The Legal Education Foundation has published a report about the collection, use and commercial exploitation of data from the justice system, focusing in particular on the new judgments database set up by The National Archives earlier this year. It reveals both a lack of public awareness and a surprising level of public concern over the re-use of such data by third parties, and recommends more transparency in how such data is managed.
The report is called Justice Data Matters: building a public mandate for court data use. It combines the results of a public survey by Ipsos with views from more focused discussion groups to explore public attitudes to the publication and use of data held in court records, including judgments.
This is interesting because it addresses a key concern about transparency, and the reason why there is often a tension between the expectation of openness and freedom of expression on the one hand, and that of privacy and confidentiality, particularly for individuals and their family life and problems, on the other. At the end of the day it’s the courts that should be transparent, not the people caught up in the cases decided there. But it’s hard to do one without the other. So in every case there’s a balance to be struck.
This is not the first time the Legal Education Foundation (LEF) has reported on matters of justice data. In October 2019 it published published Digital Justice: HMCTS data strategy and delivering access to justice, a report authored by its director of research Dr Natalie Byrom, which highlighted the need for comprehensive and consistent data collection by HM Courts and Tribunals System in order to optimise its reform of the justice system, and made a number of recommendations as to how it might be improved. One of those recommendations was that
“HMCTS should work with the judiciary and colleagues in the Ministry of Justice to commission an independent report which reviews the current arrangements for disseminating judgments to the public and maps the information flows from courts to publication. On the basis of this report, HMCTS and the MoJ should engage with key stakeholders to develop a publication solution that delivers free and comprehensive access to judgments in a structured machine-readable format.”
It was ultimately in response to that suggestion that the Ministry of Justice (MoJ) announced last year that it would cease its funding for the well established judgments database known as BAILII (the British and Irish Legal Information Institute) and instead invite The National Archives (TNA) to create a new ‘home’ for court and tribunal judgments.. However, this new home would only house cases from England and Wales (unlike BAILII which also, as its name implies, covers Scotland, Northern Ireland and indeed the Republic of Ireland and various other jurisdictions).
The new site, Find Case Law ( caselaw.nationalarchives.gov.uk) was launched in April this year. We have written about some of its teething issues here: The National Archives and Family Court transparency – a temporary glitch?
However, the new report from the LEF is not about who should publish judgments. It’s about whether and how they and other court data should be made publicly available, how that should happen, and who should be able to use it and for what purposes. These are important questions.
The launch of the new service “lays bare both the enormous opportunities and risks of sharing data of this kind” says Sir Bob Neill MP, chair of the House of Commons Justice Committee in his foreword to the report.
“In taking this body of work forward, the Government will have to juggle a difficult balancing act between, on the one hand, delivering the significant and undeniable public good of digitising court data, while on the other, avoiding the pitfalls of over-commercialisation and automation, both of which have the potential to undermine public confidence, and worse, hinder justice.”
Among the key findings of the research, the report identifies “strong public support for robust governance and oversight of commercial use of court record data” and says “transparency about access requests and data use is key to maintaining public confidence” in the use of such data by third parties.
All this makes sense, but some of the other findings raise questions about how well informed the respondents were about what goes on now. Perhaps that is part of the point. The report says “Respondents felt that the government did not keep the public informed about these matters”. It also notes “especially high levels of discomfort” over the use of court data by “third party organisations” and concerns over the “increased commodification of the justice system”. Participants in the polling were said to be “least comfortable with third party access to ‘help commercial companies to develop products and services to be used within the justice system’.” The survey showed that
- Over 70% of participants said that they knew nothing or not very much about the information contained in court records, or about who has access to court records.
- 50% of polling respondents expressed discomfort about use of court data by tech companies, credit rating agencies (42%), and insurance companies (42%).
- While 56% said they were comfortable with the information from court records being used to improve judges’ decision-making or reduce costs in the justice system, only 26% were comfortable with commercial companies having access to develop products and services.
Judgments as data
If we’re talking about court judgments, we need to bear in mind that until recently the only substantial source of such data was third party publishers, who took the trouble to record and collect the cases that they considered important enough to publish. The Incorporated Council of Law Reporting for England and Wales, as its name suggests, was set up (as an independent charity) for this very purpose, back in 1865. Prior to that there were individuals who for reasons of commerce or vanity published reports of cases under their own names (hence the collective description, the Nominate Reports). Then there were the commercial and academic publishers of law reports. When BAILII was founded at the turn of the Millennium, as an online digital resource, it was set up as a charity with the aims of furthering the Free Access to Law Movement (FALM). None of this was managed by the courts, though the judiciary played a big part in the process by contributing their judgments and approving their publication.
