This is an edited version of a presentation at the University of Ulster Legal Innovation Centre yesterday and is a summary of some of the positive and negative effects of technology on legal services.The point of this presentation is not to take the title as raising a binary choice: it is to explore the ways in which technology can increase or decrease access to justice, perhaps even at the same time. Since the pressure of technological advance is unstoppable, we should also be looking for a grid by which we can understand what is going on around us.
We should begin with an acknowledgement of the power of the legal tech movement as a whole. There were 1300 people at the British Legal IT Forum in London last week; 1700 at the American Bar Association TechShow in Chicago the week before and over 2000 in trendy London Shoreditch for the Legal Geek conference last autumn. One estimate of the current USA spend on legal tech is $1.5bn on software alone. A further stab at the value of ‘the global LegalTech market is … $15.9 billion, and growing.’ The Stanford University CodeX Techindex, last Friday listed 1138 legal start ups ‘changing the way legal is done’. The Legal Innovation Centre is itself a manifestation of this activity. There are similar centres around the world, particularly in the US. Many major law firms, including Allen and Overy (associated with the centre) have various versions of in-house tech innovation spaces like its Fuse.
By contrast, activity in developing the use of technology in access to justice has been much less – although a number of legal tech promoters, among them LegalGeek, began with hackathons designed to look at access to justice issues in order to rouse the widest range of interest. But three major problems confront those interested in developing access to justice tech. Poor people don’t by definition have any money so they cant pay for services in the same way as commercial users. Even reducing prices to vet low levels may not be enough for those totally without resources. And, traditionally, work in the access to justice sector is undertaken by a disparate and largely uncoordinated mix of law centres, advice agencies and legal aid practitioners who work on a shoestring; the area is now hit by government austerity cuts; and, perhaps as a result, little has been recorded in a form which would provide workable data on which technology could get a grip. And finally we have to deal with issues often called the digital divide. I will come back to that.
There are have been a number of waves of potential external funding which might have impacted – and might still do so – on access to justice and technology. We began with the capital behind the wave of firms, like Co-operative Legal Services, that sought to take advantage of new alternative business strategies and launch a raft of DIY web-based cheap products such as CLS’s packages on divorce. This initiative was aimed at taking a substantial bite out of the traditional High Street practice sector. However, CLS and the others encountered more consumer resistance than anticipated; they were not, perhaps, as well organised as they might have been; and the once standard-bearing CLS has been forced radically to downsize and rebrand as much more of an adjunct to the Co-op’s funeral business.
Following private capital were Legal Aid administrations looking to technology as some form of magic bullet which would simultaneously allow service extension and cost restriction. The Dutch led here with their Rechtwijzer product which promised more interactivity in advice through guided pathways and some online conciliation and mediation. It failed to meet its financial obligations and has been wound up – though a successor organisation is seeking to emerge. Less ambitiously, the US Legal Services Corporation (which puts federal money into civil legal services projects) ploughs ahead with its Technical Initiative Grants programme of around $4m a year which does impact on US developments but not through one sole product. It places great faith and a substantial amount of money into its forthcoming Legal Navigator pilots – selfhelp referral and intake statewide sites being piloted in Alaska and Hawaii.
A range of non-Justice government departments came next in what could be seen as a third wave of funding. New South Wales funded the Nadia project (to which I will return) as part of the launch of a disability benefit. And, in England and Wales, various pots of money have been made available by our Department of Business for artificial intelligence initiatives – including a £7000,000 grant to the Solicitors Regulation Authority for an AI challenge fund to be soon launched.
Finally, we now have the development of online dispute resolution in the courts where Ministries of Justice might seek to integrate small claims cases with a degree of self help. This is the model of the Civil Resolution Tribunal in British Columbia and it was argued in the preliminary reports here by Lord Briggs as a necessary part of the online courts and tribunals programme in England and Wales. It seems to have been quietly – silently – dropped as too expensive and fiddlesome. It is not clear that the massive court modernisation programme in England and Wales will have much direct connection with information and advice services – a weakness all the greater with the widespread demise of legal aid.
