(concurring) — I concur with the majority’s conclusion that based on the context in which the phrase *753“member of the family” appears, and based on the other language set forth in the declaration describing the insured, the average purchaser of this insurance policy in 19945 would have understood that the policy covered only the purchaser and those individuals connected to the purchaser by blood, affinity, or law. I write separately in an attempt to clarify why it is inappropriate to apply the very broad definition of “member of the family” advocated by the dissent.
The first and essential question before us is: Would the average purchaser of such a policy at that time reasonably have understood the phrase “member of the family,” as the policy uses that term, to include another adult living in the same residence as the insured at the time of the accident who was unrelated to the insured by blood, affinity, or law? Kitsap County v. Allstate Ins. Co., 136 Wn.2d 567, 575, 964 P.2d 1173 (1998); Allstate Ins. Co. v. Peasley, 131 Wn.2d 420, 424, 932 P.2d 1244 (1997). I believe that the answer is no and that the phrase, “member of the family,” in the insurance policy context here, is unambiguous. Thus, the various other possible definitions provided by the dissent are immaterial and create ambiguity where none exists. See Kitsap County, 136 Wn.2d at 576.
Although the dissent verbalizes the necessity of considering the language at issue in context, it looks to very dissimilar contexts to support its ambiguity conclusion. First, it notes the multiple dictionary definitions, an exercise that does not promote the context discussion. It then turns to the case of Mains Farm Homeowners Ass’n v. Worthington, 121 Wn.2d 810, 854 P.2d 1072 (1993), in which the Supreme Court grappled with a restrictive covenant limiting the use of property to “single family residential.”
Looking to the intent of the parties, the Mains court held that unrelated adults residing in an adult family home did not constitute a “family” and, thus, the operation of such a *754home would violate the covenant. The Mains court first notes that:
No purpose will be served by examining and comparing in detail the numerous cases which define “family”. Because of the widely differing documents being interpreted, the contexts in which the word is used and the fact-specific circumstances, it is impossible to arrive at a single, all-purpose definition.
121 Wn.2d at 817. It then briefly explores “the possible definitional spectrum” before turning to the specific context at issue. The opinion discusses the details of the adult home, the reasonable expectations of other lot owners, and the language in the covenant that focused on the purpose of the property use as opposed to the nature of the structure on the property.
After concluding that, in context, this was not a family home, the Mains court provided the following cautionary note, which is applicable here:
We caution that the interpretation of a particular protective covenant is largely dependent upon the facts of the case at hand. Our holding should not be construed as an encompassing declaration concerning covenants and uses under other circumstances.
121 Wn.2d at 827. Nonetheless, the dissent cites and quotes Mains as authority for a definition broader than “ ‘persons related by blood, marriage, or adoption.’ ” Dissent, at 769.1 believe this approach takes language and facts out of context and does not look at the facts before us in context.
The dissent next turns to a 1935 case, Collins v. Northwest Casualty Co., 180 Wash. 347, 349, 39 P.2d 986 (1935), which involved the interpretation of the phrase “an adult member of the named Insured’s household,” as used in an automobile insurance policy. The issue in Collins was whether the “household” survived the death of the insured. In its discussion of the various meanings for the word “household,” the Collins court notes that “household” suggests a different meaning than does “family.” This dicta neither provides context for the case at hand nor shows that the word “family” is ambiguous in the insurance policy context.
*755Finally, the dissent’s reliance on the author of a 1960 anthropological “work” is not persuasive. The Murdock description of the family as a “social group” contains no clue as to context. We can only assume from its anthropological orientation that its context is universal and nonspecific, the exact opposite of the relevant context in this insurance policy case. See Webster’s Third New Int’l Dictionary 93 (1969).6
The variety of potential meanings of “family” outside the context of the insurance policy at issue does not justify a conclusion that the relationship between Ray Edinger and Blake Matthews falls within a reasonable definition of “family.” As the dissent acknowledges, we do not look at the various potential definitions for a word in an insurance policy unless the average purchaser would understand the word, as used in the policy rather than viewed in isolation or in other contexts, to have multiple meanings.
The majority looks at the “common meaning” of family to determine how the average purchaser would understand the policy. This method is consistent with Peasley, 131 Wn.2d at 425-27, and supports the majority’s holding. The dissent challenges this approach, contending that because of the absence of a definition in the policy, the average purchaser “would not have understood that Penn-America’s policy adopted nor eliminated, expressly or impliedly, any other meaning of the word ‘family.’ ” Dissent, at 771. Assuming that the dissent is saying that the policy included any meaning of family other than one defined solely by residency, it fails to demonstrate that the average purchaser would find those multiple potential meanings applicable in this context.
The majority’s effort to interpret and clarify each clause in the policy, here specifically “member of the family” and “resident of the household,” is a helpful and reasonable *756methodology. Boeing Co. v. Aetna Cas. & Sur. Co., 113 Wn.2d 869, 876, 784 P.2d 507 (1990). The fundamental difference between the majority and the dissent is that the dissent would define as family members those individuals connected to the insured through a pseudo-family relationship because they are similar to the relationships within the definition. But good friends who enjoy each other’s company and live under the same roof do not become family members simply because their friendship has certain familial characteristics such as affection, caring, and companionship.
This case involves an even more remote relationship — a member of the family of a pseudo-relative. To describe this legal relationship as a family does not clarify ambiguity, instead it creates ambiguity where none exists. Thus, even assuming that the evidence adduced at trial was sufficient to prove that Sue Matthews was Ray Edinger’s meretricious spouse, the dissent points to nothing in the law or in the public policy of this state in 1994 that would automatically make all resident members of Sue’s family also members of Edinger’s family for insurance coverage purposes.
Because I believe that the average purchaser of this policy in 1994 would have understood this phrase to mean only individuals related to the insured by blood, affinity, or law, and not to roommates or good friends living together, I join the majority in affirming the trial court.
Ray Edinger purchased this six-month policy in 1994.
Webster’s defines anthropology as “the study considering man’s physical character, historical and present geographical distribution, racial classification, group relationships, and cultural history.”