I dissent.
The trial court properly granted a directed verdict on the coverage issue whether we view this case as one of concurrent causation or one of successive causation. If the case is viewed as one of concurrent causation, coverage is established by State Farm Mut. Auto. Ins. Co. v. Partridge (1973) 10 Cal.3d 94 [109 Cal.Rptr. 811, 514 P.2d 123]. Neither the majority nor the *432concurring opinions dispute this conclusion; their position is that the case must be analyzed as a case of successive causation.
Even when the case is analyzed as one of successive causation, there is coverage as a matter of law. Under Sabella v. Wisler (1963) 59 Cal.2d 21, 31 [27 Cal.Rptr. 689, 377 P.2d 889], there is coverage if negligent construction, the insured cause, was the “efficient” proximate cause of the loss, “ ‘the one that sets others in motion.’ ” The majority so recognize. (Maj. opn., p. 403.) However, the majority erroneously conclude that if the earth movement, the excluded cause, was the “efficient” proximate cause there is no coverage under Sabella. (Ibid.)
Contrary to the majority opinion, Sabella does not establish that there can be no recovery in cases where the efficient cause is excluded and the immediate cause, the one which directly causes the loss, is insured. Sabella did not involve an exclusion of the efficient cause. Sabella merely points out that when the efficient cause is the excluded cause and the immediate cause is the insured cause the case is governed by Insurance Code section 532. A number of cases prior to Sabella permitted recovery when the efficient cause was excluded and the immediate cause was the insured cause. Sabella cited one with approval in discussing the Insurance Code section. (59 Cal.2d 21, 33.) It did not overrule these cases, and the majority should not overrule them with or without discussion. Under those cases we look to the language of the policy, construing it in accordance with ordinary insurance principles, and when this is done in the instant case, coverage is clear.
In Sabella, a sewer line ruptured due to negligence of a third party, the insured and efficient cause, and emptied waste water into loose fill, causing the house to settle. Settlement was the excluded cause and the immediate cause. Sabella held there was coverage and that if the policy were interpreted to exclude coverage it would violate Insurance Code section 530. (59 Cal.2d at pp. 33-36.)
Sabella should not be read as holding that in the converse situation where the efficient cause is excluded and the immediate cause is the insured cause there is no coverage. The issue was not directly presented. The court said that Insurance Code section 532 applied to that situation, and in this connection cited the landmark case of Pacific etc. Co. v. Williamsburgh City Fire Ins. Co. (1910) 158 Cal. 367, 372 [111 P. 4], (59 Cal.2d at pp. 33-34.) There were a number of cases in addition to Pacific where the efficient cause was the excluded cause and the immediate cause was the insured cause, and the courts found coverage. Sabella did not discuss or overrule them. While there was unqualified language in Sabella that the loss is to be attributed to the efficient cause, the language should be read in the light of the circum*433stances of the case, as applying to the issue before it, namely, when the efficient cause is the insured cause. The language should not be read as stating a rule applicable when the efficient cause is the excluded cause.
A number of cases have permitted recovery where the insured cause is the immediate cause and the excluded cause is the efficient cause. In the landmark case of Pacific etc. Co. v. Williamsburgh, supra, 158 Cal. 367, 372, where the insured cause was fire and the excluded cause was earthquake, the insured’s building burned as a result of fire caused in a nearby structure by the 1906 San Francisco earthquake. In allowing recovery under the policy, the court relied upon other cases involving fire caused by earthquakes and accidental death policies where illness, the excluded cause, caused the insured to drown or to fall in front of a moving train. (158 Cal. at pp. 373-376.) Although in all of the cases the efficient cause was the excluded cause, the courts had allowed recovery.
In Pacific etc. Co. v. Williamsburgh, the court started by pointing out the now well-settled rules that the policy will be construed to meet the reasonable expectations of the insured, the insuring clause will be broadly construed and exclusions strictly construed, and any ambiguity will be construed in favor of the insured. (158 Cal. at pp. 369-370; see also, e.g., State Farm Mut. Auto. Ins. Co. v. Partridge, supra, 10 Cal.3d at pp. 101-102; Sabella v. Wisler, supra, 59 Cal.2d at p. 30; Prickett v. Royal Ins. Co. Ltd. (1961) 56 Cal.2d 234, 237 [14 Cal.Rptr. 675, 363 P.2d 907, 86 A.L.R.2d 711]; Continental Cas. Co. v. Phoenix Constr. Co. (1956) 46 Cal.2d 423, 437-438 [296 P.2d 801, 57 A.L.R.2d 914]; Ransom v. Penn Mutual Life Ins. (1954) 43 Cal.2d 420, 424 [274 P.2d 633].)
