Stout v. Illinois Farmers Insurance

150 F.R.D. 594 | S.D. Ind. | 1993

ENTRY ON PLAINTIFF’S MOTION TO COMPEL and DEFENDANT’S MOTION FOR PROTECTIVE ORDER.

FOSTER, United States Magistrate Judge.

In this breach of contract action, the plaintiff (“Mr. Stout”) alleges that the defendant (“Farmers”), his insurer, wrongfully and in bad faith refused to pay his fire loss claim under the terms of his homeowners insurance policy. Mr. Stout purchased an insurance policy from Farmers providing a year’s coverage for his house in French Lick, Indiana starting on March 12,1991. On May 19, 1991, a fire occurred at the house which damaged real and personal property. Mr. Stout reported the fire to Farmers the next day and Farmers began an investigation into the cause and origin of the fire and Mr. Stout’s claims history and financial condition. On November 14, 1991, Farmers formally notified Mr. Stout that it was denying his claim. Mr. Stout filed this suit in February 1992.¹

Now before the Court are Mr. Stout’s motion for an order pursuant to Rule 37(a), Federal Rules of Civil Procedure, to compel Farmers to comply with his Rule 34 request to produce “all documents within its investigatory file that were produced or obtained between March 19, 1991 and November 14, 1991”, (Plaintiffs Motion to Compel Production of Documents, p. I),² and Farmers’ motion for a protective order pursuant to Rule 26(c). Farmers produced some documents from this time period but withheld others. Its grounds for protection are that the documents are either immune from discovery under Rule 26(b)(3) as trial preparation materials (“work product”) or privileged as attorney-client communications under Indiana law, Rule 26(b)(1), and Rule 501, Federal Rules of Evidence. Mr. Stout moved to compel production on the ground that the documents withheld by Farmers do not qualify as work product. He did not make a showing of substantial need for the documents or undue hardship in obtaining their substantial equivalent and he offered no argument on Farmers’ attorney-client privilege contentions.³ Farmers did not raise any relevancy objections to production of the documents.⁴ Thus, the questions presented for our review are solely whether the withheld documents qualify as work product or are privileged as attorney-client communications.

Farmers filed and served the required index of documents, see Harper v. Auto-Owners Insurance Company, 138 F.R.D. 655, 664 (S.D.Ind.1991), although its explanation of the legal and factual grounds for protection was not as complete or individualized to each separate document as required. Farmers also submitted the documents for in camera inspection.

Work Product Standard.

Application of the work product rule to insurer investigative documents is one of the most difficult and often-litigated discovery issues because it is the very nature of an insurer’s business to investigate events which, either directly, or as a consequence of the insurers’ claims decisions, often result in litigation.⁵ Because of this confluence of business and litigation elements, it is difficult to determine with precision whether a particular investigative document was created to meet the insurer’s litigation or non-litigation needs, and, thus, whether it is qualifiedly immune from discovery under the work product rule. We recently addressed the scope and application of Rule 26(b)(3) in the insurance environment in Harper v. Auto-Owners Insurance Company, 138 F.R.D. 655 (S.D.Ind.1991), and so refer to that opinion for a full treatment of the governing standard. We will concentrate in this ease on the specific issues raised by the parties.

In Harper, we explained that Rule 26(b)(3)’s standard for determining whether a document was “prepared in anticipation of litigation or for trial” has two components, both of which must be satisfied before protection will be extended: first, the document must have been created for the purpose of preparing for litigation, the so-called “causation” or “purpose” prong, and, second, the insurer must have had a justifiable anticipation of litigation when the document was prepared, the “anticipation” prong. Harper, 138 F.R.D. at 661; see Allendate Mutual Insurance Co. v. Bull Data Systems, Inc., 145 F.R.D. 84, 87 (N.D.Ill.1992); International Surplus Lines Insurance Co. v. Willis Corroon Corp., No. 91-C-6057, Memorandum Order, 1992 WL 345051, 1992 U.S.Dist. LEXIS 17332 (N.D.Ill., November 10, 1992). We criticized the tendency of the “shifting focus” approach to work product to confuse the two prongs, usually resulting in the subsumption of the causation-purpose inquiry into the reasonable anticipation inquiry. We held that the Rule required independent inquiries into causation and anticipation: after litigation is reasonably anticipated or even underway, any document produced to evaluate an insured’s claim (or for any other non-litigation purpose) is not work product. Rather than being seen as a linear inquiry, therefore, work product analysis should be approached through independent, parallel lines of inquiry. See Harper, 138 F.R.D. at 662-63; APL, 91 F.R.D. at 21.

A, Causation-Purpose.

In making its argument on causation, Farmers apparently misconstrues a couple of related concepts. One of its reasons for contending that the withheld documents are work product is that they were not produced in the routine or ordinary course of its business because the procedures it followed in evaluating Mr. Stout’s claim varied significantly from its normal claims evaluation procedures for ordinary fire loss claims. (Farmers’ Brief, p. 17). It further contends that work product protection is warranted because these additional procedures would not have been undertaken but for its anticipation of litigation looming over its pending claims decision. (M).⁶ Farmers misinterprets the meaning of the causation component of the standard. The Rule requires causation in the sense of the purpose or motivation for the creation of documents—ie., the intended use to which the documents were to be put— not causation in the sense of a “but for” sequence of events or influences. For documents to qualify as work product, they must have been created for the purpose of preparing for litigation.⁷ Many documents are produced “because of’ anticipated litigation, in the sense that they would not have been created but for the prospect of litigation, but they were not created to prepare for that litigation. Obvious examples are requests and bills for litigation services. Another example is an investigation which is undertaken for non-litigation purposes when the incentive to pursue that investigation is enhanced by the prospect of litigation, e.g., businesses which conduct periodic safety and maintenance inspections in part to control their liability risk. If an insurer anticipates litigation over a pending claims decision and undertakes extra-ordinary investigations for the purpose of ensuring a correct decision, the resulting documentation is not work product because the insurer generated the reports to use in evaluating the claim in fulfillment of its contractual obligations. It’s interest in making a correct claims decision may have been increased by a desire thereby to avoid liability, or even litigation, but it still conducted the extra investigations in order to arrive at a claims decision and not to construct a litigation strategy.⁸ Steps that insurers, other businesses, or individuals take to ensure that their current or future conduct conforms to the law or contractual obligations cannot be held to constitute steps in preparation for litigation merely because there is a substantial prospect of litigation if another party believes they have failed. Therefore, documents generated when an insurer undertakes additional efforts to evaluate a claim because of the prospect of litigation, are not work product despite the fact that those efforts would not have been undertaken “but for” the prospect of litigation. The Rule’s scope is narrow: it looks only to the purpose or the actual use for which documents were generated.

Secondly, Farmers appears to follow a common mistake in misapplying the “routine and ordinary course of business” language prevalent in the case law. As we observed in Harper, the question under the Rule is not whether the creation of subject documents was more or less routine or ordinary in an insurer’s business, but whether the documents were produced for litigation or non-litigation applications. Harper, 138 F.R.D. at 668.⁹ Although consideration of a producer’s business or routine may be relevant under the Rule as evidence of the purpose or use for which the producer created the documents, the standard against which documents are measured for work product protection is not the nature of a document producer’s business or the history of its internal procedures, but whether they were produced for litigation or non-litigation purposes. Thus, to the extent that Farmers argues that the withheld papers are work product because the investigations which they document were unusual, non-routine, or extraordinary, their argument misses the mark.¹⁰ See APL, 91 F.R.D. at 19, 20; Atlanta Coca-Cola, 61 F.R.D. at 118; Binks Manufacturing Company v. National Presto Industries, Inc., 709 F.2d 1109 (7th Cir.1983) (documents not work product in part because they were prepared to assist settlement negotiations, not the routine or business of either party).

Although the asserter of work product immunity bears the initial burden of demonstrating that its documents qualify as work product, we, as other courts do, employ practical presumptions. As we stated in Harper, documents an insurer produces before arriving at a claims decision which consist of factual or other information reasonably related to an evaluation of its insured’s claim, are presumed to have been produced or used for claims evaluation and not for litigation preparation. Harper, 138 F.R.D. at 663-64. Insurers may overcome that presumption by presenting specific evidentiary proof of objective facts to the contrary. Id. Such a showing might consist of evidence that the documents were segregated from the insurer’s claims evaluation process (structurally or by circumstance), see, e.g., id.; McFadden v. Norton Co., 118 F.R.D. 625 (D.Neb.1988)¹¹, or that the evidence already gathered beforehand by the insurer put it beyond reasonable doubt that the claim would be denied, so that it would be unreasonable to presume that the insurer ordered creation of the documents to evaluate the claim, Dunn v. State Farm Fire & Casualty Co., 122 F.R.D. 507 (N.D.Miss.1988) (insured’s confession); Fine v. Bellefonte Underwriters Insurance Co., 91 F.R.D. 420, 423 (S.D.N.Y.1981). See Harper, 138 F.R.D. at 664 n. 4.

