[28] I do not share the majority's view that the In re Martin2 precept was manifestly applied by the trial court in this case as a "tie breaker." Nor could it have been. There was, in my opinion, no tie.
[29] Whatever else might be said about the comparison of the facts in this case with those of Martin, it must be conceded that the trial court in Martin relied solely upon conflicting circumstances in resolving the factual doubts in favor of the claimant's widow. The high court admiringly patted the trial judge on the back, paying inferential homage to a hallmark of basic justice — that if the court cannot always be right it should avoid being wrong to the detriment of the worker when it comes to determining his entitlement in factually unclear death claims.
[30] In the case at bar, the trial judge had more than the less than cogent circumstantial evidence available in Martin. He had direct evidence, or at least evidence approaching such status. The undisputed evidence is that Hurlburt at one point told his acquaintance in Oklahoma City he would not be coming there that night. Then about an hour later, he told his neighbors and his wife he had to go investigate an accident on the Turner Turnpike. Evidently because of the new development, he placed a second call to his friend in Oklahoma City and told her that he would be in Oklahoma City later in the evening after all. His death occurred about an hour later four miles west of Sapulpa — a location that was in the vicinity of where he said he would be working.
[31] The two calls to Oklahoma City in themselves strongly suggest something occurred between the first and second call. It probably was a request for Hurlburt to investigate an accident near Sapulpa. This inference, along with the undisputed facts, places Hurlburt, who was on 24-hour-call for his company, directly in the course of his employment when he died.
[32] To overcome this, the employer introduced certain evidence which tended to raise some doubt about whether deceased was on his way to or from an accident when he was killed. The majority takes this evidence, stirs in some speculations about evidence neither party sought to adduce, and comes up with an evidentiary mix they say preponderates in favor of the fact that Hurlburt was on a frolic of his own when he lost his life. However, this assumes the preponderance test controls the instant case rather than the reasonable doubt test — an assumption that forms the foundation for my disagreement with the result reached by the majority. For indeed, the facts might well be able to survive the preponderance test and yet still give rise to a reasonable doubt resolvable in favor of the claimant.
[33] In my opinion the evidentiary sufficiency tests used in either criminal or civil cases are not applicable in a workers' compensation death claim contest as suggested by the majority. The test to be employed in the latter case is a just and humane one lucidly promulgated in Martin, namely, that where "there is reasonable doubt as to whether an accident did in fact arise out of the employment such doubt shall be resolved in favor of the claimant."3
[34] I am satisfied this test was not applied by the trial judge in this case. And because I believe a reasonable doubt exists about whether Hurlburt's death was work related, I think a different result would have been reached by the trial Judge had it been applied. I would therefore vacate the order appealed and remand the case with instructions to apply the doubt test in evaluating the evidence.