Keywords and Cooking: an eDiscovery Parallel

There are not many things which we can all agree on in eDiscovery / eDisclosure. You would probably get near-unanimity as to what the rules say in any jurisdiction (at least from those who have read them), and most people would probably be able to agree as to what their bare words mean. We would probably not fight over the fact that a judgment or opinion had been delivered or by whom, and its broad implications may not provoke too much dissent, whatever argument there may be about its application to other cases.

It is probable that most would agree, if pressed on the subject, that there is an obligation to give discovery of electronic documents on the same basis (at least as to broad definition) as paper documents, though not everyone actually does so or does it properly (“properly” in this context meaning something different from saying “Here’s everything”). Finally, everyone will nod sagely when it is asserted that disclosure/discovery is an expensive component in much litigation, although the reaction of many is to blame the rules, the lawyers, the court, and the service providers – pretty well everyone, in fact, except the people who created all this crap and made no provision for culling it down.

In addition, there are significant differences between the players. Some handle only very big litigation whilst others have many small cases, and it is not value alone which determines how much discovery there is. Lawyers and judges include some of the brightest people on the planet, but there are others who are, er, less well-endowed in that department, to judge by some of the court decisions one sees. Some adapt readily to technology but others do not, and it is not age or intelligence alone which determine this. There are countless other sub-divisions and classifications which distinguish one person from another.

One final point before I turn to the comment which started me writing this post. If you seek to address a market made up of such disparate elements as I have described, then you have to pick a section of it as your primary audience and have that section in mind as you write or speak. My target audience is not the converted nor those who have no interest in eDisclosure / eDiscovery with no motive to find out about it. I do not clutter my writing with exceptions and qualifications which recognise every standpoint.

One would be failing in one’s duty, in the circumstances described above, if everyone agreed with everything one said. Only statements of the most useless blandness would find universal acceptance. My post Singing from the Same Hymn Sheet drew this comment from Mike Taylor of i-Lit:

You write as if no-one uses predictive coding or concept search or intelligent review technologies.

The reality is that in litigation, arbitration and regulatory investigations the general use of technology beyond dedupe, keyword, date and custodian filters is lower than in internal investigations but that is not to say they are non existent.

The most backward exercises are regulatory investigations which verge on the luddite but that tone is set by the regulator not the lawyer or client (in fact I have a client at the moment who is running  far more sophisticated internal investigation alongside the regulators investigation as they are so keen to unearth and extinguish wrongdoing in their organisation and they are not convinced the instructions from the regulator will achieve that).

Litigators are a cautious bunch and it is not surprising that they are slow adopters of new technology, their bloody-minded nature also means that being told to do something is more likely to make them put the brakes on even more!

I regularly use a blended approach to this but to rule out keyword searching as a load of rubbish is, I think, missing the point. A well-reasearched and thoroughly tested keyword list can assist parties in reducing the amount of data for review in an agreed and defensible manner which everyone understands. It has been hard enough getting litigators away from the “we must look at everything” way of thinking. I think it is a case of taking baby steps and litigators will use more sophisticated technologies when they are ready, and not before. In the meantime it is down to professionals within the field to make sure that clumsy errors in keyword list construction are avoided and that the limitations of the best keyword list are pointed out to the parties using them.

I would say that to suggest that these technologies are not being used is incorrect and that to suggest that keyword lists are useless is also incorrect, it is not as black and white as that.

There is not much to disagree with in this sensible comment, particularly in relation to those points which derive from Mike’s own experience. It is worth addressing some of the points where we appear to differ, on the assumption that others react in the same way to what I said (or am assumed to have meant) in my Hymn Sheet post..

I do not believe that I have anywhere “rule[d] out keyword searching as a load of rubbish” when they are used as part of a search strategy. The key word in this observation about keywords is “part”. What is condemned is the assumption that you can consider the job done because you have thrown a list of keywords at the data or even when you have agreed them with your opponents. I won’t recite it all here, but have a look at what US Magistrate Judge Peck said at Carmel about the unthinking use of keywords for a position which reflects my own. You might also look at my article published at the weekend about the Emerging Technologies Panel at ILTA, and specifically at the three paragraphs starting about half-way down beginning “Howard Sklar amplified on this…” for examples of the “blended approach” to which Mike Taylor refers.

There may well be regulators who, as Mike suggests, are still inadequately equipped and trained to deal proportionately with the data of those whom they investigate. The reverse is also true – many of the regulators, particularly the financial regulators, well know what to ask for and how a properly-run company ought to be able to manage and produce its data. The fact that it cannot is unlikely to impress the regulator; if their lawyers are similarly ill-equipped then the investigation is likely to be more prolonged and expensive than it needs to be and, perhaps, with a less appealing outcome.

If I do indeed “write as if no-one uses predictive coding or concept search or intelligent review technologies” (and I do not think that I do) it is because there is no point in preaching to the choir, as Judge Peck often puts it. Quite apart from any other source of information, the software companies specialising in this kind of technology seem to be doing well. Somebody must be buying it.

There is a further point, one which Mike does not make but which it is worth addressing. No one is suggesting that any of this is easy. My wife sometimes offers a parallel: what is difficult about producing a three course meal for six? You just buy the right quantities of all the right ingredients, prepare them properly and cook them for the length of time appropriate for each component, in the right container, by the right method, at the right heat and for the right length of time, so that they all end up simultaneously and at the right temperature on the plates. What is difficult about that?

EDisclosure / eDiscovery has parallels for all that, with the added complication that your opponent keeps trying to trip you up, your client is simultaneously demanding, unhelpful and mean, and the court (equating to the diners in this scenario) notoriously fussy. One could no doubt extend the parallel and talk of extremes varying from opening a can of beans for dinner through to setting fire to the kitchen and poisoning the guests, as outcomes which equate to some of the eDiscovery stories one hears about.

The parallel which matters is this: eDiscovery, like proper cooking, is a specialist skill. If you hold yourself out as able to do it then that is equivalent to inviting people round to dinner. If you can’t stand the heat, as they say, then get out of the kitchen: hire in the skills (employ a cook) or outsource the specialist tasks (book a restaurant). As with restaurants, there is a wide range of specialists to choose from, as the logos down the right-hand side of this page show.

To revert to my opening paragraphs whilst clinging on to my culinary parallel, litigation lawyers and cooks bring different levels of skill to very varied projects. Keywords, like flour and potatoes perhaps, are a key ingredient of exercises big and small. You don’t generally serve flour and potatoes on their own, and (as with keywords) you can foul up the use of even such simple ingredients if you don’t do it properly. To say that many people misuse keywords for searching is not the same as condemning their use.


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 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 )

Google photo

You are commenting using your Google 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