Search technology: an intelligent adjunct to the lawyer’s skills, not a black box

An article by H5 on the professionalization of search ties in with my recent suggestion that lawyers and search experts have parallel roles in e-Discovery and that clients, rather than the lawyers, will manage the process. The UK courts have not yet seen a challenge to a search algorithm, but it will come, and it won’t be the lawyers who will defend it

I do not generally have a project plan for what appears here. There is always a list of pending subjects, from part-drafts to ToDo list entries to vague ideas floating around in my head, but no overall schema which brings topics or people round in regular rotation. I resist both deadlines and urgent announcements – I have enough deadlines as it is for more formal papers or for conference slides, and breathless messages inviting me to get an embargoed scoop on next week’s announcement (“revolutionary, unique”) leave me cold, with their implication that I must drop everything and scurry around at a time of someone else’s choosing in order to get out an article on the same day as the rest of the world. Apart from announcements of pending conferences and near-contemporaneous reports of conferences or important cases, I take subjects as they reach the top of my mental in-box, where they simmer for a bit before I write.

This measured approach to it all sometimes means that sub-conscious connections appear between two or more subjects, so that a snippet about one thing finds a home in an article about another, or this week’s subject-matter stirs recollections of last year’s thinking. A subject which has raised its head a few times lately has been the question whether English courts consider that electronic means of culling, filtering and refining data collections are good enough relative to the lawyers’ duty of candour. That took me back to the resources on search on H5’s web site. H5 is not, of course, the only provider of search skills and technology which takes the subject seriously, nor are they the only ones who evidence their interest by active participation in TREC (the Text Retrieval Conference ) and the Sedona Conference. Their web site was, however, the first place on which I lighted when I started to investigate the points which arise when the academic and deeply complex principles of search meet the day-to-day business requirements of those who “just” want to find the right documents.

I put “just” in quotation marks because the Google generation feels entitled to assume that they will find all they need by bunging the right words into a simple search box – and so they will, of course, because that daily need to find things does not bring with it the threat of sanctions or of the other very real risks of non-compliance which come with a formal duty to produce documents. For the most part also, our Google searches are not accompanied by the very large budgets often needed to satisfy ourselves that we have found everything relevant for compliance purposes, nor by constraints on those budgets.

UK civil litigation has not yet seen any reported serious challenge to the tools or search algorithms used to filter document populations. The straight answer, therefore, to the question “do English courts consider that electronic means … are good enough” is that nobody has asked. We are, of course, seeing a rise in the number of disputes about the scope of disclosure – what sources should be included or excluded – but these are questions at a different level. What concerns me more for now, is the extent to which the fear of some failure in search technology acts as a brake when lawyers consider how to tackle electronic discovery / disclosure.

A proper article devoted to this would cover both jurisprudential analysis and a detailed explanation about search. I am less interested in either of those (for today anyway) and more interested in the business implications for lawyers and their clients which were identified in my recent post ILTA Insight 2010: lawyers risk becoming just part of the clients’ process. The suggestion raised there was that clients are beginning to see little point in engaging lawyers if that simply brings an expensive overhead on top of the technical business of finding documents. No one is suggesting that the lawyers have no role in this, but (quite apart from the fact that few of them seem to have grasped even the basics of the subject) we are talking of two distinct disciplines here which bring different, albeit overlapping, skills to the business (and I stress that word) of compliance with the obligations to courts and regulators. The difference between the two skill-sets appears most strongly when you consider whose evidence will carry most weight when questions arise about the search.

My post-ILTA musings reminded me of an article from last year on the H5 website by Nicolas Economou, H5’s Chairman and CEO, called The Professionalization of search and review in e-Discovery. It explains clearly the separate contribution which a professional search organisation can bring to the collection of evidence in the following terms:

Information retrieval experts can also harness attorneys’ subject matter knowledge and apply it in a principled, scientific manner to a range of search tasks. Do documents exist that substantiate the presence or absence of certain activities? Did a deponent know of certain facts at a given time? What other custodians might be relevant to a given topic? Does a production reveal certain behavioral patterns? How do I narrow down all documents relevant to tomorrow’s depositions to a hundred documents that I can thoroughly analyze? Information retrieval contributes principled approaches to answering these questions and, in so doing, helps litigators win their cases.

The question is not just about whether lawyers can do any of these things but whether they can do them as effectively and as cost-efficiently as can a professional search team. As Nicolas Economou says earlier in the article, search is a distinct academic field drawing on computer science, mathematics, linguistics and statistics. The question for lawyers is whether they wait for their clients to arrive for themselves at the conclusion that the functions need to be separated, or whether they anticipate this by teaming up with a provider of search services to offer a joint package to clients.

From the clients’ perspective, they have two broad options once they have decided to do their own thing in relation to search. They can either contract the exercise out or take at least part of it in-house. H5’s traditional model has been to offer their services outside the corporate firewall with a focus on litigation and litigation readiness, with the data transferred to H5 for culling, filtering and preliminary review by specialist teams before it is returned for import into the clients’ review platform of choice.
H5 has now augmented this service with a software solution called H5 Edge Classifier, which I wrote about briefly when it was launched, but which Nicolas Economou told me more about when I saw him at LegalTech in New York. H5 Edge Classifier takes the technology which H5 has developed over the years and puts it into a tool which sits behind the corporation’s firewall and on top of their existing data sources, aided by site-specific rules is required. The idea (to use their own words) is that it there “provides the critical intelligence to automate and replicate the judgements that attorneys, compliance professionals, and their review staff would make if they were able to assess each document themselves”.

The fact is that each document cannot be assessed individually within any meaningful timescale and at an acceptable cost. The perception of many lawyers at present is that any use of technology requires delegation to what they think of as a “black box” over whose selections they have no choice. What is described above is very different from the “black box” of popular imagination.The fact that the exercise (whether that be the management of information within the corporation or its review for litigation) cannot be done by traditional human means makes it easier in a sense to make a decision in favour of a technological solution. If the lawyers do not arrive at that conclusion for themselves, the clients will do it anyway – with or without the help of those who have been their lawyers hitherto.

There is a curious modesty about lawyers which is at odds with everything else one knows about them. They comprise one of the most solidly intelligent groups of people in business and yet they seem happy to shrug their shoulders and say that they do not understand technology. They are asking themselves the wrong question – it is not the technology they need to understand but the need to accept to some extent that technology derived from years of academic and commercial research has commercial application. H5’s website is a good place to start if you want to get your mind around how the principles of search apply to the business of litigation.


About Chris Dale

I have been an English solicitor since 1980. I run the e-Disclosure Information Project which collects and comments on information about electronic disclosure / eDiscovery and related subjects in the UK, the US, AsiaPac and elsewhere
This entry was posted in Court Rules, CPR, Discovery, eDisclosure, eDiscovery, Electronic disclosure, H5, Litigation, Litigation Support. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s