Nic Madge analyses the new Master of the Rolls’ vision for civil justice and its implications for access to justice.
A radical vision
On 11 January, Sir Geoffrey Vos became Master of the Rolls and head of civil justice in England and Wales. In recent months, he has advocated revolutionary reforms to civil procedure; ‘a fresh approach’
that goes far beyond the existing HM Courts & Tribunals Service (HMCTS) reform programme
. He proposes the creation of an entirely new civil justice system and, by implication, ripping up the Civil Procedure Rules. Sir Geoffrey is likely to be the most radical Master of the Rolls since the modern version of the office was created by the Judicature Acts of 1873 and 1881.
Sir Geoffrey’s vision can be seen most clearly in his speech to Harvard Law School
on 4 December. His starting point, when considering civil justice is ‘that almost every individual and business … can get whatever they want instantly with a few clicks on their smart devices’.
Sir Geoffrey proposes ‘a single point of online entry for every dispute however small or large, whether civil, family, commercial or administrative … A data set can be created from the outset and the dispute can then be directed towards the most appropriate resolution mechanism.’ Every case should ‘enter a metaphorical online funnel. Resolution interventions will cause many of them to settle, but if they do not settle the process will be directed at identifying and then resolving the issues that divide the parties.’
In creating a ‘modern dispute resolution system … the emphasis should be on the resolution rather than the dispute … [T]he process should not … be governed by the concept that every case will end up in a traditional courtroom with all the witnesses, parties, lawyers and a judge gathered together in the same place at the same time’ whether that gathering is actually physical or remote. The process should be about ‘achieving a resolution rather than about exacerbating or even necessarily always deciding who is right about the dispute that gave rise to the process’. Litigants should ‘think of any fee they pay as a fee for the solution of the problem not for the hearing of a dispute’.
His view is that ‘it would not be sufficient to put our existing procedures online’. ‘[W]e should not undertake reform by simply doing digitally what was previously done in an analogue way.’ He envisages a far greater role for artificial intelligence (AI), referring with approval to ‘the 60m small civil issues resolved every year by artificial intelligence on eBay’. ‘Why,’ he asks, ‘would one want online process that simply required the same old pleadings and formal documents to be drafted and uploaded to the dispute resolution platform when simple programming can arrive at the issue in an ordinary case far more efficiently?’
Sir Geoffrey recognises that as ‘the online programmes are developed within the funnel to an ever-wider category of case, some will need in the early stages to leave the online space to be dealt with under the legacy system … [but] the ultimate objective will be to create a ubiquitous online dispute resolution process that can resolve almost all kinds of issues at a proportionate cost in a reasonable time scale without the stress of unnecessarily confrontational set piece hearings’.
He argues that the present system overcomplicates issues. ‘In most cases the issue is really very simple. Was the car going too fast? Did John lend Mary $500? Did the defendant pay the mortgage instalments that were due? The objective in all cases must be to identify the real issue in dispute by the quickest possible route … However voluminous the documentation, I always found [in practice at the bar that] even the most complex case boiled down to an analysis of a very few documents and perhaps one or two difficult points of law. The trick is to identify the single point or the very few real points whether evidential or legal at an early stage. Artificial intelligence and the legal discipline of experienced lawyers can help with this.’ However, ‘[t]here is no need for lawyers to specialise, eg, in drafting repetitive court documentation when the issues that divide parties can be distilled and resolved online in half the time and at a quarter of the cost’.
Sir Geoffrey suggests that ‘in many cases traditional oral hearings may not be necessary’. What is needed is a ‘streamlining that aims to resolve online as large a percentage of the cases as possible using a series of targeted interventions’. Despite the increased use of dispute resolution processes and AI, ‘there may still be a need for remote or face-to-face hearings or meetings at some stages and in some cases but these should be tailored to resolution of specific issues rather than being allowed to transform themselves into costly and lengthy pitched battles’. He draws on lessons from the way in which courts have adapted to the pandemic, saying that, ‘Remote hearings have gone well’ and judges, lawyers, witnesses and parties were quick to realise their advantages, with many ‘seemingly reluctant to return to face-to-face hearings even when they are safely available’. However, ‘they are not in themselves a solution; they simply replicate old process in technologically enabled form’.
Sir Geoffrey concluded the lecture by saying, ‘I do not think judges and lawyers have a choice about the direction of travel. The only real question is when they get with the programme. Our present method of courthouse-based dispute resolution is simply not fit for the present era.’ He wants to ‘fashion new online dispute resolution mechanisms that can provide what my generation never had, namely access to justice for all levels of dispute whether trivial and soluble by Amazon or eBay or serious and requiring the attention of a state judiciary’.