B. Anticipation.

In Harper we observed the high standard the Seventh Circuit provided in Binks. for a party’s anticipation of litigation: the producer of the documents must show “at the very least [that] some articulable claim, likely to lead to litigation” had arisen, or “objective facts establishing an identifiable resolve to litigate” before the documents were produced. Binks, 709 F.2d at 1119. Binks involved a contract dispute between two corporations over the performance of a machine one had purchased from the other. During the long and, at times, acrimonious course of attempting to resolve their differences, an attorney for the buyer investigated the situation and recorded his findings and impressions. After the parties’ attempts at resolution failed and suit was filed, the buyer resisted the seller’s discovery request for its attorney’s memoranda claiming that they constituted work product. The Court held, in part, that because the parties were actively pursuing settlement negotiations when the memoranda were prepared and neither party threatened suit, there was an insufficient anticipation of litigation to invoke the work product rule. Despite the facts that litigation was undoubtedly anticipated by the parties should their negotiations fail, the parties to such litigation were certain, and the cause of action and issues were similarly certain, the Court found that there was only a possibility of litigation, thus emphasizing the importance of the parties’ “identifiable resolve to litigate” over the requirement of the presence of an “articulable claim”.¹² See also In re Special September 1978 Grand Jury, 640 F.2d 49, 64-65 (7th Cir.1980); Scott Paper Co. v. Ceilcote Co., Inc., 103 F.R.D. 591 (D.Maine 1984). An insurer's investigation and evaluation of an insured’s claim in fulfillment of its contractual duties is comparable, in this respect, to the parties’ resolution negotiations in Binks: despite the certain identification of parties and claims in foreseeable litigation between an insurer and its insured, it similarly can be presumed, absent more, that neither party acquires a firm resolve to litigate before the triggering event is reached—here, the claims decision, in Binks, the break down of negotiations. Binks thus supports the presumption applied in Harper that litigation is not sufficiently anticipated before the insurer arrives at a claims decision. See Harper, 138 F.R.D. at 663—64; APL, 91 F.R.D. at 21. This presumption can be overcome by a showing, for example, that litigation was already commenced or that the insured expressly and specifically indicated the imminent filing of suit. See Fine, 91 F.R.D. at 423.¹³

C. Concurrent purposes.

The question of the proper standard governing documents produced for concurrent litigation and non-litigation purposes, discussed in dicta in Harper, 138 F.R.D. at 661—62 n. 2, 669 n. 12, is presented in this case because Farmers asserts that some of the withheld documents were produced both to evaluate Mr. Stout’s claim and to prepare for anticipated litigation. By far, the most prevalent articulation of the standard in the ease law is that dual-purposed documents will constitute work product if the “primary motivating purpose” behind their creation was to aid in preparation for litigation.¹⁴ However, we were unable to discover, either in the case law or the commentaries, any rationale for the “primary purpose” standard. Rather, the case law mostly relies on citations to a few early cases—e.g., Gulf Oil, Janicker, and Davis—which merely state the proposition. (See note 14).

The Seventh Circuit has not clearly declared its position on this issue. In In re Special September 1978 Grand Jury, 640 F.2d 49 (7th Cir.1980), a state grand jury investigating political contributions by a trade association served a subpoena on the association’s law firm directing it to produce documents relating to its preparation of the association’s reports on contributions to the state’s Board of Elections. The Seventh Circuit found that the firm created the documents both to prepare the association’s filings with the Board of Elections, a non-litigation purpose, and to prepare for any criminal proceedings which might have resulted from the grand jury’s investigations, a litigation purpose. Without attempting to determine which purpose was stronger or “primary”, the Court held that the documents constituted work product, thus suggesting the standard that any amount of litigation motivation behind the creation of a document qualifies it as work product. Id., at 61-62; sec also Mission National Insurance Company v. Lilly, 112 F.R.D. 160, 164 (D.Minn.1986). The Court recognized a different approach later in Binks, supra. In defining the work product standard in that case, the Court quoted a passage it found persuasive from Janicker, supra. While this passage dealt principally with the fundamental difference between an in-house investigative report prepared in the ordinary course of business and one prepared in anticipation of litigation, it concluded with the statement that “‘[w]hile litigation need not be imminent, the primary motivating purpose behind the creation of the document or investigative report must be to aid in possible future litigation.’” Binks, 709 F.2d at 1119 (quoting Janicker, 94 F.R.D. at 650). In analyzing the attorney’s memoranda before it, however, the Court did not inquire into, and made no conclusion regarding, the primary motivating purpose behind the memoranda’s creation. It merely held that the producer had failed to carry its burden of proof. The Court made no statement of its own on the standard for dual-purposed documents, and it did not cite or refer to its earlier treatment in 1978 Grand Jury. Id. at 1120. Its decision appeared to turn principally, if not solely, on its finding that there was only a possibility of litigation at the time of the memoranda’s creation. Id.¹⁵ Because the standard for dual-purposed documents apparently did not figure in its decision, therefore, the “primary motivating purpose” quote from Janicker may be treated as dicta, but dicta that at least calls into question any implied standard from 1978 Grand Jury. We must, therefore, determine what standard should govern work product immunity for documents created for concurrent litigation and non-litigation uses.

We are guided in this inquiry by a few fundamental principles. Chief among them is the requirement that privileges and immunities which hinder the Rules’ objective of liberal discovery should be strictly construed consistent with their language and purposes. See Eureka Financial Corp. v. Hartford Accident & Indemnity Co., 136 F.R.D. 179 (E.D.Cal.1991); Zullig v. Kansas City Power & Light Co., Civ. A. No. 87-2342, Memorandum and Order, 1989 WL 7901 (D.Kan. Jan. 17, 1989); Pete Rinaldi’s Fast Foods v. Great American Insurance Companies, 123 F.R.D. 198, 201 (M.D.N.C.1988); Hartford Fire Insurance Co. v. Garvey, 109 F.R.D. 323, 327 (N.D.Cal.1985); United States v. 22.80 Acres of Land, 107 F.R.D. 20, 22 (N.D.Cal.1985); Lundy v. Interfirst Corp., 105 F.R.D. 499, 504 (D.D.C.1985). Next, case law in the Seventh Circuit counsels against reading work product immunity broadly. See 1978 Grand Jury, 640 F.2d at 65; Allendate, 145 F.R.D. at 87; Harper, 138 F.R.D. at 661; Henderson, 131 F.R.D. at 570.¹⁶ Thus, any doubts, ambiguities, uncertainties, or silences as to the scope of the work product Rule should be resolved in favor of the Rules’ overall policy of liberal discovery.

We are also informed by the policies behind the work product rule. The Supreme Court stated in the landmark case of Hickman v. Taylor, 329 U.S. 495, 67 S.Ct. 385, 91 L.Ed. 451 (1947), that the work product immunity promotes the “orderly prosecution and defense of legal claims” by allowing attorneys privacy to prepare their cases without “undue and needless interference” and “unnecessary intrusions” by opposing counsel and by encouraging recording of information by parties, 329 U.S. at 510-11, 67 S.Ct. at 393-94.¹⁷ Later explanations of the reasons for protecting work product have similarly emphasized the unfair advantage gained by an opponent who freely obtains materials developed or acquired as a result of a party’s own skills and labor.¹⁸ At least with respect to work product documents not containing impressions or opinions, the Rule has been interpreted to be designed more protect the proprietary interests of parties in their strategies, tactics, and research (which indirectly encourages full development of each party’s case) than to improve the quality of presentations to the court through separate, independent investigations. See National Union Fire Insurance Co. of Pittsburgh, Pa. v. Murray Sheet Metal Co., Inc., 967 F.2d 980, 984-85 (4th Cir.1992):

Even when documents and tangible things are prepared in anticipation of litigation or for trial, to the extent they do not involve the mental impressions, conclusions, opinions, or legal theories concerning the litigation, they may be discoverable on a showing of'“substantial need.” The documents falling within this classification are clothed with a qualified immunity that is grounded on a proprietary aspect of the work. The immunity for this class of document is little more than an “anti-freeloader” rule designed to prohibit one adverse party from riding to court on the enterprise of the other.

