Bad Words, there are more than 7 of them.

Ryan Murphy,
Managing Director of Consulting Services

The Supreme Court landmark ruling in Federal Communications Commission v. Pacifica Foundation, 438 U.S. (1978) decided that there are at least seven “indecent” and patently offensive words. Those of us that are tasked with producing responsive, non-responsive or privileged discovery based on “key words” have used all 7 of those words (and many more) to describe our frustration in failing to produce expected results. There is very little (or should be very little) “mystery” as to why the margin of error is so vast. Using intuition and deductive reasoning based on a review of the body of evidence to come up with a series of words that represents the responsive, privileged and non-responsive evidence contained in the electronically stored information is like trying to decipher the 23 flavors and other ingredients of Dr. Pepper through sips; only with less potential for tooth loss.

The evolutionary path of Electronic Discovery is becoming clearer every day. What is almost certain however is that the 2007 changes to the Federal Rules of Civil Procedure did not set out to create 2 separate types of evidence – 1) that which is determined to be responsive based on the content (traditional legal discovery) and 2) that which is determined to be responsive based on the amount of time remaining in the production deadline and counsel’s ability to come up with a magic formula of words to produce evidence.

Over the past few months, there has been a HUGE discussion brewing on how to bridge Boolean key word searches and the gaps they leave in the failure to produce discovery. Clustering, Social Networking and Artificial Intelligence attempt to describe methodologies and processes by which evidence can be organized but nonetheless require manipulation through generic grouping or key words to cull. The most effective, efficient and more importantly defensible position is a process by which evidence is culled based on the core facts, those being; A) Who and B) When. Then categorizing the evidence using the metadata to provide responsive, non-responsive or privileged discovery without any human manipulation. At Equivalent DATA, we call this Evidence Assessment and this process is readily available.