There is a point that I wish to make from this account of institutional funders. I have referred to US, England and Wales, The Netherlands and Canada. Technology knows no boundaries in the way of substantive law. So, to get an idea of what is going on we need to look internationally. Similarly, in my view, to evaluate any technology product we need to submit it to international comparison. The key evaluation question for any project is – is this the best in class globally?
The first problem with technology is digital exclusion – the numbers of people who are excluded from using digital delivery because of limitations of various skills – be they cultural, technological, linguistic, cognitive or other. Frankly, there is no hard evidence that I would rely upon to say how large in the excluded population. It might, in any event, be shrinking all the time. I could produce fancy reasoning but it is really little more than a guess – I would say a quarter of what we might regard as the financially eligible population (old legal aid limits). That seems to me to be large enough to demand the retention of traditional forms of delivery but large enough to merit exploring the use of technology. I am interested in more discussion of this.
There is a sub-issue here. MyLawBC is a guided pathway advice programme in British Columbia based on the Rechtwijzer. Useage rose when they ceased to collect personal data on users like their Post or ZipCode. People did not trust the privacy of the system – something which needs to be borne in mind by developers and presents real problems for funders who demand verifiable information about users. They may need to rethink. A report released yesterday by Transform Justice accuses HMCTS of suppressing research that indicated court users’ unease with digital delivery. And a US study reported today from the centre for state courts says the same.
A group of further interconnected issues also impede technology in the service of access to justice. Governments with bad news about legal aid funding or its equivalent have started to throw out small sums for technology almost as consolation prizes. That was part of the Ministry of Justice’s response to the LASPO review. That links with a finding which, no doubt, government finds inconvenient – all the research suggests that digital works best as a supplement or extension of traditional services not as a substitute. And a final point is the hype that has led to an overselling of products which then prove not to work – many of them asserted to run on versions of AI. Nadia was a robot that I thought would transform the advice world: she spoke with the seductive cadences of Cate Blanchett; Cate was going to answer all your questions. Well, she did not. She was using IBM Watson and it just wasnt powerful enough.The Australian government pulled the plug.
So, that are some of the downsides. What of the up? There is no silver bullet or magic app – one application of technology which is THE answer. We need to look at the use of technology for different functions. The future use of technology in A2j is likely to consist more in small incremental steps rather than characterised by a Big Bang.
The first positive is pretty basic: the advantage of upgrading basic office productivity tools. The Legal Education Foundation is funding a major upgrade to law centres’ basic digital tools. The US LSC has help to fund its own case management system – LegalServer. Most solicitors firms have some form of digital CRM or case management spine these days. Coming soon will be add ons such as one which allow pre-initial interview questionnaires to be sent to clients so that the cost of the initial interview is kept down.
The second positive is also fairly basic. A number of organisations in different jurisdictions have used video for remote service of advice provision. University House in Bethnal Green, East London has an arrangement with Falmouth CAB and Truro Court to provide remote services. An Ontario legal clinic does much the same. Similar arrangements happen in the US. The interesting implication of this form of delivery is that, potentially if taken to the extreme, it would threaten the basic model of a neighbourhood legal clinic developed in the US in the 1960s and then exported to the English-speaking world. You could have a national clinic on what used, at one time, to be called ‘the hub and spoke’ model – based in one place and serving a whole jurisdiction through outposts shared with other organisations.
The third area is the improvement of website information. The Government Digital Service has a creditable track record here of setting standards for legibility and accessibility. The Citizens Advice Service in England and Wales has put considerable work into its website which it is clearly turning into a digital first resource – including ‘heat maps’ which can actually monitor what part of page people are actually reading.
The Citizens Advice site – like most information sites – remains, however, stubbornly linear – based on the format of a traditional book. By contrast, MyLawBC is an example of a site which seeks to make the process of finding relevant information interactive. So, the user follows guided pathways in much the same way as you would to book an airline ticket – you dont just get the schedule of BA flights around the world. You limit the parameters of the search first.
Interactivity of provision must be the next frontier. Relate has plans to introduce guided pathway information and there are other plans around the world. To work well, guided pathways do take considerable preparation. Other agencies are poised to follow. The difficulty is that the information has to be seriously reworked if it is to move from simple linear presentation to something more directed to the ensurer’s need. That is expensive in time and resources.