To the extent, therefore, that the existence of documents or tangible things is not due to a party’s litigation preparation efforts, the policies supporting work product do not apply because release of those documents would not disclose specific litigation plans.

Although “primary motivating purpose” is the majority standard, it is not the only standard in the case law, and even among those cases articulating this standard, there are some apparent differences in interpretation. “Primary motivating purpose” appears to provide a standard which finds in favor of the purpose which has the greater weight (a “fifty percent plus one” analysis), however some cases suggest a different understanding. In Kelly v. City of San Jose, 114 F.R.D. 653 (N.D.Cal.1987), the court said:

The work product doctrine does not apply to information collected or communications made in the normal course of business. It applies only to material generated primarily for use in litigation, material that would not have been generated but for the pendency or imminence of litigation. Since police departments are under an affirmative duty, in the normal course of serving their public function, to generate the kind of information at issue here, the policies that inspire the work product doctrine are wholly inapplicable.

Kelly, 114 F.R.D. at 659 (citations omitted, first emphasis added). To similar effect are Allendate, 145 F.R.D. at 87 (quoting the primary motivating purpose standard from Sinks (quoting Janicker) but stating that “to apply the doctrine to those materials that insurers, or other parties, would produce in the ordinary course of business even if litigation were not imminent is an unwarranted extension of the doctrine”) and International Insurance Co. v. Certain Underwriters at Lloyd’s London, No. 88-C-9838, Memorandum Opinion and Order, 1990 WL 205461, *6 (N.D.Ill., Nov. 27, 1990) (after articulating the primary purpose standard, the court found that the documents were created in the ordinary course of business because “[t]he documents sought would have been created whether litigation had been anticipated or not”). Whether a company’s litigation interest in undertaking an investigation was greater than its business or other non-litigation interest at a particular time is not the same inquiry as whether the investigation would have been undertaken regardless of its litigation usefulness. Another standard was articulated in Soeder v. General Dynamics Corp., 90 F.R.D. 253 (D.Nev.1980). The Court there held that documents resulting from an aircraft manufacturer’s investigation into a crash of one of its planes did not qualify as work product because, in addition to preparing for litigation, the manufacturer had an “equally reasonable desire ... to protect future pilots and passengers of its aircraft, to guard against adverse publicity in connection with such aircraft crashes, and to promote its own economic interests by improving its prospect for future contracts for the production of said aircraft—” Id. at 255. See also McMahon v. Eastern Steamship Lines, Inc., 129 F.R.D. 197, 198-99 (S.D.Fla.1989). This “equally reasonable desire” standard most likely provides a lower threshold than the “regardless of litigation purpose” test because a person may have a reasonable interest in the results of a particular investigation for business as well as litigation purposes, but not a sufficient interest to motivate it to undertake the non-litigation purpose in the absence of its litigation uses. Finally, another approach requires a court to determine the “driving force” behind the creation of a document, though there may be little difference between this and the “primary purpose” standard. See, e.g., National Union, 967 F.2d at 984.

We believe that the appropriate standard for evaluating dual-purposed documents to be one in line with Kelly, Allendate, and International Insurance. If a document or thing would have been created for non-litigation uses regardless of its intended use in litigation preparation, it should not be accorded work product protection. Because the document would have been created for non-litigation reasons anyway, disclosure of the information therein would not disadvantage its creator, or advantage his opponent, by revealing the creator’s legal strategy or tactics; thus, the document’s release in discovery would not contravene the policies supporting the work product rule. There is no proprietary interest to protect, no threat to the orderly and fair administration of litigation, and no loss of incentive for attorneys or parties to creatively develop their cases. Since neither the language nor purposes of the work product Rule would be contradicted by this standard, we are constrained to so narrowly construe it.

Because the Rule restricts work product protection only to those documents and things produced for the specific purpose of litigation preparation—but not all documents produced “because of’ anticipated litigation in a “but for” sense—so this standard for concurrent documents focuses on the intended or actual uses of the documents regardless of role the prospect of litigation played in sparking the creator’s non-litigation interest. As discussed above, this principle accounts for the insurer who, because of the prospect or existence of litigation, decides to pursue additional or extra-ordinary investigations of a claim in order to assure a correct and defensible claims decision. See APL, 91 F.R.D. at 20, 21. On the other hand, documents produced for concurrent purposes where the intended business uses were not a sufficient motivation for the documents’ creation would be protected as work product.¹⁹

Because it is reasonable to assume that a reasonable insurer would make a claims decision only after it possessed the minimum amount of information it required to make that decision, we will presume that documents which were produced by an insurer for concurrent purposes before making a claims decision would have been produced regardless of litigation purposes and therefore do not constitute work product. As with the other presumptions discussed above, this presumption may be overcome by specific evidentiary proof of objective facts to the contrary.

This standard best comports with the objectives of the work product Rule and meets the desire for practicality and predictability. We believe that it would be an easier, more focussed, and more predictable approach to determine whether a particular investigation or document would have been ordered by an insurer to evaluate an insured’s claim in the absence of litigation preparation based on reasonable and actual insurer practice and the nature and chronology of evidence bearing on the claim. Attempting to determine the proportions of an insurer’s motivations behind the creation of a document at a particular point in time is not only extremely difficult, but has led to widely divergent and often seemingly arbitrary results. As we observed, the level of an insurer’s interest in a particular document for litigation purposes is difficult to measure because it varies greatly depending on many factors which change with time. By contrast, determining whether an insurer would have produced a document for claims evaluation purposes should be an easier process since the criteria for evaluating a claim should be more stable and the chronology of evidence obtained on those criteria readily determinable—even given the fact that the insurer’s criteria and desired quantum of evidence may vary with the prospect of litigation, the amount of the claim, and other factors. More reasonable and practical guideposts for insurers and plaintiffs are provided by the presumptions based on the claims decision and the available rebuttal showings of segregation and existing sufficient information, for example.

Application of Work Product Standard.

A. Argument.

Farmers claims work product protection for certain documents produced by it on and after May 24, 1991 (five days after the fire) because, at that point, it had evidence indicating that the fire was incendiary and that Mr. Stout was involved in its setting, and thus began to investigate his claim principally to prepare for litigation. The facts on which Farmers relies are set out in the margin.²⁰ Farmers’ interpretation of these facts as grounds for work product protection follows:

Therefore, Farmers reasonably anticipated that, based upon the information it had learned by May 24, 1991, there was a likelihood that the claim may be denied if the objective facts continued to hold true and, therefore, that there was an obvious likelihood that litigation would ensue with the insured. Indeed, Mr. Stout had already threatened to sue Farmers. At this point, the nature and primary focus of the investigation changed. For example, after the claim was referred to Harrison & Moberly on May 24, 1991, representatives of Farmers regularly consulted with John Beeman and Jeffrey Oberlies, attorneys for Farmers, with regard to the scope, details, and findings of the investigation of Plaintiffs claim in order to make a good faith decision regarding the claim of Richard L. Stout, Jr.

Further, investigation of a fire loss claim in the ordinary course of business where litigation is not anticipated does not include requesting documents regarding the insured’s detailed financial history, detailed information of the insured’s prior loss history, and detailed verification of information received by Farmers from the insured.

Farmers’s investigation of a fire loss claim that is done in the routine and ordinary course of business requires the insured to submit to Farmers Insurance Exchange a Sworn Statement In Proof Of Loss and a Personal Property Inventory Worksheet. Such claim is then adjusted and a draft is issued to the insured. Only when the cause and origin of the loss is determined to be arson and evidence exists that connects the insured to the cause of the loss does the insurer engage in an extensive and detailed investigation for the primary purpose of litigating the claim with the insured, if necessary. Evaluating and processing claims wherein it is suspected, based upon objective facts, that arson is the cause of the loss and the insured is connected to such arson vary substantially from the procedures for evaluating and processing claims wherein such facts do not exist.

The more intensive and specialized investigation prior to the denial of the claim was undertaken primarily to prepare for litigation in the event the additional investigation confirmed the initial information supporting suspicions of Mr. Stout’s involvement.

Based upon the objective facts known by Farmers on May 24, 1991, Farmers anticipated that litigation was likely to occur with regard to this claim, although a decision had not been made on this claim. While not all of the documents were obtained, prepared and/or generated by Farmers after May 24, 1991, to aid in possible future litigation, the only documents withheld by Farmers were documents whose primary motivating purpose was to aid Farmers in future litigation regarding the claim.

