Judgment day for BAILII

First published by the Society for Computers and Law October 2011

A recent Guardian editorial criticised the status quo in relation to the publication of court judgments and called for more open access. In so doing BAILII came across as the villain of the piece rather than the saviour of free law which most lawyers know it to be. Nevertheless, the editorial did raise valid questions about free and open access to case law which deserve answering. I asked Sir Henry Brooke for his response to the specific questions raised and these are his responses.

Why does the MoJ release judgments through a contract with BAILII?

It is impossible to understand the present situation without knowing a bit of history. Historically, the responsibility for disseminating judgments was vested in the law reporters. Court-appointed shorthand writers would take a record of an oral judgment, and if anyone wished to publish them, they would transcribe them and were allowed to charge a fee for selling the transcripts to law reporters — the Incorporated Council of Law Reporters (for the official law reports or the Weekly Law Reports) or Butterworth’s (for the All England Reports) or an increasing number of specialised law reporters — and to anyone else who wished to purchase them. Once a judgment of either Division of the Court of Appeal was transcribed, the shorthand writers were bound to deliver a copy to the Supreme Court Library and the Bar Library. One could read the transcripts in the Bar Library, but it was impossible for anyone to copy one for use in court without the permission of a judge, and any copy could only be obtained on payment of a very large fee. Further publication or recopying was strictly prohibited. This is why the Supreme Court Library possesses copies of these transcripts which go back about 60 years (although a number of criminal appeal judgments about 20 years ago were destroyed, so that it is not a continuous set).

If a judge (or more usually his/her clerk) prepared a reserved judgment, then the text was read out in court, or, through a very slow process of change (inspired initially by Lord Donaldson), handed down in court without being read out. If a judgment was handed down a copy was made available increasingly to the immediate parties to the action or appeal, as well as being sent to the reporters without the need for transcription. This practice had the unwelcome consequence that leading sets of chambers and law firms, whose representatives often appeared in this class of litigation, would copy the transcripts internally, while nobody else had access to them until they appeared in a law report some months later. This distinction between the ‘haves’ and the ‘have nots’ was one of the unfair anomalies I was determined to destroy when I joined the Court of Appeal in 1996.

In those days, standalone PCs were slowly taking over from the typewriter: the Royal Courts of Justice was not networked until about 2003, and prior to that time dissemination of judgments was by floppy disk, accompanied by plenty of virus problems.

The next challenge was to persuade the courts to adopt a common format for their judgments, and I was the driving force behind the preparation of two Practice Statements, in 1998 and 1999, which introduced paragraph numbering and a system of neutral citation (which the House of Lords and the Privy Council then adopted). We then persevered with in-house attempts to produce a standard judgments template, but this did not have its wrinkles finally ironed out until 2001 or 2002.

It follows that the Lord Chancellor’s Department (‘LCD’: the lineal predecessor of the Ministry of Justice) never thought it had any responsibility to do anything about publishing judgments except to make contracts with the shorthand writers, permitting them to exploit their transcriber’s copyright by selling the transcripts on, while retaining (I think for no charge) a copy for the Supreme Court Library.

The Court of Appeal and the High Court deal with very many more judgments every day than courts such as the Court of Session in Scotland and the Courts in Northern Ireland. The arrangements in the House of Lords and the Judicial Committee of the Privy Council were different, but those courts, which hardly ever delivered oral judgments, were traditionally much better equipped with expert staff who would check the judgments carefully before they were issued — a cause of the very much longer delays in the availability of judgments at that level. They, too, did not publish their own judgments in those days: they left this to the Law Reporters, while retaining a copy for their records. (In 2009 BAILII ‘saved’ the entire library of Privy Council transcripts and made them all freely available for the first time after it became obvious that there would be no room for them on the move from Downing Street to the new UK Supreme Court building).

Until recently, the arrangements for tribunals were very similar. I remember a big breakthrough ten years ago, when with the help of funding from Matrix Chambers BAILII made publicly available every judgment of the Employment Appeal Tribunal, going back to the late 1970s when they were first retained in that tribunal’s library, in a way that had never happened before.