Other jurisdictions have had to pay more attention to referral than have we have. Traditionally, a referral to a legal aid lawyer was all you needed and they were spread nationally. Where services have been more patchy and variable, there is more of a need for a more targeted referral. A number of pro bono bodies like Justice Connect in Australia are pulling together sites which will match a user’s problem with a service that might be appropriate. This will also be part of the Legal Navigator package.
Unsurprisingly perhaps, artificial intelligence is working its tentacles into the field. An interesting US project which is a collaboration between Suffolk and Stanford law schools seeks to use natural language processing and machine learning. It seeks to categorise different legal problems; then take a series of questions asked in ordinary language by users of Reddit; and match the two. That would be the first step of an approach which would turn information provision on its head. Prospectively, you would ask a question in ordinary language; have it translated into a machine readable version; facilitate machine learning within the system; and then have an answer in ordinary language. This is being developed in a collaboration between Suffolk and Stanford Law Schools in the context of the ‘legal navigator’ project.
Less ambitious than AI has been the advance of the chatbot – of which Nadia was but the most advanced. Considerable publicity has been obtained for a range of bots designed by Joshua Browder of which the best known is Do Not Pay – a way of challenging parking tickets. A chatbot can vary in sophistication from a very simple programme (which Do Not Pay is) to something much more complicated and with the chatbot presenting itself with a personality. They clearly have fantastic potential. I am not sure that is yet really manifest in many of the examples around the world. They tend to the simplistic. But there must be potential here.
Fourth is interactive self help document assembly. In the USA, the LSC has funded a project called A2J author which is: ‘a cloud-based software tool that delivers greater access to justice for self-represented litigants by enabling non-technical authors from the courts, clerk’s offices, legal services organizations, and law schools to rapidly build and implement user friendly web-based document assembly projects. This allows organisations to use a basic template to draw up a simple guided interview that generally takes a user through half a dozen steps to a courthouse where their objective is achieved – e.g. to issue proceedings of some kind.’ One of its products, a motion to modify child support of spousal maintenance in Minnesota won recognition as the ‘best automated form’ in 2017 from the Self Represented Litigants Network.’ That reflects a move toward the provision of self-assembly documentation.
The UK has followed into the self-assembly field with caution. CourtNav, however, is very similar to projects fuelled by A2J author – without the visuals. It is an online tool developed by a specialist Citizens Advice office in the Royal Courts of Justice. The system has now been taken up by the whole Citizens Advice and can be accessed from local offices. It relies on pro bono lawyers to check the self-assembled documents.
There has also been some exploration in England and Wales of the possibility of interactive self- assisted letters rather than court interventions eg for a disability payment known as PIP (Personal Independence Payments) where one app will help the users with a letter of claim and another provider will produce a similarly interactive request for a mandatory reconsideration. A user can be guided to complete a standard letter with information that is relevant to the matter in hand – and given ‘just in time’ resources to help them understand what is required. Rather more may be made of these programmes in promotion than in actual use and a challenge in this area is to get innovative provision actually used and tested.
Fifth and finally, we have a category which we might call serendipity. This covers the use of technology in a range of disparate ways which do not really hang together yet show an innovative approach. That, for me, pretty well covers where we are at generally. Let me give three examples. A family practitioner in Brighton has an IBM Watson subscription. He has input all his cost data into it and he uses it to shape and limit the risk on fixed fee quotations. Just fix NYC has an app which has allowed 1200 families across New York City to take action on housing disrepair. It structures the capture of evidence, the taking of photos, the making of a case and a referral to an assisting organisation. It won an international award from the Hague Institute for Innovation of Law last year. Project Callisto seeks to address sexual harassment on campus universities by using the recording possibilities of the net. A university subscribes and a person who considers themselves harassed puts up a report which remains uncoded and unread until the name of the alleged perpetrator reappears in another report. A counsellor then raises with both parties whether they wish to take the matter further and, to assist them to do so, the system has preserved their initial complaint made without any contact between them.
In conclusion, technology cannot compensate for the widespread loss of individual legal services which we have experienced in England and Wales. Technology does, however, give us the opportunity to raise our game and increase the value of our services. We will see as we progress how many people are excluded from digital and we will have to make provision for a considerable number. That doesn’t mean we shouldn’t try out what we can do. And be inspired by what people around the globe have attempted.