Where the insurance company’s investigation indicates that the fire was an incendiary or intentionally-set fire, along with other indications which tend to indicate the possibility of the insured being responsible for the loss, there is a strong likelihood that litigation will occur and the insurance company is entitled to protect the fruits of its investigation. Although the insurance company may not have completed its investigation, if the further investigation indicates that the insured is responsible for the loss, and therefore that the claim should be denied, courts in Indiana consistently hold that such an investigation is in anticipation of litigation and protected.

* * # * * *

... the investigation shifted from mere claim evaluation to anticipation of litigation on May 24,1991. By this date, the investigation of Richard Stout’s claim had resulted in “detailed suspicions from both in-house and independent investigators” and the investigation switched from the adjustment of an ordinary claim to one in which most of the documents generated would most likely be used in eventual litigation, due to the suspicious nature of the fire, Richard Stout’s financial condition before the loss, Mr. Stout’s prior fire losses, and the amounts involved.

(Farmers’ Brief, pp. 15-17, 21 (citations omitted, original emphases)).

B. Analysis.

Distilled down, Farmers claims that the following facts led it to create the withheld documents in preparation for litigation:

1. Mr. Stout had a previous fire loss in January, 1991 under similar circumstances: (1) “almost identical” or “very similar” burn patterns, though neither the nature nor the significance of the similar burn patterns is further explained, (2) both fires occurred on Sundays, and (3) both occurred between 3:00 a.m. and 5:00 a.m. Surprisingly, we are not told in Farmers’ brief whether the January 1991 fire was determined to be arson or whether the house was insured at the time by Farmers or another company (we discovered from reviewing the withheld documents in camera that Farmers did insure and pay on the previous loss). If the earlier fire was determined not to be arson and a claim was paid, the probative value of these unexplained similarities would be minimal absent further explanation. Mr. Stout further stated that he had an earlier fire loss about twelve to fifteen years earlier.

2. Mr. Stout’s financial condition was unknown and he was evasive and aggressively secretive about his finances: (1) during a recorded statement, he said he could not remember how much income he reported on his last tax return or whether he had filed the return one month earlier; (2) during the recorded statement, he refused to answer questions concerning any loans or debts; (3) Mr. Stout requested an increase in advanced monies but refused again to answer questions about loans and debts, and (4) Mr. Stout demanded that a new claims representative be assigned because the current one was asking questions about his finances.

3. Circumstances surrounding the fire: (1) Mr. Stout, who has the only keys to the house, locked it up when he left the previous day at 3:00-4:00 p.m. and he wasn’t aware of any forced entries after the fire; (2) the house was listed for sale at the time of the fire; and (3) Mr. Stout obtained insurance for the house a little over two months before the fire.

4. An anonymous phone call- indicated that Mr. Stout was “involved” with having “various properties” burned.

5. Mr. Stout said he would retain an attorney and sue Farmers if it didn’t assign a new claims representative to his ease.

6. In an oral preliminary report five days after the fire, an outside investigating firm told Farmers that it had a “strong suspicion” of arson based on its scene inspection.²¹

Our analysis of the significance of these facts is made easier by Farmers’ own interpretations. In its argument, Farmers admits that, on May 24, 1991, (1) it had insufficient information to make a claims decision,²² (2) it hadn’t concluded that the fire was arson,²³ (3) it wasn’t certain that Mr. Stout was the arson-setter,²⁴ and (4) that it believed that there was only a likelihood of litigation over the claim, which likelihood was further dependent on the results of its later investigations and claims decision.²⁵ We have to agree with Farmers that the facts and circumstances it knew on May 24, 1991 were insufficient to warrant a reasonable insurer in denying Mr. Stout’s claim. We therefore find that it would be unreasonable to assume that Farmers did not undertake or consider the later investigations in order to arrive at a good faith claims decision—which Farmers states it hired Harrison & Moberly on May 24 to assist it in accomplishing. Farmers does not assert that the later investigations were not used or desired to evaluate Mr. Stout’s claim. Farmers has therefore failed to rebut the presumption that it ordered and considered later documents in arriving at its November 14, 1991 claims decision.²⁶ The only report of an investigation into the fire itself, Wolf Technical Service’s oral report on May 24, 1991, indicated only a “strong suspicion” of arson—no conclusions or findings. As Harper held, even if Farmers had formally and reasonably concluded that the fire was arson, this fact alone would have been insufficient to warrant a reasonable insurer in concluding that Mr. Stout was involved in setting the fire. Harper, 138 F.R.D. at 667. The facts on which Farmers relies to suggest Mr. Stout’s involvement in the fire—similarities to an earlier fire loss, evasiveness about debts and loans, presence at the house, etc.— while they support Farmers’ early suspicions, were insufficient to reasonably support final conclusions; they did, however, indicate the need for further investigation, as Farmers itself admits. Regarding anticipation, as discussed above, it is clear in this Circuit that a mere likelihood, even an obvious or substantial likelihood, of litigation is insufficient to satisfy the requirement of the Rule. The required identifiable resolve to litigate is entirely absent here. Farmers anticipated litigation with Mr. Stout only “if the objective facts held true” and “in the event the additional investigation confirmed the initial information”. (Farmers’ Brief, pp. 15, 17 (emphases added)). That much of an uncertainty is too much under Binks. There was nothing from Mr. Stout to indicate that litigation was imminent or that he was resolved to litigate. Farmers’ attempt to rely on his “threat” of litigation is mere reaching. For what appears here, Mr. Stout could have been upset because he believed that Farmers’ adjuster was asking for irrelevant or private information. That he was motivated by an attempt to hide incriminating information is certainly a possibility, but nothing in the facts known by May 24, 1991 tend to indicate that alternative was any more probable than other “innocent” alternatives. At any rate, Mr. Stout’s threat to sue was directed solely at Farmers’ choice of claims adjusters, not its claims decision. If this suit were about Farmers’ bad faith or improper choice of adjusters rather than its claims decision, perhaps the threat would be informative; as it is, however, it is unpersuasive.

Farmers’ argument that the investigations it undertook in this case after May 24, 1991 are work product because they were beyond its routine and ordinary course of business in evaluating fire loss claims must be rejected. First, as discussed supra, such a claim misunderstands the standard as differentiating between routine and extra-ordinary activities, rather than between litigation and non-litigation activities. Second, at any rate, the environment for determining Farmers’ routine and ordinary course of business is its suspected arson claims.²⁷ Its argument and Branch Claims Manager’s affidavit reveal that undertaking further investigations in those cases is Farmers’ routine and ordinary course of business. See Harper, 138 F.R.D. at 668.

Farmers argues that even if its investigations after May 24,1991 were undertaken for the purpose of evaluating Mr. Stout’s claim, they still constitute work product because they were primarily motivated by a desire to prepare for litigation. In addition to finding an insufficient anticipation of litigation above, we easily find that Farmers would have undertaken investigations after May 24,1991 to evaluate Mr. Stout’s claims regardless of its intention to use them to prepare for litigation. Farmers admits that it had only suspicions on May 24, 1991 and that the later investigations were undertaken to either confirm or debunk those suspicions—and the facts known at the time do not justify any more than reasonable suspicions.

Farmers’ contentions that the withheld documents it created or obtained between May 24, 1991 and November 14, 1991 are protected as work product are rejected. The specific document rulings are set forth below.

Our conclusion that the documents withheld by Farmers do not qualify as work product renders moot its plea that we protect against disclosure of the mental impressions or opinions of its attorneys or agents. 8 C. Wright and A, Miller, Federal Practice and Procedure, § 2026, p. 129 (1993 Supp.) (“If documents are not trial preparation material, as defined in Rule 26(b)(3), they are not entitled to any special protection from discovery under that rule even though they contain mental impressions, conclusions, and legal theories.”).

Attorney Client Privilege.

Farmers also moves for a protective order covering certain documents it contends are privileged from discovery as attorney-client communications. As Farmers relies in part on the standard set forth in Harper, and Mr. Stout does not contest it, we repeat that standard here:

A party asserting the attorney-client privilege must show: (1) the asserted holder of the privilege is or sought to become a client; (2) the person to whom the communication was made (a) is a member of the bar of a court, or his subordinate and (b) in connection with this communication is acting as a lawyer; (3) the communication relates to a fact of which the attorney was informed (a) by his client (b) without the presence of strangers (c) for the purpose of securing primarily either (i) an opinion on law or (ii) legal services or (iii) assistance in some legal proceeding, and not (d) for the purpose of committing a crime or tort; and (4) the privilege has been (a) claimed and (b) not waived by the client.