All these “third parties” were developing “products or services to be used within the justice system”, because how else could the courts have access to the legal precedents they were required to follow in accordance with our common law system? The fact that the Ministry of Justice has belatedly woken up to the realisation that it should be archiving judgments as public records, and has made arrangements for that to happen, should not suddenly deprive their use by third parties of legitimacy, simply because the public don’t properly understand or consent to it.
It seems astonishing that barely more than half of the general public (56%) understand something as basic as how citing precedents and using other court data might be beneficial to the administration of justice. Yet (as the graphic above shows) only 32% were “comfortable with law firms and lawyers not involved in those cases” having access to such content, while another 32% were actually “uncomfortable”. This seems to betray a pretty fundamental misunderstanding of open justice.
In this regard the report makes a couple of points about restrictions on access to judgments. First:
“When informed about the existence of private publishers of court judgments participants were alarmed about the lack of transparency of the data sets held by them and the possibility that they have more court judgments than are publicly available. Some questioned whether we currently have genuine open justice with private publishers charging for access to judgments that are not available to the public.”
This may sound sinister but there is a perfectly simple explanation for it: the additional content held by “private publishers” mainly consists of oral judgments which they have paid to transcribe, but which has not been provided to BAILII or The National Archives. That is wholly within the MoJ’s power to remedy.
Secondly (and much as you’d expect):
“Participants were informed that judgments from certain cases, particularly family court cases involving children, are not generally available to the public. This was something many participants were supportive of, feeling that such cases are more personal and sensitive.”
But it’s worth bearing in mind that some litigants and campaigners, as well as journalists and legal commentators, would like even these private cases to be subject to proper public scrutiny.
Other types of data
Obviously judgments are only one type of court data, and atypical in the sense that they are mostly unstructured long form texts rather than the kind of data points, or little nuggets of data, that might be processed by way of statistical analysis. Although bulk analysis of judgments data is possible, it is those other, more statistical kinds of data – such as details of the personal characteristics of those caught up in the justice system, or the types of claim they make, the time it takes to process their cases and the outcomes they receive – that were so woefully lacking when the Byrom report on Digital Justice was published. That data is essential to the future development of the justice system. As Sir Bob Neill pointed out:
“The more we know and understand the justice system – where it is operating well, how it might be unfairly impacting some, and when and where it requires further resources – the better positioned we are, as legislators, policy makers and practitioners, to take steps to improve it. Data is the main tool at our disposal to make those inroads.”
The problem is that once such data has been collected and made available for re-use, there is always a risk that it might be re-used for the wrong purpose. The report finds that
“Not all purposes for using the data in court judgments were considered acceptable or justifiable by the public. … The use of data for predictive purposes, and to develop low-cost alternatives to court for those on low incomes, were particularly controversial.”
There was a concern that “the growth in products to predict case outcomes might be used to encourage people on low incomes not to pursue their cases in court”. The fact that small claims are put through an alternative, digital system also prompted suspicions of a “two-tier” justice system, one for the rich and another for the poor.
But predictive analysis is controversial for other reasons too. The judiciary are wary of it, perhaps nervous about the exposure of unconscious biases or the fact that it accords too little weight to judicial discretion and reduces their decisions to a purely mechanical equation. (Such analytics, though widely used in some jurisdictions, such as the United States, are banned outright in others, such as France, where legislation has made it a criminal offence.)
Ultimately what the report demonstrates is that people are nervous about personal data being misused or exploited in ways that might benefit commercial players without necessarily contributing a public benefit. In that sense, it is very similar to public attitudes towards the use of health data. The aim of the report is therefore to develop what it calls a “public mandate” for the use of justice data. “With trust in the justice system on the line, what does the public consider acceptable when it comes to sharing and using data?”
The report makes a number of recommendations aimed at improving public understanding of the use of justice data, through greater transparency and better communication, and at ensuring that such use is regulated in accordance with public expectations. This is as much as matter of public legal education as anything else. (It also calls for “further research” – so perhaps we can expect another report in due course.)
It is of course important that HMCTS collects data for its own purposes – something it failed to do before, and which it was urged to do in order to inform and shape its ongoing digital reform of the justice system. The question that this report addresses is whether and how to allow third parties to exploit that data. Should it be sold to the highest bidder for anyone to use, such as credit rating agencies, insurers, litigation funders and so on? Or should the courts be more costive about it, and only allow its use on very restrictive, stringent conditions?
Whatever your views, this is a debate worth having. But it needs to be an informed debate, and it’s clear from this report that public understanding is currently insufficient.
This post originally appeared on the Transparency Project blog and is reproduced with permission and thanks