Until the advent of the Internet, neither the MoJ nor its predecessors ever recognised any duty to issue case law and make it public, a duty which certainly isn’t contained in any statute. The iniquitous Treasury-driven policy of ‘full cost recovery’ in the civil and family courts, which is not heard of in any other common-law country, has impelled them to spend as little as they can in connection with the publication of judgments, through the system of devolving responsibility to profit-making contractors which I have described.

During the judge-driven reforms of the late 1990s which I have also described, the LCD (or, more accurately, the recently created Courts Service Agency (later known as HMCS or, now, HMCTS)) started its own judgments web site, but this was always a slow and rather hit-and-miss affair, tending to publish only those leading judgments which were sent to it by the judges themselves.

The circulation list for new judgments is not restricted to BAILII, and BAILII does not have any exclusive right to receive them. A number of other organisations receive judgments in the same way as BAILII.

When practice under the two new Practice Statements had settled down, and all the judgments of the High Court and the Court of Appeal were issued in a standard format, the authorities at the Royal Courts of Justice decided about nine years ago that they wished to publish immediately on the Internet every substantive judgment of the Civil Division of the Court of Appeal and the Administrative Court, and all the judgments of the other courts at RCJ which the judges (or some other responsible authority) identified as being appropriate for publication. After considering the different options, they decided that publication by BAILII would provide a much more satisfactory and cost-effective facility than the HMCS web site for those who wished to access these judgments (for instance BAILLII provided a much faster service, better search facilities and the added bonus of automatic hyper-text linking to other judgments on the BAILII database).

It was in those circumstances that BAILII entered into contractual arrangements with HMCS whereby BAILII received the additional funding it needed to publish every such judgment, instead of publishing only those leading judgments which the judges sent to BAILII for publication. With this funding BAILII (a charitable, non-profit-making company) has been able to engage a part-time member of staff to handle the extra volume of cases the contract provides for. These arrangements have been renewed from time to time, and come up for reconsideration next spring.

Why does BAILII not allow search engines to index its judgments?

As the BAILII FAQ explain, the prohibition on indexing is because it may be necessary to remove or amend a judgment after initial release. We are not only concerned with simple corrections to the text of a judgment. Sometimes a judgment is circulated which should not have been circulated at all, for example a judgment which the court has directed, perhaps for reasons of national security, should not be published to anyone (other than the parties themselves). Inexperienced judges (or more usually their clerks) release the judgments to BAILII (and to all the other recipients of the same e-mail) when they should not have been released at all, or should not have been released without some necessary instruction being placed on the front cover of the judgment.

There is also a problem when any requisite anonymisation has not been carried out in accordance with a judge’s direction. Practice Direction 39A provides that ‘the court may order that the identity of any party or witness must not be disclosed if it considers non-disclosure necessary in order to protect the interests of that party or witness’. Sometimes it is found that a judgment which should have been anonymised has not been anonymised correctly, or has not been anonymised at all, and that it contains sensitive information from which children or witnesses — or sensitive medical details which should not have been published — can be identified. Such mistakes may be discovered for the first time years later, not just within the first month after the judgment has been given.

To minimise the damage which might be caused by errors of this kind it is very important that once the error has been spotted the judgment should be made no longer available to the public without any delay. This would not be possible if the judgment has been indexed by a search engine, as there would be a delay during which the judgment would still be accessible using the search engine.

It must also be remembered that Google is not the only search engine. There can be no assurance that every search engine will respect a request that pages should not be cached.

It has been the consistent belief of BAILII’s trustees that the provision of a searchable database of judgments on one web site is sufficient to make this source of law freely available to the public. We do not consider that taking the extra step of making the judgments searchable, using Google or other search engines, is necessary in order to achieve the primary objective of making judgments of the courts freely available to the public. And such a move forward would create the much greater risk of making judgments freely available when their publication, in whole or in part, has been expressly prohibited by a court order, so that issues of contempt of court may come into play.