Diversified Industries, Inc. v. Meredith, 572 F.2d 596, 601-02 (8th Cir.1977).... To the extent that [counsel] acted as a claims adjuster, claims process supervisor, or claim investigation monitor, and not as a legal advisor, the attorney-client privilege would not apply. See Mission National, 112 F.R.D. at 163.

Harper, 138 F.R.D. at 671. We add that if a document was pre-existing or otherwise not produced for the purpose of obtaining or communicating legal advice, it does not become privileged merely because it was passed on to, or through, an attorney in the course of representation. Because our determination on the privileged nature of the documents depends on review, the specific rulings and explanations are set forth below.

Document Rulings.

The documents are identified by number as they are listed in Farmers’ Brief, pp. 3-9, and, in parentheses, as they are labelled on the copies submitted to the Court for in camera inspection. The Court’s order follows the identification which is then followed by an explanation.

l(B-l) and 2(B-2): PRODUCE. Not shown to be work product.

S(C-l): PRODUCE. Not shown to be work product. Farmers also claims the document is privileged as attorney-client communications. The document is a one-page pre-printed form with filled blanks addressed to G.C.I. from Farmers’ Branch Claims Manager requesting G.C.I. to investigate Mr. Stout’s claim and attaching a copy of Farmers’ file (not attached to this memo). Farmers argues in part the following grounds:

This document represents the results of communications between attorney and client within the scope of the attorney-client relationship regarding the investigation of this claim. Providing counsel with information developed during the course of investigation by Farmers Insurance Exchange is necessary to further develop and to aid the attorney-client relationship.

(Memorandum attached to document 3(C-1), p. 2). Farmers’ conclusory statement that this document represents the results of attorney-client communications regarding the investigation of the claim is insufficient. On the face of it, this document is merely a transmittal form. It was not sent to or from counsel or counsel’s agents; it was not sent to or from Farmers’ employees or agents in order to transmit advice from, or information to, counsel; and it clearly was not sent for the purpose of seeking or transmitting legal advice. The attorney-client privilege is not so broad as to cover all of a client’s actions taken as a “result[] of communications between attorney and client.” If counsel advised Farmers that, based upon the law, a good faith evaluation of a claim like Mr. Stout’s would require that certain minimum steps be taken, all of Farmers’ documents produced in order to implement that advice would not be privileged as attorney-client communications. This document is not shown to be privileged as an attorney-client communication.

lf(C-2): PRODUCE. Not shown to be work product. Not shown to be privileged as attorney-client communication. This is a one-page transmittal memo from Farmers’ Branch Claims Manager to G.C.I. enclosing Farmers’ file. Same reasons as for document 3(C-1).

5(B-3), 6(B-Jf), 7(B-5), 8(B-6), 9(B-7), 10(B-8), ll(B-9), 12(B-10), 13(B-11), U(B-12), 15(B-1S): PRODUCE. Not shown to be work product.

16(B-U): TAKEN UNDER ADVISEMENT. This is the State Fire Marshal’s Report on Mr. Stout’s fire. While this document has not been shown to be work product, Farmers does indicate that the Indiana Arson Reporting Statute prohibits it from copying or disseminating the document. If Mr. Stout does not already have a copy of this report or cannot readily obtain a copy from the State Fire Marshal’s office, we will entertain later briefing on whether to order Farmers to produce it. See Jackson v. Brinker, 147 F.R.D. 189 (S.D.Ind.1993).

17(B-15): PRODUCE. Not shown to be work product.

18(C-3): PRODUCE. Not shown to be work product. Not shown to be privileged as attorney-client communication. Same reasons as for document 3(C-1). This is a letter to G.C.I. from Farmers requesting that specific inquiries be pursued during its investigation. It clearly was not sent for the purpose of obtaining legal advice or services, or to transmit within Farmers advice from, or information to, counsel.

19(G-k): PRODUCE. Not shown to be work product. Not shown to be privileged as attorney-client communication. This is one of G.C.I.’s investigation reports to Farmers. Same reasons as for document 3(C-1).

20(B-16), 21(B-17), 22(B-18), 23(B-19), 2MB-20), 25(B~21): PRODUCE. Not shown to be work product.

26(C-5): PRODUCE. Not shown to be work product. Not shown to be privileged as attorney-client communication. This is another of G.C.I.’s investigation reports. Same reasons as for document 3(C~1).

27(B-22), 28(B-23), 29(B-21): PRODUCE. Not shown to be work product.

80(C-6), 81(C-7), 32(C-8), 83(C-9), SUC-IO), 85(C-11): PRODUCE. Not shown to be privileged as attorney-client communications. These are communications between Farmers and G.C.I. regarding G.C.I.’s investigations. Same reasons as for document 3(C-1).

[NOT LISTED] (C-12): PRODUCE. Not shown to be work product. Not shown to be privileged as attorney-client communication. This is Wolf Technical Service’s August 29,1991 supplement to its June 21,1991 report (document 20 (B-16)). It is addressed, however, to Farmers’ outside counsel. Because (1) Wolfs previous report was addressed directly to Farmers and Farmers offers no explanation why this report was sent to counsel, (2) Wolf Technical Services was hired by Farmers on May 20, 1991, the day after the fire, to conduct a cause and origin investigation, before counsel was retained on May 24, 1991, and (3) the report is consistent with Wolfs original mission in that it consists of cause and origin findings, we find that the report does not constitute an attorney-client communication. A party cannot cause a document to become privileged merely by sending it, or having it sent, to, 'or through, counsel. As its showing, Farmers only offers its rote, non-individualized attorney-client argument with respect to this document; it fails to explain how this supplement sent to counsel furthered the rendering of legal advice or services and such is not evident from the document itself or the context of other investigative reports which were sent directly to Farmers. To the extent that Farmers may rely on its counsel suggesting that Wolf undertake this supplementary report, we reject the argument for the same reasons as for document 3(C-1) and because, if counsel was thus acting as a claims adjustment process monitor or advisor, it was not acting in a legal capacity.

36(B-25): PRODUCE. Not shown to be work product.

87(C-1S), 38(C-U): PRODUCE. Not shown to be work product. Not shown to be privileged as attorney-client communication. These are communications between Farmers and G.C.I. and Wolf. Same reasons as for document 3(C-1).

[NOT LISTED] (C-15): PRODUCE. Not shown to be work product. Not shown to be privileged as attorney-client communication. This is Wolf Technical Service’s August 27, 1991 supplemental report which is addressed to Farmers’ outside counsel. Same reasons as for document [NOT LISTED] (C-12).

39(B-26), WB-27), U(B-28): PRODUCE. Not shown to be work product.

12(0-16): PRODUCE. Not shown to be work product. Not shown to be privileged as attorney-client communication. This is a letter from G.C.I. to Farmers regarding information gathered from Mr. Stout’s previous employers, with attached records. Same reasons as for document 3(C-1).

)S(B-29): PRODUCE. Not shown to be work product.

1UC-17): PRODUCE. Not shown to be work product. Not sfiown to be privileged as attorney-client communication. Report from G.C.I. to Farmers. Same reasons as for document 3(C-1).

U(B-SO): PRODUCE. Not shown to be work product.

16(C-18): PRODUCE. Not shown to be work product. Not shown to be privileged as attorney-client communication. Report from G.C.I. to Farmers. Same reasons as for document 3(C-1).

17(B-S1), )8(B-32), 19(B-33): PRODUCE. Not shown to be work product.

50 (NOT INCLUDED): Work product immunity denied because not shown to be work product. Attorney-client privilege claim is TAKEN UNDER ADVISEMENT. This document is described as a “Facsimile Confirmation Report from Farmers Insurance Group to Harrison & Moberly dated 11/14/91” but it was not included with the documents submitted for in camera inspection. Farmers shall submit this document for our review.

52(0-19), 53(0-20), 5)(B-35), 55(B-S6), 56(B-37): MOOT. Mr. Stout requested Farmers’ claims files to November 14, 1991. These documents were generated after that date and will not be ordered produced.

57(B-38), 58(B-S9), 59(B-ltQ): PRODUCE. Not shown to be work product. These documents are either undated or the dates are unrecognizable. Because Farmers does not assert that they were generated after November 14, 1991 and, on their face, they could have been generated earlier, we order them produced.

[UNLISTED] (D-l): PRIVILEGED. This is a communication dated May 24, 1991 from Farmers’ outside counsel to Farmers’ Branch Claims Manager. It renders legal advice.

Conclusion.

