that case a justifiable decision would be to protect the minimum conditions for the basic rights and life chances of the most vulnerable party, which in this case is the child who lost its parents or withers away in institutions as a social orphan. Of course, the first imperative is the prevention of this deplorable situation. But in a world with millions of children in this situation, the ethical perspective combined with much replicated empirical evidence about the negative developmental prospects of institutionalized children, international adoption can be argued to be warranted as a last resort child protection measure.

One of the reasons Balk and colleagues do not take into account the empirical evidence collected during the past seven decades is their skepticism about the validity of those research findings. They state that developmental researchers are biased and have used predominantly small samples: “Most researchers have a background in American academia, which in itself colors their perspective. What is more, many key researchers are adoptive parents themselves, a positionality that inevitably bears on the analysis.” (p. 3). If it is suggested that impartiality is only possible without being a member of the population studied, it might be reassuring that some of the most influential adoption researchers (as indicated by citation impact) are non-American, non-adoptive parents, including the authors of the current paper. Furthermore, to counter-act biased conclusions we proposed to take only meta-analytic findings into account when translating research to policy.19 Our set of meta-analyses reported in The Lancet Psychiatry covers literally all empirical studies with the largest variety of results about the (potentially positive and negative) effects of institutionalization and adoption.20 We do agree that the statistical power of adoption studies is often below par because of small sample sizes, but combined in meta-analyses they leave little room for false positives or false negatives, and meta-analyses show the strong positive impact of adoption on child developmental outcomes.21 Furthermore, in the Netherlands some of the largest surveys have been conducted. For example, Van Ginkel and colleagues studied a population-based cohort of 2,360,450 Dutch youth, including 10,602 internationally adopted peers, and found that Dutch adoptees generally do not use more medication for mental or physical health issues than their non-adopted peers, which illustrates the children’s impressive catch-up growth and development.22

5 Conclusions

It is critically important to give weight to research about the impact of institutional care on child development and effects of adoption on previously institutionalized children when it comes to recommendations about governmental policies. Only based on the historical record of adoption abuses, such recommendations run the risk of neglecting the damaging effects of structural neglect on children who have no choice but growing up in institutional care. Even in well-equipped and small institutions the damage done is substantial; children do need family-based care arrangements to grow up.23-24 Having said that, it would be short-sighted to recommend intercountry adoption without considering adoption abuses, whether they are ‘systemic’ or not. No single adoption abuse can be tolerated, and any legal, socio-economic and political measure (except the total abolishment of intercountry adoption) should be used to prevent such abuses from happening. We missed a recommendation that the national government(s) take their responsibility for closer and more careful systematic monitoring of compliance with the legal procedures for intercountry adoption. That would more logically follow from the committee’s findings than the recommendation of a moratorium on intercountry adoption. As long as millions of children are still subjected to institutional care, intercountry adoption into families functions as a last resort child protective measure. Our recommendation would be to first fight adoption abuses and start tearing down orphanages, and when de-institutionalization is successfully completed, then abolish international adoption because it has become obsolete as a last resort for the protection of vulnerable children.25


MHvIJ is supported by an award from the Netherlands Organization for Scientific Research (Spinoza prize). MHvIJ and MJB-K are additionally supported by the Gravitation program of the Dutch Ministry of Education, Culture, and Science and the Netherlands Organization for Scientific Research (NWO grant number 024.001.003). Their work on The Lancet Commission on (de-)institutionalization was partly supported by Lumos.


C. Desmond, K. Watt, A. Saha, J. Huang, and C. Lu, “Prevalence and number of children living in institutional care: global, regional, and country estimates.” Lancet Child and Adolescent Health 4(5) (2020), 370–377.


S.D. Hillis, H. Unwin, Y. Chen, L. Cluver, L. Sherr, P.S. Goldman, O. Ratmann, C.A. Donnelly, S. Bhatt, A. Villaveces, A. Butchart, G. Bachman, L. Rawlings, P. Green, C.A. Nelson and S. Flaxman, “Global minimum estimates of children affected by COVID-19-associated orphanhood and deaths of caregivers: a modelling study.” Lancet Child and Adolescent Health 398(10298), (2021). 391–402.



M.H. van IJzendoorn, M.J. Bakermans-Kranenburg, R. Duschinsky, N.A. Fox, P.S. Goldman, M.R. Gunnar, D.E. Johnson, C.A. Nelson, S. Reijman, G.C.M. Skinner, C.H. Zeanah and E.J.S. Sonuga-Barke, “Institutionalisation and deinstitutionalisation of children 1: a systematic and integrative review of evidence regarding effects on development.” Lancet Psychiatry 7(8) (2020), 703–720.


M. Dozier, J. Kaufman, R.R. Kobak, T.G. O’Connor, A. Sagi-Schwartz, S. Scott, C. Shauffer, J. Smetana, M.H. van IJzendoorn, and C.H. Zeanah, “Consensus statement on group care for children and adolescents: A statement of policy of the American Orthopsychiatric Association.” American Journal of Orthopsychiatry, 84 (3), 219–225.


J. Bowlby, Attachment and Loss, Vol. 1: Attachment. (London: Penguin, 1969).


S.B. Hrdy, Mothers and Others: The Evolutionary Origins of Mutual Understanding. (Cambridge, MA: Harvard University Press, 2009).