In other jurisdictions the courts take on the responsibility of ensuring the requisite quality control so that judgments and transcripts leave the court after rigorous scrutiny and nobody apart from staff employed by the court sees any restricted information. In such jurisdictions indexing of judgments, caching and republication is often allowed.

Needless to say, if MoJ was to provide the volume of staffing which could scrutinise in an expert way every judgment before it is released, in the way that happens in those jurisdictions and also in the Supreme Court (with its 12 judges sitting in panels of five producing comparatively few judgments every year), the risk of error would be greatly reduced, but it is living in cloud cuckoo land to think that this would ever happen. The arrangements BAILII has made with HMCS and HMCTS (which include an indemnity for BAILII in the event that mistakes of this kind ever occur) represent a pragmatic solution to this problem. I would not like to be a publisher who republishes one of these judgments without consent and without the benefit of such an indemnity.

You will find that Google does report some 63,700 pages indexed on BAILII out of the 200,000 or so pages BAILII publishes. But Google has not indexed the content of these pages, only their titles: BAILII takes steps to prevent Google indexing the contents of judgments and we believe that we achieve this.

The most common explanation for Google displaying results for BAILII judgments is that there is a link to the BAILII judgment on some other web site: Google indexes that link, and points to the BAILII page which is the target of the link. There is nothing which we can do to prevent this, but it does not mean that BAILII has failed in its steps to prevent Google indexing the content of judgments. It is still the case that the contents of the judgment cannot be searched using an Internet search engine.

Who owns copyright in judgments?

Professor Philip Leith, who is one of BAILII’s trustees, has touched on the issues in an article here.

Over the years I have spoken to judges and leading QCs who have had very great experience in the field of copyright law, and they are of the unanimous opinion that the copyright in a judge’s words is vested in the judge. The LCD once obtained an opinion from counsel to contrary effect, and we agreed to disagree. But since both the government and the judges wish all judgments to be available freely without copyright constraints, there seemed to be nothing to be gained by litigating (or arbitrating) the issue.

What obstructs the free republication of all judgments is the private copyright rights asserted by those who say they have acquired those rights further down the line (whether as transcribers of oral judgments or as publishers of law reports). Now that I am too old ever to sit again as a retired judge I can express my personal opinion that if the matter was ever litigated in England the English courts would follow the Supreme Court of Canada and hold that the raw text of a judgment can be freely republished so long as it is not simply photocopied from a law report, because the copyright is vested in the judge. However, BAILII cannot afford to litigate the point, and it would take a very rich commercial litigant or clarifying legislation to put the matter beyond doubt. I remember once comparing the original text of Donoghue v Stevenson (from the House of Lords Library: now on the BAILII web site) with the text that appears in the official law reports. They were virtually identical apart from a very few, completely immaterial, alterations by the law reporter which were introduced to accommodate the house style of the law report.

Whether transcribers acquire copyright in transcribed judgments depends on the terms of the contract under which they were appointed. In recent times the practice of the LCD and its successors was to assert what they called ‘Crown copyright’ on the front page of every judgment, but to allow the transcriber to exploit the value of the transcribed product by charging a fee when it permitted republication. This practice had the effect of reducing the overall cost of the contract to the LCD. I was instrumental in ensuring that when the contract for the firm which transcribes every oral judgment of the Civil Division of the Court of Appeal and the Administrative Court came up for renewal about eight years ago the new contract made it plain that HMCS reserved the right to permit republication freely and without any constraints. This provision was not at that time incorporated in the contracts of the other members of the panel of approved transcribers who might be engaged to transcribe other oral judgments of the High Court on an ad hoc basis.

Why does BAILII forbid reproduction on other sites?

The judgments of many courts and tribunals in the United Kingdom and the Republic of Ireland are now freely available to the public, through the efforts of the trustees and staff of BAILII, at a single web site. (The judgments of these courts are also provided to a number of other recipients, who do not choose to make them available to the public without charge, as BAILII does.)

