eDiscovery Misconduct Uncovered
eDISCOVERY
(This is a rather long case study but it is well worth reading, especially for an attorney involved in eDiscovery. There were so many nuances to the case that one can easily write a book about it.)
Background:
We were retained as eDiscovery consultants, on a David versus Goliath case where the defendant was the Goliath. The case was filed in federal court. We were approached by a small law firm and asked to analyze whether the production tendered by the defendant was complete, accurate, and whether the discovery was undertaken and conducted with professional care and judgment, and more importantly, if there was any evidence of misconduct committed by the defendant during discovery production. Initially, we were asked to perform a cursory review of the opposing counsel’s production, which consisted of over 50,000 pages of paper documents which appeared to be totally scrambled. We initially identified missing attachments and a high number of duplicate emails and documents.
It was quickly determined that this was worse than a simple data dump. Our initial impression was that it seemed that the discovery was incomplete and may have been performed in bad faith. We could not understand how it was possible for an attorney who performed a very thorough privilege review, not to detect attachments that were missing from the emails. We felt that the overall condition of the discovery production warranted a more thorough analysis as we suspected the defendant’s council may have attempted to exploit the apparent inexperience of our client in eDiscovery procedures and withhold critical documents. Based on our initial review, we suggested to our client that a more thorough analysis was necessary to determine the true extent of the irregularities. We suggested a two-prong approach: the first would be technical in nature and the second, investigative. The technical approach would entail developing a database to potentially identify: a) email gaps, b) the quantity of missing attachments and c) duplicate emails and email chains, as well as other technical irregularities. The investigative approach would entail reading all relevant information to determine what was missing and the significance of the missing ESI.
Challenge:
We first questioned why the defendant’s production was in paper instead of an electronic format. Production in the “native format” on a CD or DVD would have cost far less than printing everything out, especially for the defendant firm, where they were likely to have extensive resources to accomplish this with minimal cost consequences. Our investigation showed that the defendant had the eDiscovery tools, programs, and employees in place to respond to such requests with minimal effort. Considering how “native format” production is required by the Federal Rules of Civil Procedure and, as per all publicly available information, the fact that the defendant likely had extensive internal eDiscovery programs and resources to respond to such requests, it made no sense for the defendant to tender the discovery production in paper. Our investigation revealed that the defendant had a dedicated eDiscovery counsel and employed full-time employees to respond to such discovery demands with utmost efficiency and minimal cost consequences. Why then, was it necessary for the defendant to go through the trouble of taking an extra step to convert ESI from native format to paper? The most obvious advantage in converting native format ESI to paper is hindering the recipient’s efforts to find relevant information. At this point, courts understand and recognize the significance of this issue. However, if the defendant’s goals are to create a burdensome situation for plaintiffs, conceal the existence of critical documents, conceal misconduct, and frustrate the plaintiff's ability, paper production will provide a venue for acts including:
- Scrambling the contents and tendering the production in a random manner to confuse the reader and render the production incomprehensible.
- Add random duplicates to increase review time and confuse plaintiff's counsel. If in native format, duplicates can be identified very rapidly as opposed to reading thousands of additional pages. What could have been accomplished with a couple of mouse clicks took us hundreds of hours to accomplish.
- Tender the emails in a non-chronological order to confuse the reader and make sorting the emails a rather time-consuming task. Emails in native format can be sorted in a chronological order with the click of a mouse. It took the plaintiff's counsel hundreds of hours to accomplish this task in paper format.
- Make it especially difficult, if not impossible, to search in an efficient manner. This goes beyond simple keyword searching as even the plaintiff had OCR’ed the production, the resulting data set would never allow the reader to manipulate the data set to follow conversations in a logical manner. It will never replace the ease in which the native emails can be isolated by topic and individuals, and sorted in a chronological order.
- Make the discovery of withholding and/or spoliation of ESI remote.
- Make it extremely difficult to determine the source of the ESI tendered.
- Make it impossible for the requester to determine when a particular document was created if there is no date on the document body.
- Genuineness of the documents cannot be determined as paper documents do not contain any metadata.
- Critical attachments in email messages may be withheld in paper format. When an email is tendered as “ordinarily maintained,” all attachments will be embedded within the email message and, without committing outright fraud and tampering with the file, they cannot be removed.
- Emails can be withheld or deleted with no (or remote) chance of detection.
- Make it extremely difficult to identify broken email chains.
- Make it extremely difficult to detect gaps in emails.
- Make it especially difficult to follow the natural flow of conversations.
- Make it especially burdensome to the requestor and exhaust his financial resources hoping that they will have to give up and settle.