For the reasons set forth in this Entry, Farmers’ motion for a protective order is denied and Mr. Stout’s motion to compel is granted for those documents ordered produced above. The motions are otherwise granted and denied respectively except for those matters which are taken under advisement or declared moot. Because a party has ten days after being served with a copy to file objections to a magistrate judge’s pretrial rulings as clearly erroneous or contrary to law, 28 U.S.C. § 636(b)(1)(A); Fed.R.Civ.P. 72(a), Farmers’ duty to produce shall be suspended until after the ten days for objection has expired and pending resolution of any objections which it files. Production to which Farmers does not intend to object shall be made forthwith.

SO ORDERED.

1 . Invoking this Court’s diversity jurisdiction, 28 U.S.C. § 1332, Farmers removed the action to this Court from the Circuit Court for Orange County, Indiana, 28 U.S.C. § 1441.

2 . Because the Rule 34 motion was not presented to the Court, we rely on Mr. Stout's description of the documents requested. Farmers did not contest this description. However, we assume that Mr. Stout actually requested documents Farmers generated between May (not March) 19, 1991, the date of his fire loss, and the date his claim was denied, correctly stated as November 14, 1991.

3 . Mr. Stout's motion to compel came first, then Farmers filed its combined response and its motion for a protective order. Mr. Stout made no reply.

4 . In addition to suing for the proceeds under the policy, Mr. Stout originally sought punitive damages alleging that Farmers refused to pay his claim in bad faith. It would appear, however, that Indiana no longer recognizes a cause of action for punitive damages for an insurer's bad faith refusal to pay on an insurance contract. See Miller Brewing Company v. Best Beers of Bloomington, Inc., 608 N.E.2d 975 (Ind.1993), rehearing denied (June 23, 1993); Erie Insurance Co. v. Hickman, 610 N.E.2d 283 (Ind.App.1993). Mr. Stout has since dropped his claim for punitive damages, without prejudice.

5 . See APL Corp. v. Aetna Casualty & Surety Company, 91 F.R.D. 10, 17 (D.Md.1980); Thomas Organ Co. v. Jadranska Slobodna Plovidba, 54 F.R.D. 367, 373 (N.D.Ill.1972); Atlanta Coca-Cola Bottling Co. v. Transamenca Insurance Co., 61 F.R.D. 115, 118 (N.D.Ga.1972).

6 . Farmers described how the investigations pursued in this case varied from its routine course of business:

"Further, investigation of a fire loss claim in the ordinary course of business where litigation is not anticipated does not include requesting documents regarding the insured's detailed financial history, detailed information of the insured's prior loss history, and detailed verification of information received by Farmers from the insured.

"Farmers’s investigation of a fire loss claim that is done in the routine and ordinary course of business requires the insured to submit to Farmers Insurance Exchange a Sworn Statement In Proof Of Loss and a Personal Property Inventory Worksheet. Such claim is then adjusted and a draft is issued to the insured. Only when the cause and origin of the loss is determined to be arson and evidence exists that connects the insured to the cause of the loss does the insurer engage in an extensive and detailed investigation for the primary purpose of litigating the claim with the insured, if necessary. Evaluating and processing claims wherein it is suspected, based upon objective facts, that arson is the cause of the loss and the insured is connected to such arson vary substantially from the procedures for evaluating and processing claims wherein such facts do not exist.”

(Farmers’ Brief, p. 16-17 (citations to affidavit omitted; original emphases)).

7 . Perhaps confusion could be avoided by label-ling this component of the Rule the "purpose” or "use” prong.

8 . In the same way, an insurer's interest in an accurate claims evaluation may be sparked by the amount of loss and exposure involved, thus motivating it to undertake investigations which it might not otherwise undertake.

9 . The "routine and ordinary course of business” idea appears in two influential sources, but neither supports the use made of it by Farmers. Professors Wright and Miller include this often-quoted passage in their treatise:

Prudent parties anticipate litigation, and begin preparation prior to the time suit is formally commenced. Thus the test should be whether, in light of the nature of the document and the factual situation in the particular case, the document can fairly be said to have been prepared or obtained because of the prospect of litigation. But the converse of this is that even though litigation is already in prospect, there is no work product immunity for documents prepared in the regular course of business rather than for purposes of litigation.

8 C. Wright and A. Miller, Federal Practice and Procedure, § 2024, p. 198 (1970) (footnotes omitted, emphasis added). Although this passage opposes "regular course of business” against "purposes of litigation”, the authors emphasize throughout their discussion of the work product doctrine and Rule that its chief distinction is between documents created for litigation and non-litigation purposes, whatever those purposes may be, whether business or non-business, routine or non-routine. The Advisory Committee Note to the 1970 amendment to Rule 26 states that "[mjaterials assembled in the ordinary course of business, or pursuant to public requirements unrelated to litigation, or for other nonlitigation purposes are not under the qualified immunity provided by this subsection”. (Emphasis added). The emphasized portion reveals, however, that the Committee likewise saw the distinction as being between litigation and non-litigation motivated documents, with those produced in the ordinary course of business being but one example of non-litigation motivated documents.

10 . Further, Fanners makes the same factual error as did the insurer in Harper when it states that,

Only when the cause and origin of the loss is determined to be arson and evidence exists that connects the insured to the cause of the loss does the insurer engage in an extensive and detailed investigation.... Evaluating and processing claims wherein it is suspected, based upon objective facts, that arson is the cause of the loss and the insured is connected to such arson vary substantially from the procedures for evaluating and processing claims wherein such facts do not exist.

(Farmers' Brief, p. 16-17 (bold emphasis added; italics emphasis original)). In this passage Farmers actually reveals that the "extensive and detailed investigations” it undertakes for "evaluating and processing claims” under arson-plus-insured-involvement cases is a matter of Farmers' policy, routine, and ordinary course of business. Harper, 138 F.R.D. at 668 ("A process that has become a company 'policy' and is undertaken as a matter of course is that company's routine.”). Whether a certain procedure is a company's routine or ordinary course of business depends not on how often the procedure occurs, but on whether the procedure is a planned or designed activity which is implemented as a matter of policy once the triggering circumstances or events occur. We repeat that a document producer’s routine is not the work product standard—we only intend here to indicate the inconsistency in the argument as presented.

11 . As to the reasonableness of expecting segregation of documents more in the context of dual-purposed documents, see Lawyers Title Insurance Corp. v. United States Fidelity & Guaranty Co., 122 F.R.D. 567, 569-70 (N.D.Cal.1988), which opines that the costly duplication and risk of inconsistent decisions entailed with segregation makes it an economically and bureaucratically infeasible alternative; thus, it would be unfair and unreasonable for courts to "force insurance companies to employ and deploy two wholly separate teams of adjusters and lawyers to conduct at the same time two separate processes that have the same subject matter.”, id., at 570. The standard for work product protection of concurrent litigative and non-litigative documents is discussed infra, but here it may be observed that the Rule must be interpreted in accordance with its language and the policies it is intended to serve, and it must be applied equally to all litigants. If the Rule is interpreted in such a way that, when applied to insurers, requires disclosure of documents which would otherwise be protected if produced by other parties, solely because of the nature of insurers' business, then courts cannot be charged with unreasonably "forcing" insurers to adopt costly and clumsy internal procedures to protect their documents. In this light, the question more properly becomes whether courts should extend special protections to insurance carriers. See Thomas Organ, 54 F.R.D. at 373. If insurers wish to undertake extraordinary measures to protect the confidentiality of otherwise disclosable documents, they are free to do so. It should be observed, however, that far from being unique to insurers, the problem of dual-purposed documents affects many types of parties—whoever produces investigative or evaluative documents for purposes other than litigation. See Soeder v. General Dynamics Corp., 90 F.R.D. 253, 255 (D.Nev.1980) (manufacturer's investigation of crash of one of its aircraft); Janicker v. George Washington University, 94 F.R.D. 648, 650-51 (D.D.C.1982) (university's investigation of fire at dormitory); Binks, supra. Further, Lawyers Title’s statement that segregation is infeasible because claims adjusters could not "responsibly” process a claim during litigation without considering "the claims, assertions and arguments made in the complaint and other documents filed in the litigation”, Lawyers Title, 122 F.R.D. at 569, cannot be accepted because such court pleadings are produced by the insured, not the insurer, so they are irrelevant to insurers' work product protections and, because such documents are matters of public record, they would not constitute work product.

12 . Under this high threshold, the cases often appear to turn more on proximity to litigation than anticipation of it.