BAILII is a subscriber to the Montreal Declaration on Free Access to Law (2002, as amended in 2007) which states that:

Public legal information from all countries and international institutions is part of the common heritage of humanity. Maximising access to this information promotes justice and the rule of law;

Public legal information is digital common property and should be accessible to all on a non-profit basis and free of charge;

Organisations such as legal information institutes have the right to publish public legal information and the government bodies that create or control that information should provide access to it so that it can be published by other parties.

The Declaration obliges legal information institutes, like BAILII, not to impede others from obtaining public legal information from its sources and publishing it, and BAILII places no such impediment. Nor could it. All it seeks to forbid is the publication of its ‘value-added’ material, for the reasons I have given. It is for HMCTS (and the judges) to decide what republication policy best serves the public interest.

I have read of the Judgmental group who aim to build up a library of freely-redistributable judgments which are accessible via a normal Google search. This policy would be unobjectionable if they were to obtain the case law, in the form in which HM Courts Service and the judges issue it, and devote the same resources as BAILII does in converting it into a suitable format and making efforts to prevent the publication of material that must not be published. However, Judgmental’s policy seems designed to ride roughshod over all the risks which I have described previously. BAILII has committed resources to making copies of judgments available in a presentable format, and we very much regret that others should be taking these copies from our web site without our agreement and with no acknowledgement of BAILII’s efforts. We certainly regard this as being damaging to BAILII.

Why BAILII needs your support

During my 15 years of responsibility for these matters, I identified only one government minister, Michael Wills, who ever expressed any direct interest in the work which BAILII was doing, and shortly afterwards (in May 2002) he lost his job (for other reasons!). I can see not the slightest prospect of any foreseeable government making any extra funds available for the proper publication of judgments.

In April 1997, in our ground-breaking judgment in Bannister v SGB plc, Lord Saville and I (who have both been trustees of BAILII since its inception), together with Lord Justice Waller, wrote:

If this country was in the same happy position as Australia, where the administration of the law is benefiting greatly from the pioneering enterprise of the Australasian Legal Information Institute (AustLII), we would have been able to make this judgment immediately available in a very convenient electronic form to every judge and practitioner in the country without the burdensome costs that the distribution of large numbers of hard copies of the judgment will necessarily impose on public funds.’

That was the dream. It started to become a reality one unforgettable evening at Chatham House two and a half years later. Since its inception in 2000 BAILII has found ways of overcoming most of the complex problems I have described in these answers. With only two and a third members of staff and some remarkable software, it has punched far beyond its weight since it was first created. In the context of current public funding realities, the best way forward, in my view, would be to retain the current arrangements, to strengthen BAILII (through an influx of private funding) so as to enable it to build up a far larger database of pre-1996 judgments than a £250,000 project allowed us to create five years ago, and to explore ways in which very large savings of public money could be made by making BAILII the original freely accessible source of tribunal judgments in the same way as it is for the judgments of what we now have to call the Senior Courts at RCJ.

As I look back on 50 years in the law, the creation and sustenance of BAILII is one of the achievements of which I am most proud. Now that I have reached the age of statutory senility, I will be standing down as chairman of BAILII’s trustees in November this year. I would be very sad if any of the activities described in the recent article in the Guardian were to damage the service we have created. A well-informed blogger recently wrote:

Without BAILII, we would be left with expensive proprietary services such as Westlaw and LexisNexis, which are already a luxury for law firms and certainly not affordable to the person on the street. It is also no understatement to say that if BAILII ceased to exist, so would the legal blogs and other free-to-access commentary services.

Such a treasure should not be destroyed wantonly or unadvisedly when so many under-funded lawyers depend on it.

Sir Henry Brooke was Chairman of the Bar’s first Computer Committee in 1985, President of the Society for Computers and Law between 1992 and 2001, and successively judge in charge of IT and judge in charge of modernisation for seven years from 1997 onwards. He has been chairman of BAILII’s trustees since 2000, and stands down from this office in November 2011.