These ideas, and much of the language used, appear to draw heavily on the work of Professor Richard Susskind,1See, for example, Online Courts and the Future of Justice, OUP, November 2019; ‘The future of courts’, The Practice: Remote courts, vol 6, issue 5, Harvard Law School, July/August 2020; and speaking at the Court Excellence and Innovation conference in Dubai, reported at: Michael Cross, ‘No basis for face-to-face justice, says digital court guru Susskind’, Law Society Gazette, 8 November 2018.
IT adviser to the Lord Chief Justice, who questions whether courts are a service or a place and argues that justice requires that most disputes should be resolved by judges working online and not in courtrooms. He expects technology to transform the court service within 10 years. ‘[B]y default, all cases, in not many years to come, should be conducted online unless there are compelling reasons to assemble in a courtroom'.2‘The future of courts’, The Practice: Remote courts.
He argues that self-service online courts will open up a latent market among claimants who today see justice as too slow, costly and antiquated. ‘No basis exists in jurisprudence or legal philosophy for conducting justice face to face in physical premises.'3See: ‘No basis for face-to-face justice, says digital court guru Susskind’.
‘Online is not an alternative to the courts system, it is the courts system.’4See: ‘No basis for face-to-face justice, says digital court guru Susskind’.
He describes online judging as ‘an asynchronous process involving human judges receiving arguments and evidence in electronic form and delivering their decisions in kind—with no oral hearings, either in physical courtrooms or by video’.5See Online Courts and the Future of Justice and ‘The future of courts’, The Practice: Remote courts.
Where does this leave access to justice?
It is an understatement to suggest that Sir Geoffrey Vos wants to bring about rapid root-and-branch reform. Rather, his intention is to uproot all the trees in the civil justice forest and plant new species in their place. And he is not just a simple woodcutter living in a hut on the edge of the woods. He is the powerful and energetic Lord of the Forest who is likely to have the clout to achieve much of what he wants. These proposals raise many questions for those of us concerned about access to justice for all; questions that need to be considered urgently.
Is fundamental reform of the civil justice system needed? Undoubtedly, ‘Yes’. Court buildings are run down and, after the court closure programme, often geographically remote. Delays in hearing cases are too long. Costs and fees are too high. Legal aid is rarely available. Documentation is too voluminous. Existing IT systems are poor in the civil courts.
Is there a role for improved IT systems and the use of AI?
Again, undoubtedly, ‘Yes’. To misquote Aneurin Bevan, ‘people who stand in the way of progress, get run down’. The future does not lie in opposing progress. Automated systems have been introduced to replace humans in the banking, insurance, education, and employment sectors, as well as in armed conflicts. Google Translate has become a very effective tool. AI programs are capable of beating chess grandmasters. It is claimed that they have also had considerable success in predicting the outcome of cases before the European Court of Human Rights. 6Donna Lu, ‘AI learns to predict the outcomes of human rights court cases’, New Scientist, 8 August 2019, although it should be noted that the predictions are based on the judges’ statements of the facts, which may well give a clue as to which way the wind is blowing.
But can judicial reasoning, or the activity of judging, be replicated by algorithms, operating with machine learning capabilities? Probably, in time, in some cases, assuming that the algorithms are sophisticated enough. But to what extent is it acceptable to reduce the human and social element in justice?7Note the concerns of John Morison and Adam Harkens, ‘Re-engineering justice? Robot judges, computerised courts and (semi) automated legal decision-making’, Queen’s University Belfast, Legal Studies, 39(4), 618–635, 2019.
How would the public respond to life-changing events (such as possession orders evicting families with children from their homes) being determined with little or no human intervention? How could an algorithm exercise judicial discretion, such as deciding whether or not it is reasonable to make a possession order or to suspend it? And given that algorithms essentially make decisions on the basis of past experience, how would they respond to novel legal and social situations?
Will the new systems be properly funded? Will new AI programs be developed and/or managed by HMCTS or outside providers?
What Sir Geoffrey and Professor Susskind are proposing will not come cheap. Devising effective new systems is likely to cost more than maintaining the current system. The courts modernisation programme has already come in for much criticism.8See, for example, Court and tribunal reforms. Second report of session 2019, HC 190, Justice Committee, 31 October 2019.
Will the algorithms ask the correct questions? Will they be sophisticated enough to deal with all the subtleties of disputes? How do we ensure that new AI programs do not incorporate the inbuilt prejudices of those who design them?