The court analyzed Civil Code sections 2626 and 2628 (now Ins. Code, §§ 530 and 532) and reasoned that the two sections must be read together and that the latter section does not permit the insurer to claim an exemption broader than that specified in the policy, and that no “rule of construction would justify us in holding that the section broadens the scope of the original exception.” (Pacific, supra, 158 Cal. at p. 373.) The court then interpreted the language of the policy and allowed recovery.
Other cases allowing recovery by the insured when the efficient cause was an excluded cause and the immediate cause an insured cause include Wilson v. Travelers' Insurance Co. (1920) 183 Cal. 65, 67 [190 P. 366]; Kinsey v. Pacific Mutual Life Ins. Co. (1918) 178 Cal. 153, 155 [172 P. 1098]; State Farm Fire & Cas. Co. v. Kohl (1982) 131 Cal.App.3d 1031, 1035 et seq. [182 Cal.Rptr. 720]; Chambers v. Kansas Life Ins. Co. (1957) 156 Cal.App.2d 265, 267 et seq. [319 P.2d 387] and Tierney v. Occidental Life Ins. Co. (1928) 89 Cal.App. 779, 782 [265 P. 400].
*434When an insured looks at his policy and sees specific insured causes and numerous exclusions, he often may reasonably conclude that the insured causes and the excluded causes are mutually exclusive. As to many of the excluded causes, it may be difficult, if not impossible, to conceive of a situation where the excluded cause and the insured cause could combine to cause a loss. Thus he may reasonably conclude that the insurer out of an abundance of caution has sought to make clear that there is no coverage where an excluded cause alone causes the loss and never turn his mind to the rare possibility that the excluded cause and the insured cause might combine to cause a loss. For example, an insured purchasing a fire policy on his home believes that, if his home burns, he will be paid; he does not think that the 25 or 50 exclusions in his policy exclude 25 or 50 types of fires unless the policy makes it abundantly clear either by express statement or necessary implication.1 Accordingly, when the insured cause is the immediate cause of loss as in the fire illustration, coverage should exist unless the language of the policy makes it clear that coverage is excluded when the excluded occurrence causes the insured cause to in turn cause the loss.
This principle, which has long been applied to policies that specify the risk, such as fire insurance policies which were the predecessor of the instant all-risk policy, should be generally applied to all-risk policies where the risks insured are not specified individually but only as the risks not excluded. To adopt a rule less favorable to the insured would render the “all risk” policy misleading. By its terminology this type of policy purports to broaden coverage as compared to a specified-risk policy covering the same type of losses. Further, the incentive for the insurer to read the coverage as mutually exclusive of the exclusions is greater because the format of the policy means that if a cause is not excluded it is an insured cause, so that if the loss were caused solely by it there would be coverage. It is all the more reasonable for the insured to conclude that the purpose of the exclusion is to exclude losses caused solely by one or more of the excluded causes rather *435than a purpose to exclude cases of multiple causation when an insured cause occurs.
When we interpret the policy in the instant case under the settled principles applicable to interpretation of insurance policies, Wilson v. Travelers’ Insurance Co., supra, 183 Cal. 65 is controlling. There the policy provided for recovery for personal injury caused by train wreckage but included within the long list of exclusions “injury resulting wholly or partly from any of the following to wit: . . . explosives of any kind.” (183 Cal. at p. 67.) The insured was seated in a passenger car of a train when a violent explosion occurred in the toilet room, with parts of the passenger car causing injury to the insured. The court reasoned: “If we must look for the cause of the injuries to the first cause, it would never be the wreckage of the car which is itself, a result and not a cause.” (183 Cal. at p. 68.) The court went on to conclude that if the insured had been directly injured by a bomb placed in his lap which then exploded, the exclusion would have been effective to eliminate coverage but where the explosion caused parts of the wrecked car to injure the insured there was coverage under the established rule of construction requiring ambiguities to be construed in favor of the insured. (183 Cal. at p. 69.)
In Wilson, supra, 183 Cal. 65, the efficient cause was obviously the excluded cause, explosives, and the immediate cause was the insured cause, train wreckage. The court allowed recovery based on the language of the policy, contrary to the majority’s rule which would exclude coverage whenever the efficient cause was excluded. Further, Wilson points to the proper interpretation of the instant policy.
In the instant case, the policy states: “This policy does not insure against loss: . . . caused by, resulting from, contributed to, or aggravated by any earth movement, . . .” The policy also excludes “loss: . . . by . . . settling, cracking, shrinkage. . . .” The words refer to a “loss” related to the earth movement or other exclusions, and do not deal with a situation where the excluded causes do not themselves produce the loss but merely set in motion the insured cause which in turn causes the loss. The exclusion clause is ambiguous as to whether the excluded causes alone or in conjunction with other excluded causes limit coverage or whether coverage is also limited when the excluded cause sets in motion an insured cause which in turn causes the loss. An insured can reasonably conclude that only the former is excluded from coverage and that when the immediate cause is an insured cause there is coverage.2
*436It is true that the majority has adopted rules which provide symmetry— there is coverage if the efficient cause is insured but not if it is excluded. But we are not architects or engineers who often work towards symmetry. As judges, our goal is to search for justice. The basic rules governing interpretation of insuring clauses and exclusion clauses discussed above make it clear that insuring clauses and exclusions are not treated equally, and there is no justification for symmetry.