13 . An insured's statements of intention to sue must be carefully evaluated. The kind of statement which is required is notice that suit will be imminently filed, amounting to a demonstration that the insured has an "identifiable resolve to litigate.” Threats made only for tactical or other reasons, e.g., to induce action or to forcefully make known the insured’s position, must be distinguished. An unrepresented insured who threatens suit if his insurer doesn't make a quick claims decision, denies the claim, or continues to pursue the adjustment process in an improper or dilatory way, does not demonstrate the imminence of litigation or a firm resolve to litigate, but only the contingency or possibility of litigation. See Binks, 709 F.2d at 1120 (threatening tone of buyer’s letter to seller which concluded with statement that ”[i]f you persist in your choice not to make the necessary corrections, we shall proceed on our own and continue to hold you fully responsible for any damages and expenses incurred” did not justify a sufficient anticipation of litigation on seller’s part because "it still falls short of stating that [buyer] intended to institute litigation concerning the System. Rather, the letter clearly states that [buyer] still intended to persuade [seller] to correct the problems in the System and, falling short of that, to withhold full payment of the purchase price of the System.”).

14 . See United States v. Davis, 636 F.2d 1028, 1040 (5th Cir.), cert. denied, 454 U.S. 862, 102 S.Ct. 320, 70 L.Ed.2d 162 (1981); United States v. Gulf Oil Corp., 760 F.2d 292, 296 (Temp.Emer.Ct.App.1985); Binks Manufacturing Co. v. National Presto Industries, Inc., 709 F.2d 1109, 1119 (7th Cir.1983) (quoting Janicker); Blockbuster Entertainment Corp. v. McComb Video, Inc., 145 F.R.D. 402, 403-04 (M.D.La.1992) (citing Davis); Allendate Mutual Insurance Co. v. Bull Data Systems, Inc., 145 F.R.D. 84, 87 (N.D.Ill.1992) (quoting Binks quoting Janicker); Colonial Gas Co. v. Aetna Casualty & Surety Co., 144 F.R.D. 600, 605 (D.Mass.1992) (citing Atlantic Financial); United States v. Rosenthal, 142 F.R.D. 389, 393 (S.D.N.Y.1992) (citing Gulf Oil); Martin v. Valley National Bank of Arizona, 140 F.R.D. 291, 304 (S.D.N.Y.1991) (citing Binks); Ferguson v. Lurie, 139 F.R.D. 362, 367 (N.D.Ill.1991) (citing Commonwealth Edison and Janicker); Exxon Chemical Patents, Inc. v. Lubrizol Corp., 131 F.R.D. 668, 670 (S.D.Tex.1990) (citing Davis); Henderson v. Zurn Industries, Inc., 131 F.R.D. 560, 570 (S.D.Ind.1990) (citing Binks); McMahon v. Eastern S.S. Lines, Inc., 129 F.R.D. 197, 198-99 (S.D.FIa.1989); Zullig v. Kansas City Power & Light Co., Civ. A. No. 87-2342, Memorandum and Order, 1989 WL 7901, *4 (D.Kan. Jan. 17, 1989) (citing Hardy, Scott Paper, Janicker, and Soeder); In re Atlantic Financial Management Securities Litigation, 121 F.R.D. 141, 144 (D.Mass.1988) (citing Gulf Oil); E.E.O.C. v. Commonwealth Edison, 119 F.R.D. 394, 395 (N.D.Ill.1988) (quoting Janicker and citing Binks); Independent Petrochemical Corp, v. Aetna Cas. & Sur. Co., 117 F.R.D. 292, 298 & n. 11 (D.D.C.1987) (citing Janicker); Hardy v. New York News, Inc., 114 F.R.D. 633, 644 (S.D.N.Y.1987) (citing Gulf Oil); Kelly v. City of San Jose, 114 F.R.D. 653, 659 (N.D.Cal.1987); Scott Paper Co. v. Ceilcote Co., Inc., 103 F.R.D. 591, 595 (D.Maine 1984) (quoting Janicker); Janicker v. George Washington University, 94 F.R.D. 648, 650 (D.D.C.1982); In re International Systems and Controls Corporation Securities Litigation, 91 F. R.D. 552, 557 (S.D.Tex.1981) (citing Davis), order vacated on other grounds, 693 F.2d 1235 (5th Cir.1982).

15 . The Court’s holding consisted of the following (footnote omitted):

"It is often difficult to determine with precision when a particular attorney's work is 'in anticipation of litigation’ rather than 'an investigative report developed in the ordinary course of business.’ Janicker, 94 F.R.D. at 650. We conclude though, that while there may have been 'the remote prospect of litigation’ when Lavers [the creator of the documents] prepared his memoranda, the appellant has failed to meet its burden of proving that the memoranda were 'prepared ... because of the prospect of litigation,' Diversified Industries [Inc. v. Meredith ], 572 F.2d [596] at 604 [ (8th Cir.1977) ] (emphasis added), or, that 'some articulable claim, likely to lead to litigation,’ Coastal States Gas Corp. [v. Dept. of Energy ], 617 F.2d [854] at 865 [(D.C.Cir.1980) ] (emphasis added), had arisen. Therefore, we hold that Presto failed to meet its burden of proof necessary to invoke the work product privilege for the January 21 and January 25 memoranda.”

16 . We regret our misquotation in Harper of a statement from Special September 1978 Grand Jury, which was also misquoted in Henderson. Both of these decisions quoted the Seventh Circuit as stating that it does "not read the 'in anticipation of litigation' requirement broadly." Harper, 138 F.R.D. at 661; Henderson, 131 F.R.D. at 570. By omitting the word "that” which appears before "broadly” in Special September 1978 Grand Jury, 640 F.2d at 65, we inadvertently converted the Seventh Circuit's response to a specific argument of counsel in that case into a general rule of interpretation. However, the underlying principle of narrow interpretation in that case is still valid, though weaker, support for our broader point. 'Attention to detail' and 'always check the original’ are good advice for judges as well as attorneys to follow.

17 . One commenter misunderstood the policy behind the work product rule as applied to insurers:

"In Upjohn Co. v. United States [449 U.S. 383, 101 S.Ct. 677, 66 L.Ed.2d 584 (1981)], the United States Supreme Court devoted a portion of its discussion to the work product doctrine and the policies behind it. The Court noted that a key reason for protecting certain materials is the need to encourage parties to develop all of the facts and compile a thorough record of each case. Protection of these records allows a party to carry out his quest for information without fear of having his written materials revealed to an opponent.

"The foregoing policy is certainly applicable to insurance litigation. In order to provide deserving claimants with fair protection under their insurance policies and to avoid payment of illegitimate claims, insurers should be encouraged to conduct complete investigations into possible claims and maintain accurate files of surrounding facts and conclusions. In the Hickman case, for example, the Court warned that allowing one party free access to any relevant information possessed by the other would undermine such a policy. If claimants are readily allowed to discover files kept by insurers, the insurers might decide to limit the preparation of documents regarding a particular claim. A request for discovery of such limited files would not likely produce beneficial information for a claimant, and society as a whole would suffer from the incomplete services that insurance companies would provide.”

"Work Product Discovery in Insurance Litigation”, 18 Indiana Law Review 547, 558-59 (1985) (footnotes omitted). All authorities recognize that the immunity from discovery accorded to work product is designed only to promote parties’ and attorneys’ litigation preparation efforts. No decision that we have come across has held or opined that the work product immunity was designed to, or should be applied to, promote the accuracy or efficiency of insurers’ private business operations or any party’s private endeavors, as suggested in this passage. The incentive for insurers to conduct complete and efficient investigations into claims and maintain accurate files is adequately provided by the contractual obligations they assume, the possibility of legal enforcement of those obligations, and economic incentives.

18 . See, e.g., Westinghouse v. Republic of the Philippines, 951 F.2d 1414, 1428 (3rd Cir.1991) (work product rule "promotes the adversary system by enabling attorneys to prepare their cases without fear that their work product will be used against their clients”); Shields v. Sturm, Ruger & Co., 864 F.2d 379, 382 (5th Cir.1989) ("The work product privilege ... promote[s] the adversary system by safeguarding the fruits of an attorney's trial preparations from the discovery attempts of an opponent”); Allendate, 145 F.R.D. at 87. Cf. Hewlett-Packard Co. v. Bausch & Lomb, Inc., 116 F.R.D. 533, 538 (N.D.Cal.1987) ("The Hickman Court appreciated that our system of adjudication is at its heart dialectical. If one side relies (for ideas or information or legal research) on the other, half of the dialectic is lost. Half a dialectic is no dialectic at all. Thus the principal purpose of the work product doctrine is to preserve the vigor of the dialectic by creating an environment in which each lawyer feels not only free, but pressured, to do his own work and to make both his investigation and his evaluation of the case as aggressive, as thorough, and as internally frank as possible”).