Although it is right to suggest that in many road traffic claims speed is a factor, there may be many others. In the average county court mortgage possession claim, the important question is not generally whether there are arrears, but whether they can be paid within a reasonable period so that any possession order should be suspended. Further, AI may eliminate human prejudices and mental shortcuts in making any individual decision, but there is a risk that the preconceptions of the humans who initially devise the AI system are built into it. While it may be acceptable to outsource some of the elements of the provision of justice, it cannot be acceptable for decision-making that determines the outcome of disputes to be taken outside the justice system.9Professor Susskind, who has proposed that ‘front-ends’ should be slotted in front of conventional court services with the aim of dissolving or diverting simmering legal disagreements into online dispute resolution (‘putting a fence at the top of the cliff rather than an ambulance at the bottom’), now ‘reluctantly’ supports their provision by private-sector entities: ‘The future of courts’, The Practice: Remote courts.
It is vital that court users, lawyers and judges play a key role in devising new systems. That cannot be left to outside consultants who do not use and know the state of the justice system.
Do asynchronous hearings and the determination of some issues by AI comply with the right to a fair hearing enshrined in European Convention on Human Rights article 6 and the common law?
Perhaps, in theory, but what will happen in practice? In order to ensure effective participation in a trial in which AI plays a role, parties must be able to challenge the algorithmic score that is the basis for any decision.10See, for example, State of Kansas v John Keith Walls, 116,027, where the Court of Appeals of the State of Kansas decided that a defendant must be allowed access to the complete diagnostic Level of Service Inventory-Revised assessment on which the court relied in deciding what probation conditions to impose on him.
The EU General Data Protection Regulation, which continues to apply in the UK despite Brexit, includes a right ‘not to be subject to a decision based solely on automated processing, including profiling, which produces legal effects’ without any human intervention.11See article 22(1) of Regulation (EU) 2016/679 on the protection of natural persons with regard to the processing of personal data. That may be a reason why eBay appears to retain a human element in its digital dispute resolution process. See Jeremy Barnett and Philip Treleaven, ‘Algorithmic dispute resolution – the automation of professional dispute resolution using AI and blockchain technologies’, The Computer Journal, vol 61, issue 3, March 2018.
What of those who do not have ready access to the internet? The proportion of those who have internet access is increasing year on year. It has risen to 96 per cent of households, but only 84 per cent of the UK population can access the internet via a smart phone. That falls to 53 per cent of those who are 65 and over. That is not a reason to oppose reform, but it would be wrong to deny anyone access to the single point of online entry to civil dispute resolution. It must also be recognised that access via a computer in (say) a library does not provide equality with litigants who have online access in their own homes or offices. An alternative route of access must be provided for such people.
More importantly, what of those who are disadvantaged by poor literacy and/or linguistic skills or by mental health problems?
The Lord Chief Justice has recognised
that the new systems developed during the pandemic, including remote hearings, have caused particular problems for litigants in person. Those problems will be even greater when trying to access a single online entry point or completing online forms, especially for those who speak English as a second language. Likewise, a significant number of people, who appear to function well in society, conceal difficulties in writing or typing (eg, due to dyslexia). All those with experience of litigants in person, especially in small claims, have seen cases that, on the papers, seem hopeless, but when explained orally by litigants in person become compelling. Measures must be taken to ensure that proper services are available to assist vulnerable people and that any new systems and procedures do not breach the ‘reasonable adjustment’ provisions in the Equality Act 2010. Arguably, one of the criteria for eligibility for assistance in any reformed legal aid and advice system should be ‘special litigation needs’ (perhaps irrespective of the merits of the dispute) in the same way that children can be assessed for ‘special educational needs’. Such special litigation needs could be assessed according to language and literacy skills and mental health issues. Again, ensuring equality in any new system will not come cheap.
What will be the role of the legacy system? It should not be allowed to become second-class justice for those who are unable to enter the online funnel or who are diverted out of it.
How is value to be assessed? The implication is that ‘low-value’ cases are more likely to be resolved by a more basic process. A claim for £3,900 may appear small, but it amounts to more than the annual income of someone receiving income support. A claim for £100,000 may appear sizeable, but it amounts to a fraction of the annual income of some investment bankers. In general, the jurisdiction of the Dubai International Financial Centre Small Claims Tribunal is limited to AED 500,000 (slightly over £100,000), yet that is more than the annual profit of many small UK businesses.
It is crucially important that in moving towards new dispute resolution mechanisms, we do not lose the ‘justice’ element in civil justice. As recognised in the preface to LAG’s 2020 book, Justice Matters: essays from the pandemic
, we need ‘a larger discussion about the future, about equality and justice’. Sir Geoffrey’s proposals emphasise the need for that to happen urgently. Legal Action Group should be at the forefront of those discussions.