dissenting:

This appeal does not call on us to resolve definitively the coverage dispute between the parties. The only issue is whether National Casualty’s duty to defend has been triggered. To prevail on that issue under California law, National Casualty must affirmatively establish that no reasonable reader could have understood the insurance contract to provide coverage. Because I do not believe that National Casualty has met that burden on this record, I respectfully dissent.

I

Under California law, the duty of a liability insurer to defend claims asserted against an insured is extremely broad. Montrose Chemical Corp. v. Canadian Universal Ins. Co., Inc., 6 Cal.4th 287, 295 , 24 Cal.Rptr.2d 467 , 861 P.2d 1153 (Cal.1993). An insurer must defend its insured against a suit “which potentially seeks damages within the coverage of the policy.” Gray v. Zurich Insurance Co., 65 Cal.2d 263, 275, 54 Cal.Rptr. 104 , 419 P.2d 168 (Cal.1966). This obligation can be excused only when the third party complaint “can by no conceivable theory raise a single issue which could bring it within the policy coverage.” Montrose Chemical Corp., 6 Cal.4th at 287 , 24 Cal.Rptr .2d 467, 861 P.2d 1153 . “In other words, the insured need only show that the underlying claim may fall within policy coverage; the insurer must prove it cannot.” Id. Any doubt as to whether the facts give rise to a duty to defend is resolved in the insured’s favor. Id. at 299-300 , 24 Cal.Rptr.2d 467 , 861 P.2d 1153 . The insurer bears the burden of proof and must affirmatively establish the absence of coverage. Id. at 287 , 24 Cal.Rptr.2d 467 , 861 P.2d 1153 .

Further, under California law, when ambiguity arises from competing interpretations of the insurance contract, the insurer cannot succeed by showing that the insured’s interpretation is less reasonable than some alternative; instead, the insurer must show that the insured’s interpretation is objectively unreasonable. American Internat'l Specialty Lines Ins. Co. v. Continental Casualty Ins. Co. (“AISLI”), 142 Cal.App.4th 1342 , 49 Cal.Rptr.3d 1, 13 (2006). If conflicting interpretations demonstrate a policy ambiguity, then California law requires that the ambiguity be *530 construed against the insurer since the insurer drafted the contract. Cal. Civ. Code § 1654 ; AIU Ins. Co. v. FMC Corp., 51 Cal.3d 807, 821 , 274 Cal.Rptr. 820 , 799 P.2d 1253 (1990). In other words, when policy ambiguities are at issue, as they are in this case, the insurer must show that no reasonable reader could have understood the insurance contract to provide coverage.

Because this case was decided at summary judgment, we must construe all of the facts in the light most favorable to Launch Media.

II

Launch Media has offered a reasonable construction of the Media Special Perils Policy Multimedia Coverage policy provision at issue. The specific dispute in this case is a narrow one. The parties disagree only as to the proper interpretation of the aggregating provision — or “each loss” provision — in Launch Media’s liability policy. While National Casualty argues that all claims in the underlying Arista Records action can be aggregated into a single loss, Launch Media contends that the Arista Records claims constitute eleven separate losses, ten in the first policy period and one in the second.

The parties here agree that neither National Casualty’s causal theory of aggregation nor Launch Media’s eleven losses theory of aggregation strictly “comportfs] with the actual policy language.” That fact alone should be enough to trigger a duty to defend since ambiguity should be construed against the insurer. However, in my view, Launch Media’s policy construction is more persuasive than National Casualty’s.

A

Construing the facts in the light most favorable to Launch Media, I conclude that Launch Media has established occurrences in two policies periods. The policy provision at issue provides coverage for losses attributable to acts “committed in the utterance or dissemination of matter by or with permission of the Insured during the Policy Period in multimedia activities and related advertising, regardless of when claim is made or suit is brought.”

The policy limit applying to “each loss” is the “Policy Limit and Self-Insured Retention in effect on the date on which matter is first uttered, disseminated, gathered, acquired or obtained.” The typical purpose of this clause in a media policy is to limit liability for republication to the first policy period in which the original publication was made, or “first uttered.”

This case involves a number of matters that were “first uttered” during the initial policy period. However, Launch Media has demonstrated that it first played the song Liquid Dreams during the later policy period, and it has shown that a record company called J Records owned the copyright to Liquid Dreams at the time that the song was first played. For the purposes of summary judgment, Launch Media has therefore sufficiently demonstrated that it first uttered new matter during the second policy period and that the matter was owned by a person who did not own any of the previously uttered matter.