19 . However, regardless of the producer’s intended use for a document, if it actually uses the document for non-litigation purposes, the document should not qualify as work product. See Harper, 138 F.R.D. at 663, n. 3. If an insurer's adjusters later acquire and use reports of investigations or witness statements which were earlier ordered by its attorneys to prepare for litigation (because, either due to neglect, delay, or other reasons, the adjusters failed to order the investigations previously), the distinction between original intention and actual use would not justify denial of disclosure since disclosure would not contravene the Rule’s policy of protecting a party’s or counsel's litigation preparation efforts. Such previously segregated documents may certainly be deemed "prepared” or "acquired” by the insurer for non-litigation purposes. Otherwise, work product immunity would turn on fortuitous (or planned) circumstances such as an insurer employing counsel to begin preparation for litigation (assuming a sufficient anticipation of litigation exists) and commencing serious claim evaluation only after the common investigations and interviews had been ordered by the attorneys. In that case, a court would have to attempt to determine whether the particular investigations or interviews would have been ordered by the insurer for claims evaluation purposes. Surely the best indication of that circumstance would be the fact that the insurer did eventually acquire and use the documents for that purpose. Absent a contrary showing, however, it may be presumed that a document’s intended use was also its actual use.

20 . From Farmers' Brief, pp. 13-15, taken from the affidavit of Farmers' Branch Claims Manager, attached to the brief (original emphases):

a. On May 19, 1991, Richard L. Stout's residence was destroyed by fire.

b. On May 20, 1991, Farmers Insurance Exchange received a report of the fire loss.

c. On May 20, 1991, Defendant instructed Wolf Technical Services to conduct an origin and cause investigation of the fire loss.

d. On May 21, 1991, Farmers Insurance Exchange discovered that the Plaintiff had a previous fire loss in January, 1991, with almost identical bum patterns.

e. On May 21, 1991, Farmers Insurance Exchange Claim Representative, Mark Oswald, inspected the scene.

f. On May 22, 1991, the Defendant obtained a recorded statement from the Plaintiff during which the Defendant learned the following:

1. Mr. Stout was a self-employed carpenter claiming to earn close to 2,000.00 a month, but he could not remember how much income he reported on his last tax return or whether he filed his last tax return in April, just one month earlier.

2. Mr. Stout admitted that his house was locked up on Saturday, May 18, 1991, when he left at approximately 3:00-4:00 p.m., and that he is the only one who has keys to his home. Mr. Stout was not home at the time of the fire.

3. Mr. Stout stated that he was not aware of any signs of forcible entry into his home.

4. In addition to his previous fire loss in January, 1991, Mr. Stout admitted that he had another fire loss in Mitchell, Indiana 12-15 years ago (it was subsequently learned to have occurred in 1984).

5. The home on Olive Street which suffered the fire loss was listed for sale prior to the loss.

6. Mr. Stout refused to answer questions concerning whether he had "other loans or debts." Then refused to answer all questions concerning loans and debts.

g. On May 23, 1991, Claims Representative Mark Oswald reviewed the file regarding Plaintiff’s previous fire loss in January 1991, which had very similar burn patterns under very similar circumstances. For example, Mr. Stout's previous fire loss occurred on Sunday, January 13, 1991, at 3:27 a.m., while the most recent fire loss occurred on Sunday, May 19, 1991, at 4:45 a.m.

h. Mr. Stout obtained insurance on March 12, 1991 for his new home which suffered a loss.

i. On May 23, 1991, the Defendant received an anonymous telephone call informing the Defendant that the Plaintiff was involved with having various properties burned.

j. On May 24, 1991, Mr. Stout called Defendant requesting an increase in advanced monies, and yet would not disclose if he had any debts or loans.

k. On May 24, 1991, the Defendant referred this matter to Greater Cincinnati Investigation, Inc. (GCI) requesting additional investigation of this claim.

l. On May 24, 1991, Defendant talked to Gerry Mang of Wolf Technical Services, who had conducted the cause and origin investigation, and who indicated that based upon the scene inspection, there was a strong suspicion of arson.

m. On May 24, 1991, Mr. Stout called Defendant and spoke to Patrick Ortman, Branch Claims Supervisor, and demanded a new claim representative be assigned to his case because the current claim representative was asking questions about his finances. Mr. Stout stated that if he did not get a new adjuster assigned to his case he would retain an attorney and file suit.

n. On May 24, 1991, the Defendant referred this claim file to Harrison & Moberly for legal advice.

21 . In the absence of further explanation, we see little if any significance to our inquiry of the facts that Mr. Stout claimed to be a self-employed carpenter earning about $2,000 a month, that Farmers hired G.C.I. to investigate the fire (we aren't told why, or when in the chronology of events on May 24), or that Farmers referred the case to attorneys for legal advice. The fact that Farmers later discovered that Mr. Stout’s pre1991 fire occurred more recently than he described is not relevant unless we are told that it discovered the inaccuracy by May 24, 1991.

22 . On May 24, 1991, Farmers had only suspicions and undertook further investigations to arrive at conclusions.

"... based upon the information [Farmers] had learned by May 24, 1991, there was a likelihood that the claim may be denied if the objective facts continued to hold true----” (Farmers' Brief, p. 15 (emphasis added)).

"... after the claim was referred to Harrison & Moberly on May 24, 1991, representatives of Farmers regularly consulted with ... attorneys for Farmers, with regard to the scope, details, and findings of the investigation of Plaintiff's claim in order to make a good faith decision regarding the claim of Richard L. Stout, Jr.” (Id. at 16 (emphasis added)).

"Evaluating and processing claims wherein it is suspected, based upon objective facts, that arson is the cause of the loss and the insured is connected to such arson vary substantially from the procedures for evaluating and processing claims wherein such facts do not exist.” (Id. at 16-17 (first emphasis added)).

Farmers' post-May 24 investigations were "undertaken primarily to prepare for litigation in the event the additional investigation confirmed the initial information supporting suspicions of Mr. Stout's involvement.” (Id. at 17 (emphases added)).

See also id. at 21, 24.

23 . Id.

24 . Id.

25 . "... based upon the information it had learned by May 24, 1991, there was a likelihood that the claim may be denied if the objective facts continued to hold true and, therefore, that there was an obvious likelihood that litigation would ensue with the insured." (Farmers’ Brief, p. 15).

Fanners’ investigations were undertaken "for the primary purpose of litigating the claim with the insured, if necessary,” (id. at 16), and "to prepare for litigation in the event the additional investigation” confirmed its suspicions, (id. at 17).

"... Farmers anticipated that litigation was likely to occur with regard to this claim....” (Id. at 17).

"Based upon the facts revealed prior to May 24, 1991, Farmers had a reasonable basis for anticipated [sic] that litigation with the insured was likely to occur on this claim if these facts continued to be verified through additional independent sources.” (Id. at 24 (emphasis added)).

Farmers also states that it felt that litigation was "substantially imminent” by May 24, 1991, (id. at 17), and there was a "strong likelihood” that litigation would occur, (id. at 21).

26 . Farmers withholds the post-May 24 Wolf Technical Services reports and the Greater Cincinnati Investigation reports. It claims that it began anticipating and preparing for litigation after that date based on the facts discovered by that time. If we were to interpret the Rule as ignoring actual uses of documents and concentrating solely on original motivations, it is clear that, because these investigations and their resulting reports were ordered by May 24, 1991, when Farmers contends it began to prepare for litigation, the later reports would not qualify as work product under Farmers’ contentions or the facts. The Wolf investigation was ordered on May 20, 1991, one day after the fire and before Farmers had any indications as to the nature of the fire or Mr. Stout’s involvement and before Farmers could have developed a litigation purpose. The G.C.I. investigation was ordered on May 24, 1991, the day Farmers claimed to have begun operating in anticipation of litigation.

27 . Farmers states that it undertakes its "extensive and detailed investigation” only when the cause of a fire is "determined” to be arson and "evidence exists” connecting the insured to the cause of the loss. (Farmers’ Brief, p. 16). However, in this case, by May 24, Farmers had not "determined” that Mr. Stout’s fire was arson; it had only suspicions and undertook the further investigations to confirm those suspicions. We also question whether Farmers would not order further investigations into a suspected arson fire even in the absence of evidence at the time implicating its insured.