Commentators in other jurisdictions, me included, have expressed wonder at the procedural hurdles which US case law developments in the past six years have placed in the way of parties preparing to give discovery. We don’t criticise the need to do the job properly – to preserve that which ought to be preserved and to collect it in a manner which is capable of being defended in court – but the procedures as they have developed appear to have paid more attention to formal niceties than to the actual needs of most cases. Requirements developed to meet the worst cases seem to have been applied to every case, regardless of the degree of actual risk that important documents will be missed or destroyed.
There are signs, perhaps, that US courts are stepping back from positions which are now seen as major contributors to wasted expense, as rule-makers and judges rethink the balance between cost and utility and consider how to control (and if necessary punish) those whose conduct puts really does put justice at risk without burdening every party to every case with expensive processes which are disproportionate to the value of the case and the risk of injustice.
Some of the things said at the Georgetown Advanced eDiscovery Institute CLE in December give clues as to the thinking which is going on as to the future. I will come back to that in another articlee (or two) but it is worth reviewing where the US has got to so far and to consider how parties can minimise the risk of getting it wrong. One of the specific issues is the extent to which it is prudent for companies to collect their own documents and data.
This is one of the subjects covered by Daniel Lim, VP and General Counsel of Guidance Software, in an article called Rolling the Dice on Legal Hold and Self-Collection: Not Worth the Risk on the Corporate Counsel web site. The article gives a good summary of the present state of the US law relating to legal hold obligations for civil proceedings and to self-collection, that is, the collection of relevant documents and data by the party itself rather than by experts directed by lawyers. Although the article’s focus is on US civil proceedings, much of it applies in other jurisdictions.
Daniel Lim’s article brings us up to date on the present state of play following opinions in Howard Chin v. The Port Authority of New York & New Jersey and in National Day Laborer Org. Network v. United States Immig. & Customs Enforcement Agency. Between them, says Daniel Lim, these cases address the following points:
- Failure to issue a legal hold notice for relevant documents is not gross negligence per se, but a factor in determining whether an organization should be sanctioned for preservation failures.
- “Custodians cannot be trusted to run effective searches without providing a detailed description of those searches,” NDLON, slip op. at 36 (internal quotations omitted).
- There is a need for the testing and refinement of e-discovery search terms.
The impressions we get is that common sense is now the arbiter rather than procedural nicety. In place of the near-automatic adverse inference on the failure to issue a written litigation hold, we see consideration whether the failure to preserve, when taken with other factors, makes it appropriate to make an adverse inference – in practice, we are back to a more sensible notion (one which is recognisable to a UK audience) that actual harm is needed to found an adverse inference. If that was in fact always the case, you would not know it from some earlier cases and from the reaction to them.
In relation to keyword searches, the point is not that employee-run keyword searches are ipso facto inadequate, but that the potential for incorrect or inconsistent keywords is the greater when spread amongst possibly untrained personnel. The focus is more on the adequacy of the the description given to those making searches and on their skills.
Daniel Lim identifies one particular quotation from Judge Scheindlin’s NDLON Opinion which is worth picking out:
Searching for an answer on Google (or Westlaw or Lexis) is very different from searching for all responsive documents in the FOIA or e-discovery context.
This is fundamental for anyone, in any jurisdiction, who relies on keywords alone to find potentially disclosable documents. Keywords serve their purpose, but they take no account of misspelling, of nicknames or of all the other informal ways which those creating the documents use to describe their subject matter.
Two messages result. One is that the need for preservation, hold and collection processes are not simply a requirement under stringent FRCP procedures but make sense any jurisdiction, as does the use of appropriate tools to ensure defensible consistent results. The other is that reliance on keywords must be accompanied by an understanding of their limitations. There is good reason why experts use software which take account of context, such as clustering and predictive coding.