- In the highly unlikely event that the misconduct is discovered, the respondent can claim an innocent mistake. None of the above listed advantages to an opponent can be accomplished without committing an outright fraud and tampering when production is made in “native format” where such acts can be detected by forensic analysis.
Response:
Because of his inexperience in eDiscovery procedures, the plaintiff's counsel failed to specifically request the ESI in native format. However, authors of the Amendments to the FRCP apparently built in safeguards to prevent the exploitation of inexperienced counsel by enacting Rule 34(b)(2)(E)(ii)("If a request does not specify a form for producing electronically stored information, a party must produce it in a form or forms in which it is ordinarily maintained or in a reasonably usable form or forms."). In addition, Joint Discovery Plan section 6 which addressed ESI, stated “..such information shall be produced in the form most readily and reasonably accessible by the producing party.” To suggest that the purest form of ESI, “email,” is readily accessible in paper to the defendant was simply absurd.
Rule 34(b)(2)(E)(i) mandates that a party must produce documents as they are kept in the usual course of business or must organize and label them to correspond to the categories. The default production format for ESI is reasonably usable or as ordinarily maintained. Rule 34(a)(1)(A) allows the discovery of any documents or electronically stored information stored in any computer storage media either directly or, if necessary, after translation by the responding party into a reasonably usable form. “Reasonably usable form” may not be interpreted as what is reasonable or convenient for the producing party. In other words, there is no “as deemed reasonable by the respondent” exception to the FRCP rule 34(b)(2)(E)(ii).
The Advisory Committee noted that if the responding party ordinarily maintains the information it is producing in a way that makes it searchable by electronic means, the information should not be produced in a form that removes or significantly degrades this feature. [Our investigation showed that the counsel for the defendant had published an excellent article on this subject and claimed to be an eDiscovery specialist.] Furthermore, the advisory committee recognized the fact that under certain circumstances, ordinarily maintained ESI, in the cases of encrypted ESI or old mainframe-based databases, etc, may not be usable to the requesting party. In those circumstances, the burden is on the producing party to translate it to a reasonably usable format so that the requestor can access it.
Accessdata Corp. v. Alste Techs. Gmbh, (D. Utah Jan. 21, 2010) Even though the party did not specify a form, the Court required the defendant to disclose ESI in native format after disclosing documents in a non-searchable PDF format.
During our analysis of the defendant’s discovery production, we did not find any “Litigation Hold” instructions which should have been sent to the custodians on this case. Analysis of the “Privilege Log” failed to indicate such instructions were issued but deemed privileged. At this point, it was our opinion that the defendant’s counsel had failed to issue proper “Litigation Hold” instructions to the defendant. It appeared that no such instructions were sent to the custodians and the IT staff who are responsible for preservation of ESI. Unlike paper discovery, failure to preserve ESI immediately can result in its loss. ESI can easily be inadvertently or intentionally deleted or altered. Loss of ESI can occur during daily operation of a computer system, typical rotation of backup tapes, and deletion of user files and emails. ESI can be inadvertently or intentionally deleted, modified, or overwritten in seconds. Based upon our analysis of the discovery produced, there was evidence of potential spoliation and/or withholding of ESI.
Pension Comm. of the Univ. of Montreal Pension Plan v. Banc of Am. Secs, No. CIV. 05-9016, at *24 (S.D.N.Y. Jan. 11, 2010)
The court has stated a party, after a duty to preserve has been attached, has to issue a written litigation hold, to identify the key players and to ensure that their electronic and paper records are preserved, to cease the deletion of email or to preserve the records of former employees that are in a party’s possession, custody, or control, and to preserve backup tapes when they are the sole source of relevant information or relate to key players.
For example, one of the principal's emails showed an interruption of the natural flow of conversations at a critical point. Furthermore, the number of emails produced from his mailbox appeared to be incomplete. Even though he received numerous emails on this matter, not a single "sent email" was produced by this individual. For someone who certified under perjury that he was very familiar with this case (he was actually the person in charge of the entire operation) to send no emails for a 31 month period was highly irregular.