Moreover, if symmetry were a goal, the majority has it backward. Assuming that in multiple-cause cases we must choose between coverage where the insured cause is the efficient cause and where it is the immediate cause, common sense should tell us that there should be coverage when the immediate rather than efficient cause of loss is insured. An insured who has his house burn as a result of a fire from a nearby falling tree has a greater expectation that his fire insurance will provide coverage than he has when a fire causes the tree to fall crushing the house. But under the majority’s rules, an exclusion for loss by falling trees will mean that the fire policy will provide coverage when the house is crushed but not when it burns.
As Sabella points out, a number of earlier decisions had held that the insured would recover when, ’ as in that case, the efficient cause was the insured cause. A number of other earlier cases had recognized that the insured would recover when the efficient cause was an excluded cause where the immediate cause was an insured cause, depending on the language of the policy. The language of insurance policies is not ordinarily directed to the multiple-cause situation and rarely specifies expressly or by clear implication that there is no coverage when an excluded cause sets in motion an insured cause which in turn causes the loss. Two Court of Appeal cases had observed generally that it “has been held that when two causes join in causing an injury, one of which is insured against, the insured is covered by the policy.” (Zimmerman v. Continental Life Ins. Co. (1929) 99 Cal.App. 723, 726 [279 P. 464]; Hughes v. Potomac Ins. Co. (1962) 199 Cal.App.2d *437239, 244 [18 Cal.Rptr. 650] [disapproved on another point in Sabella v. Wisler, supra, 59 Cal.2d at p. 34].)
Today’s decision in effect repudiates long-settled principles of insurance policy interpretation without discussing the applicable case law. Applying those long-settled principles, the trial court properly concluded that there was coverage as a matter of law and ordered a directed verdict on the coverage issue. It is unfortunate that the majority do not discuss or distinguish the applicable cases, and accordingly they err in affirming the judgment of the Court of Appeal.
I would reverse the judgment of the Court of Appeal.
Cases of necessary implication in multiple-cause cases differ on the basis whether the policy insures specified risks or is an all-risk policy. When the policy insures specified risks, the exclusion will be effective in multiple-cause cases where the exclusion necessarily assumes existence of the insured cause. For example, in an automobile liability policy an exclusion of teenage drivers necessarily assumes that the insured cause, use of the automobile, is present. When the policy is an all-risk policy, the exclusion will be given effect when the asserted insured cause necessarily relates to the excluded cause. For example, when a homeowner’s comprehensive liability policy excludes coverage for injuries arising out of the use and ownership of motor vehicles, the insurer is not liable for causes of action based on negligent entrustment because the alleged insured cause, negligent entrustment, is necessarily related to the use and ownership of the vehicle. (Safeco Ins. Co. v. Gilstrap (1983) 141 Cal.App.3d 524, 527 et seq. [190 Cal.Rptr. 425]; see Hartford Fire Ins. Co. v. Superior Court (1983) 142 Cal.App.3d 406, 413 et seq. [191 Cal.Rptr. 37, 39 A.L.R.4th 189]; Allstate Ins. Co. v. Jones (1983) 139 Cal.App.3d 271, 275 [188 Cal.Rptr. 557]; State Farm Fire & Cas. Co. v. Camara (1976) 63 Cal.App.3d 48, 50 et seq. [133 Cal.Rptr. 600].)
The majority state that they do not overrule Pacific etc. Co. v. Williamsburgh City Fire Ins. Co., supra, 158 Cal. 367. Rather, the majority say, “We approve of the fundamental principles of insurance law discussed in Williamsburgh, we recognize that it would be unrealistic *436to discuss at length a case analyzing property policy language that is now obsolete.” (Maj. opn. at p. 403, italics added.)
I agree that no purpose would be served by analyzing the “obsolete” language of the Williamsburgh policy. But we should analyze the "modem” language of the exclusion clause in the policy before us under the fundamental principles of insurance law discussed in Williams-burgh. I have done so above. However, the majority do not look at the language of the exclusion but, contrary to the fundamental principles, merely adopt an automatic rule denying recovery when the immediate cause is insured and the efficient cause is excluded.
Both before and after Sabella, coverage has been held to exist in California decisions when the efficient cause was excluded. The majority cite no California case, before or after Sabella, which held that there was no coverage when the immediate cause of loss was the insured cause. In short, apart from the one ambiguous sentence in Sabella, there is no precedent in California for the majority’s rule, and every successive causation case where the immediate cause is insured has found coverage, contrary to today’s majority holding.