concurring.
I agree with the majority’s conclusion that National Surety’s denial of the claim is the operative occurrence in this discovery case, and that documents made prior to that occurrence are discoverable. I also agree with the dissent that Rule 166b(3)(d) requires a trial court to answer three questions in determining whether investigative documents should be exempt from discovery:
(1) was the communication made among a party and its agents?
(2) was it made subsequent to the occurrence or transaction upon which the suit is based?
(3) was the subject of the communication connected to the prosecution, investigation, or defense of the claim or to the investigation of the occurrence or transaction out of which the claim has arisen?
There is no serious dispute in this mandamus proceeding about questions (1) and (3). We are concerned with the second question here: how far into the history of dealings between parties to a lawsuit may one of the parties keep its internal communications secret?
Rule 166b(3)(d) incorporates the prior provision of old Rule 186a. This new rule has somewhat broadened the investigative exemption by expanding the language relevant to the third question above. However, the rule has not waivered in its requirement that the communication be made subsequent to the occurrence or transaction that is the basis of the suit.
In Texas Employers’ Insurance Association v. Fashing, 706 S.W.2d 801 (Tex.App.—El Paso 1986), an employee was discharged after filing two worker’s compensation claims. In his wrongful discharge suit against his employer, he sought discovery of the investigative files of the compensation carrier. The carrier resisted, the *71trial court ordered discovery, and the carrier applied for a writ of mandamus to get the discovery order rescinded. The compensation carrier claimed that the investigative communications regarding the prior worker’s compensation claims were exempted because proof that the injury claim was in good faith is an essential element of the wrongful discharge suit. The El Paso Court disagreed with this “essential element” analysis. It held that the discharge was the occurrence upon which the suit was based; the discharge “triggered” the cause of action. See also State v. Clark, 695 S.W.2d 673 (Tex.App.—Austin 1985).
This case is similar, though the question is closer. The surety company’s denial of the claim on the bond triggered the lawsuit, not the malfeasance of the bank officer almost ten years before that, as the dissent maintains. The malfeasance is merely an element of proof in this suit on the bond.
An insurance policy is at issue in this case, with the unique relationship between insurer and insured. When a person buys a contract of insurance, he has the insurer’s promise to pay certain amounts upon the occurrence of certain events. In the normal scheme, there is no dispute between them until the insurer denies a claim for loss. After this, the insured files a lawsuit, not when the loss occurs. Before this, discovery should be allowed.
Where the dealings between the parties are complex and protracted, as in this case, the time determination which a trial court must make, based on statutory construction of Rule 166b(3)(d), is exceedingly difficult to make. In a case like the one that is before us, any number of events may occur which ultimately lead to litigation. Under the present wording of the rule, and the lack of cases construing it, too many lawsuits are being held up for additional months while the discovery disputes are resolved. There has to be a definite test, set out by the Supreme Court, which practitioners and trial courts can easily apply to these recurring problems. Otherwise, our intermediate appellate courts will continue to be faced with what amounts to interlocutory appeals of discovery issues by means of original proceedings.