Furthermore, the defendant has claimed in a letter to the plaintiffs’ counsel that Bates D000001 – D031758 were collected from custodians (which are to this date unidentified) and Bates D031759 – D051332 originated from case files in paper form. If we assumed that this statement was correct and the litigation hold was properly implemented, every email or other ESI that are contained in case files should have been produced from the custodian’s mail boxes and file server home directories as well. This was simply not the case based upon the discovery tendered. There were emails that were produced from the case file which we could not find within the custodians mailbox productions. Therefore, it was clear that the litigation hold was never implemented and the custodians were free to delete any ESI which they may have deemed damaging. We have shown that they in fact did delete such emails. Furthermore, Information Technology personnel who were responsible for suspending the rotation of the archival tapes were not aware that the tape backups containing email and other ESI that belong to the custodians should have been removed from rotation. The significance of this act is that the complete unaltered versions of email stores before such deletions took place were probably not preserved and were overwritten by newer versions that did not contain the deleted emails. This would cause spoliation of critical ESI to a point that it is no longer available. We identified numerous instances where emails existed in the “paper” category but not in the electronic category. Based on the defendant’s counsel's written assertions, we could only infer that the defendant may have been withholding responsive ESI and/or spoliation had occurred.
We also discovered production was not certified by an attorney. Rule 26(g)(1) requires an attorney to sign all disclosures, discovery requests, responses, and objections. By signing, an attorney or other party certifies the accuracy and completeness of the production to the best of that person’s knowledge, information, and belief formed after a reasonable inquiry. Considering the overall condition of the discovery tendered, this was not surprising.
The Supreme Court has confirmed that a court may impose an appropriate sanction on an attorney for the harm caused by the failure to abide by Rule 26(g). Chambers v. NASCO, Inc., 501 U.S. 32, 51, 111 S. Ct. 2123, 115 L. Ed. 2d 27 (1991)
Based on our email gap analysis we saw a five month gap in emails by all involved at a critical point. When the email chains were followed, they came to an abrupt end right before the significant event (cause of the lawsuit) occurred. When the emails resumed again, the conversations concerned how to deal with the plaintiff and how the defendant could get out of it. During the time in question, it appears that the true nature of the event was discovered. However, the defendant, even during the litigation, was attempting to conceal the true nature and significance of the events that took place during the five month period; a timespan which involved a significant environmental disaster.
Our investigators, while reading most of the production during review, had identified key events, individuals, and government organizations that were involved in the case. We consequently uncovered what had happened, who was involved, and what the defendant was doing to avoid the responsibly of cleaning efforts.
Very little ESI other than emails was produced. There were numerous emails referring to other ESI, yet none were produced. Using our “Timeline Analysis,” we identified 68 emails where documents were sighted but not provided in production.
Understanding the expansive nature of the definition of Electronically Stored Information, the Advisory Committee note states: Because of the wide variety of computer systems in use and rapid changes in technology, there is no limit or precise definition of ESI. All electronically stored information is discoverable. Kleiner v. Burns, No. 00-2160, 2000 WL 1909470, at *4 (D. Kan. Dec. 15, 2000)
As stated by the advisory committee, ESI includes, but is not limited to: word processing documents, spreadsheets, databases, presentations, voice mail messages, back-up voice mail files, email messages and files, backup email files, deleted emails, data files, program files, backup and archival tapes, etc. Yet the defendant did not produce a single piece of ESI other than what appeared to be benign emails.
Actually, the production was scrambled so badly, it had an unintended consequence of confusing the defendant and their counsel. They failed to recognize the significance of three emails. If read alone the emails appeared to be benign, however, when they are placed in a chronological order with other emails from other individuals, they simply solved the case in favor of the plaintiff and showed that the defendant was acting in a calculated way to harm the plaintiff. We uncovered the fraud they attempted to perpetrate on the plaintiff – a fraud which would have shocked the conscience of any decent human being.
We identified 90 emails with 166 missing attachments.
We found and demonstrated, on numerous occasions, the natural flow of conversations were interrupted.
In the document, “Initial Disclosures Pursuant to Rule 26(a)(1),” provided by opposing counsel, several key individuals were identified as likely to have discoverable information. However, we have not seen a list which identified and confirmed the true custodians from whom the ESI tendered was actually collected from.
There was no way to determine how many custodians the defendant had collected from. Unless the production is tendered as required by law as “ordinarily maintained,” there is no way of knowing if they even collected from all custodians. The email can originate from the custodian's “Outbox,” as well as the recipient’s “Inbox.” In this case we had no way of identifying what came from where.
According to FRCP 34, the party producing documents must organize and label them to correspond to the categories in the request. Additionally, the opposing counsel did not provide details on the origins of the documents produced (i.e. custodian and source).
We had serious reservations about the ethics of the defendant's counsel. We coached our client on the relevant portions of FRCP. At that point he was well equipped to deal with all the issues. However, we advised him to play dumb and pretend that he really does not understand it. Moreover, we advised our client to initiate contact with the defendant's counsel (who is an eDiscovery specialist) to see how far she would go in her attempt to exploit his inexperience.