M.H. van IJzendoorn and M.J. Bakermans-Kranenburg, “ ‘Tear down your institutions’. Empirical and evolutionary perspectives on institutional care in SOS Children’s Villages.” PsyArXiv (2021).


P.S. Goldman, M.J. Bakermans-Kranenburg, B. Bradford, A. Christopoulos, P. Ken, C. Cuthbert, R. Duchinsky, N.A. Fox, S. Grigoras, M.R. Gunnar, R.W. Ibrahim, D. Johnson, S. Kusumaningrum, N. Agastya, F.M. Mwangangi, C.A. Nelson, E.M. Ott, S. Reijman, M.H. van IJzendoorn, C.H. Zeanah, … E. Sonuga-Barke, “Institutionalisation and deinstitutionalisation of children 2: policy and practice recommendations for global, national, and local actors.” Lancet Child and Adolescent Health 4(8) (2020), 606–633.


J. Palacios, S. Adroher, D.M. Brodzinsky, H.D. Grotevant, D.E. Johnson, F. Juffer, L. Martínez-Mora, R.J. Muhamedrahimov, J. Selwyn, J. Simmonds and M. Tarren-Sweeney, “Adoption in the service of child protection: An international interdisciplinary perspective.” Psychology, Public Policy, and Law 25(2) (2019). 57–72.


Y. Balk, G. Frerks, and B. De Graaf, “Investigating Historical Abuses. An Applied History Perspective on Intercountry Adoption in the Netherlands, 1950s–Present.” Journal of Applied History (2022) 1–28. DOI: 10.1163/25895893-bja10020.


P. Selman, “Global Statistics for Intercountry Adoption: Receiving States and States of origin 2004–2020.” February 2022 (accessed March 10, 2022).


N.A. Dobrova-Krol and M.H. van IJzendoorn, “Institutional care in Ukraine: Historical underpinnings and developmental consequences.” In Child maltreatment in residential care: History, research, and current practice, eds. A.V. Rus, S.R. Parris and E. Stativa. (Cham: Springer International Publishing, 2017), 219–240.


P. Selman, “Global Statistics for Intercountry Adoption: Receiving States and States of origin 2004–2020.” February 2022 (accessed March 10, 2022).


M.H. van IJzendoorn and M.J. Bakermans-Kranenburg, “Replication crisis lost in translation? On translational caution and premature applications of attachment theory.” Attachment & Human Development 23 (2021) 422–437.



J. Rawls, A theory of justice. (Boston: Harvard University Press, 1980).


M.H. van IJzendoorn and F. Juffer, “The Emanuel Miller Memorial Lecture 2006: Adoption as intervention. Meta-analytic evidence for massive catch-up and plasticity in physical, socio-emotional, and cognitive development.” Journal of Child Psychology and Psychiatry, 47, (2006), 1228–1245. doi:10.1111/j.1469–7610.2006.01675.x.


M.H. van IJzendoorn and M.J. Bakermans-Kranenburg, “Replication crisis lost in translation? On translational caution and premature applications of attachment theory.” Attachment & Human Development 23 (2021) 422–437.


M.H. van IJzendoorn, M.J. Bakermans-Kranenburg, R. Duschinsky, N.A. Fox, P.S. Goldman, M.R. Gunnar, Johnson, D.E., C.A. Nelson, S. Reijman, G.C.M. Skinner, C.H. Zeanah and E.J.S. Sonuga-Barke, “Institutionalisation and deinstitutionalisation of children 1: a systematic and integrative review of evidence regarding effects on development.” Lancet Psychiatry 7(8) (2020), 703–720.


M.H. van IJzendoorn, M.J. Bakermans-Kranenburg, R. Duschinsky, N.A. Fox, P.S. Goldman, M.R. Gunnar, D.E. Johnson, C.A. Nelson, S. Reijman, G.C.M. Skinner, C.H. Zeanah and E.J.S. Sonuga-Barke, “Institutionalisation and deinstitutionalisation of children 1: a systematic and integrative review of evidence regarding effects on development.” Lancet Psychiatry 7(8) (2020), 703–720.


J.R. van Ginkel, F. Juffer, M.J. Bakermans-Kranenburg and M.H. van IJzendoorn, “Do internationally adopted children in the Netherlands use more medication than their non-adopted peers?” European Journal of Pediatrics 175(5) (2016), 715–725.


M.H. van IJzendoorn and M.J. Bakermans-Kranenburg, “ ‘Tear down your institutions’. Empirical and evolutionary perspectives on institutional care in SOS Children’s Villages.” PsyArXiv (2021).


M.H. van IJzendoorn, M.J. Bakermans-Kranenburg, R. Duschinsky and G.C.M. Skinner, “Legislation in search of “good-enough” care arrangements for the child: A quest for continuity of care.” In The Oxford Handbook of Children and the Law, ed. J.G. Dwyer (Oxford: Oxford University Press, 2020), 129–153.


M.H. van IJzendoorn and M.J. Bakermans-Kranenburg, “Tearing down or fixing up institutional care for abandoned children? Comment on Rygaard (2020).” American Psychologist 77 (2022), 135–137

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 826 362 31
PDF Views & Downloads 1077 355 24