Under the plain terms of the policy, National Casualty owed a duty to defend Launch Media as to liability claims arising from matters “first uttered” during the second policy period. Launch Media has therefore established the potential for coverage, triggering National Casualty’s duty to defend Launch Media under the second policy, as well as the first policy.

B

National Casualty argues that it does not have any duty to defend because the *531 incident in the second policy period relates back to the claims arising during the first policy period and must therefore be aggregated with them for purposes of analyzing its duty to defend. But this interpretation conflicts with the plain language of the policy, which provides for coverage “regardless of when claim is made or suit is brought.” It also defies a common sense application of the policy. There is no doubt, for example, that had a separate suit been brought based on the Liquid Dreams infringement alone, National Casualty would have had the duty to defend the action under the terms of the policy. There is no basis in the policy for excusing that duty based on the existence of other losses and claims during a different policy period.

National Casualty asserts a causal theory interpretation of the policy, under which its duty to defend would be limited to a single action, asserting that the multiple infringements at issue here arise from a single decision to webcast copyrighted songs. This interpretation has no basis in the text of the policy. Further, the cases that National Casualty cites in support of this theory have the opposite effect from the one that National Casualty intends. Rather than demonstrating a legal definition of “occurrence” that applies to all insurance aggregation provisions, the cases demonstrate that insurers are capable of writing causal aggregation clauses and that National Casualty completely failed to do so in this case. The provisions at issue in Bay Cities Paving & Grading, Inc. v. Lawyers’ Mut. Ins. Co., 5 Cal.4th 854 , 21 Cal.Rptr.2d 691 , 855 P.2d 1268 (Cal.1993), EOTT Energy Corp. v. Storebrand Int’l, 45 Cal.App.4th 565 , 52 Cal.Rptr.2d 894 (Cal.App.1996), and American Intemat’l Specialty Lines Ins. Co. v. Continental Casualty Ins. Co., 142 Cal.App.4th 1342 , 49 Cal.Rptr.3d 1, 21 (2006), tie aggregation to “acts,” “happenings,” “events,” “circumstances,” “transactions,” and “decisions,” all of which are terms that support a causal theory of aggregation. The provision at issue here, by contrast, includes no such language, aggregating instead by reference to “utterances or disseminations of matter ... relating to the same subject, person, or class of persons.” The policy language at issue here cannot support National Casualty’s theory that its duty to defend is limited to the one action filed based on a single decision to infringe. Rather, the language more reasonably supports Launch Media’s policy interpretation, understanding the language in the policy aggregating losses by “subject, person or class of persons” to mean that claims pertaining to eleven discrete companies should be aggregated by company. In the context of the lawsuit at issue, this construction would mean that the infringement of multiple songs owned by one record label would be treated as one loss, while the infringement of multiple songs owned by another record label would be treated as a separate loss. Because there are eleven copyright owners involved, there are eleven separate losses covered by the policy.

Launch Media’s interpretation not only conforms with the policy language, but also makes eminent sense. If, for example, the eleven owners had filed eleven different actions, National Casualty would doubtless have to defend them all. The fact that the eleven losses were essentially caused by one corporate decision would not relieve National Casualty of its duty to defend the separately filed lawsuits. However, under National Casualty’s theory, it would only have to defend one suit. This construction makes no sense and finds no support in the policy language.

National Casualty also argues that the fact that Launch Media paid only one SIR in the first policy period demonstrates that *532 Launch Media’s interpretation of the policy in practice was different from the theory it now asserts. However, at best, this fact shows Launch Media’s subjective interpretation at the time it paid the SIR. It is not relevant to the objective reasonableness of Launch Media’s eleven losses theory, nor is it relevant to the question of what the reasonable expectations of the insured were at the time the policy was issued. Our inquiry must be limited to an objective analysis, asking what an objective insured might reasonably have believed (at the time that the policy was drafted) regarding the scope of coverage. See AISLI, 49 Cal.Rptr.3d at 13 . It is therefore irrelevant that Launch Media paid only one SIR, rather than ten SIRs, in the first policy period; that fact does not establish that the eleven losses theory is objectively unreasonable.

Finally, National Casualty asserts that J Records was not a different “person” from the other ten record labels because it was later acquired by BMG, which was an existing plaintiff in the Arista Records action. However, the trigger for policy coverage is the utterance of matter, and Liquid Dreams indisputably belonged to J Records, not BMG, when Launch Media first uttered it.

Ill

For the foregoing reasons, I conclude that Launch Media’s policy construction is more reasonable than National Casualty’s. However, as noted, we need not actually resolve that question. The duty to defend is triggered when an insured demonstrates the potential for coverage under a policy. In my view, Launch Media clearly has.

Therefore, I respectfully dissent.