During one of the phone conversations we advised the plaintiff's counsel to ask the defendant’s counsel as to how this discovery was performed and by whom. The defendant's counsel was extremely evasive in terms of who performed the collection and would not provide details. Her only response was “we will never tell you that.” Also, during the same phone conference she told the plaintiff's counsel that paper format was considered ‘readily accessible’ and production in native form was simply not required by the law. Since she was someone who claimed to be knowledgeable enough to sit in eDiscovery panels [and regularly boasted about working in that very courthouse for nine years where this matter was pending] we had no option but to conclude that the defendant's counsel was acting in bad faith by trying to exploit her perception that the plaintiff’s counsel was inexperienced with regard to eDiscovery issues, by misstating the law.
We advised our client to keep pushing for "native format" production, which would reveal the discovery violations committed and force the defendant to turn over the critical ESI they were withholding.
When this issue was brought before the court, the defendant's counsel also made a statement to the court regarding her firm’s inability to view emails in the format in which the defendant maintains them, ie., Lotus Notes. The defendant's counsel stated, “Production is in paper format, we received it from the defendant in paper because it is in “Lotus Notes” and we can’t read it.” First and foremost, Lotus Notes documents can easily be converted to the MS Outlook format by numerous software utilities that cost a few hundred dollars at the most. This process of document conversion is a standard operating procedure for anyone involved in eDiscovery. Furthermore, to even suggest that her firm could not read emails from one of their largest clients was absolutely absurd.
We investigated to see if what she had stated was true. We found job postings that her firm was seeking paralegals who were experienced in Summation and/or Concordance, popular ESI review tools. It appeared that the firm was utilizing at least one of these review tools. Summation can read Lotus Notes emails with the simple use of the available add-on and Concordance can process Lotus Notes after conversion by using readily available tools from Microsoft.
Accordingly, we found her explanation to the court as to why this production was tendered in paper to be simply incomprehensible. Furthermore, if her explanation to the court was deemed reasonable, the consequences of such a decision would allow conduct of this sort from parties in the future, based on a mere claim that they cannot read a client's ESI. Parties would then be free to produce ESI in non-searchable paper instead of an electronic format. Our legal research failed to reveal the counsel's inability to access a client's ESI in native format as being an exception to the requirements of Rule 34(b)(2)(E).
During a later conference, as she was confronted with the fact that her firm can indeed read Lotus Notes emails, she changed her position once again. This time she took the position that what she meant was that she could not read it on her own workstation. Considering her first reasoning of why this production was presented in paper to the plaintiff's counsel [during the phone conference wherein she stated that the production was in paper because “paper is considered readily accessible and the native format production is simply not the law”] it seemed that her reasoning changed to whatever was convenient at the time.
We provided the plaintiff’s counsel with the list of deficiencies found, so it be forwarded to the defense counsel for corrective action. When it became available, we reviewed the defendant’s response to the deficiency letter. Our analysis of the response basically confirmed our findings that the discovery tendered by the defendant was inaccurate, incomplete, and most likely was performed in bad faith.
The defendant’s response to the “deficiency letter,” in essence, has shown that this discovery was inaccurate, incomplete, and was not performed with due care. Their position of not addressing any one of the issues in a satisfactory manner showed that there was an element of bad faith. The defendant responded with some ridiculous explanations to certain items and did not even bother to address some of the issues that were clearly un-addressable.
We demonstrated that the defendant, based upon the above stated facts, had apparently failed to fulfill its obligation and responsibilities. Our investigation also revealed that acting in bad faith and disobeying rules during discovery were not new to the defendant. On another case in California, the defendant was caught intentionally destroying critical ESI and was sanctioned by the court. In that case, the defendant had deleted and destroyed critical ESI and was sanctioned by the court (adverse inference and monetary sanctions) as a result of its actions during discovery. It seemed clear that the sanctions received were not severe enough to deter one of the world’s largest corporation from engaging in similar conduct in the future. We provided the judge’s decision on that case to the plaintiff's counsel.
We demonstrated with clear and convincing evidence that the defendant and their counsel had acted in bad faith and their conduct was egregious and intentional. It appeared that the defendant took calculated risk as it did not believe the plaintiff would have the means to uncover the misconduct set forth herein.
Results:
We filed an expert opinion of our findings with the court. Motion for sanctions was filed. Our investigative efforts also solved the case in the plaintiff's favor and equipped them with all the smoking gun evidence they needed to win the case.
Back to Case Studies