The problem created by Rule 34(b)(2)(E)(i) and (ii)
Federal Rule of Civil Procedure 34(b)(2)(E) provides:
Producing the Documents or Electronically Stored Information. Unless otherwise stipulated or ordered by the court, these procedures apply to producing documents or electronically stored information:
(i) 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 in the request;
(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; and
(iii) A party need not produce the same electronically stored information in more than one form.
Rule 34(b)(2)(E)(i) addresses production of “documents” and states that they must be produced “as they are kept in the usual course of business” or “organiz[ed] and label[ed] . . . to correspond to the categories in the request [for production].” Rule 34(b)(2)(E)(ii) addresses production of “electronically stored information” and states that it must be produced “in a form or forms in which it is ordinarily maintained” or “in a reasonably usable form or forms.”
The question that has divided the courts is whether electronically stored information is subject to both Rule 34(b)(2)(E)(i) and Rule 34(b)(2)(E)(ii) or only to Rule 34(b)(2)(E)(ii). If both (E)(i) and (E)(ii) apply to ESI, then ESI that is produced, under (E)(ii), in a reasonably usable form, but not in the form in which it is ordinarily maintained, would seem not to be produced, under (E)(i), as it is kept in the usual course of business — which means that the ESI, under (E)(i), must be organized and labeled to correspond to the categories in the requests for production.
In practice, responding parties almost never want to produce documents organized and labeled to correspond to the categories in the requests for production, because this shifts the expense of reviewing and categorizing responsive documents from the requesting party to the responding party. But neither do responding parties want to produce ESI, under (E)(i), in the form in which it is ordinarily maintained, since such native-format ESI is malleable (e.g., a Word document subject to editing); difficult to stamp, index, and refer to (the Bates-ing process) in litigation; and difficult to use in depositions and at trial. They would much rather produce Bates-stamped images with accompanying metadata, which would qualify as a “reasonably usable form” of the ESI under (E)(ii).
The practical question is thus whether responding parties are put to a choice between producing natives or producing images + metadata, but being forced to categorize the images and metadata according to the requests for production.
District courts are divided
Some courts hold that “Rule 34(b)(2)(E)(i) governs hard copy documents, and (E)(ii) governs ESI, with no overlap between.” Anderson Living Trust v. WPX Energy Production, LLC, 2014 WL 930869 at *13 (D.N.M. 2014); see also, e.g., National Jewish Health v. WebMD Health Services Group, Inc., 2014 WL 2118585 at *6 (D. Colo. 2014). These courts hold that one is never required to produce ESI categorized and labeled according to the requests for production under Rule 34(b)(2)(E)(i) because that rule simply never applies to ESI. Instead, with respect to ESI, the only question is whether it is produced as it is kept or in a reasonably usable form under Rule 34(b)(2)(E)(ii). The idea is that the responding party should never be required to categorize and label ESI because “parties requesting ESI [are] able to organize it themselves — in their own way, to their own satisfactory level of thoroughness, and at their own expense — through the use of text-searching technologies like filtering, grouping, and ordering.” Anderson Living Trust, 2014 WL 930869 at *13.
Other courts hold that both (E)(i) and (E)(ii) apply to ESI: “Rule 34(b)(2)(E)(i), which provides that parties ‘must produce’ documents as they are kept in the usual course of business or ‘must label’ the documents to correspond to the categories in the request, applies to productions of ESI as well as traditional hard copy documents.” City of Colton v. American Promotional Events, Inc., 277 F.R.D. 578, 584 (C.D. Cal. 2011); see also, e.g., MGP Ingredients, Inc. v. Mars, Inc., 2007 WL 3010343 at *4, *4 n.12 (D. Kan. 2007). The idea here is that (E)(ii) addresses format, which is a problem that simply doesn’t arise in the context of paper documents. That is why (E)(ii) refers only to ESI and not to documents generally. Because (E)(ii) addresses format, while (E)(i) addresses organization (ordinary course v. according to the requests), and because format is a question that arises only in the context of ESI, (E)(ii) and (E)(i) can apply cumulatively to ESI even though only (E)(ii) mentions ESI specifically. And this reading is to be preferred because (E)(i) is as sensible a rule for ESI as it is for paper documents.
(E)(i) should not apply to ESI
The better reading of Rule 34(b)(2)(E) is the one in Anderson Living Trust and National Jewish Health under which Rule 34(b)(2)(E)(i) does not apply to ESI.
As a matter of sound lawyering, the textual argument is dispositive: Rule 34 throughout treats “documents” and “electronically stored information” separately (e.g., Rule 34(a)(1)(A), allowing a party to request production of “documents or electronically stored information,” and Rule 34(b)(2)(E), providing “procedures [for] producing documents or electronically stored information”); thus, because Rule 34(b)(2)(E)(i) refers only to “documents” and not to “electronically stored information,” it applies only to documents and not to ESI. The rule uses both “documents” and “electronically stored information” when it means to include both of these; the rule’s failure to use “electronically stored information” in (E)(i) means that (E)(i) does not apply to ESI.
This reading is also to be preferred from a policy perspective, for the reasons stated in Anderson Living Trust. Paper documents are difficult to search, sort, and categorize; producing them in an unordered mess imposes an artificial burden on the requesting party to sort through the mess and find evidence. Thus, it makes sense for (E)(i) to require the responding party to produce the documents either in the way in which they are kept (i.e., the documents will be no less organized when the requesting party receives them than they were in the responding party’s possession) or, if the responding party is going to reorganize them anyway, then in the organization suggested by the requesting party’s requests.
Unlike paper documents, ESI is easily searched, sorted, and reviewed using tools like Disco. The ability to manipulate ESI using software obviates the organizational problem that (E)(i) addresses: the recipient of ESI, using modern legal technology, can create whatever organization of the ESI it wants regardless of the organization of the ESI coming in. This is why it shouldn’t matter in a production of ESI whether the files appear in the same order they appeared on the hard drives from which they were collected; rather, what matters is that the files are accompanied by metadata indicating where they came from (“original file path” metadata in Disco), so that if this is the order in which it makes sense to review them, the files can be sorted that way in the ediscovery tool of the requesting party. And, likewise, if the order that makes sense is instead by custodian or instead according to the prevalence of certain keywords or instead according to some more sophisticated relevance ranking, these orders, too, can be created on the fly in modern ediscovery platforms.
What is important about ESI productions is not how they are organized (the (E)(i) problem), but rather the format of the ESI (the (E)(ii) problem) — and, in particular, that the format include as much information about the ESI (metadata) as possible to enable the requesting party to create whatever organization of the ESI it wishes.
Here, then, is an example of how understanding legal technology helps make sense of a rule.