Opinion
Franklin S. Chadwick and Charlene D. Mandel (homeowners) were the insureds under a broad peril homeowners insurance policy issued by Fire Insurance Exchange (FIE). They noticed cracking in their walls, the cause of which was found to be substandard design and construction of the house’s wall and floor framing. FIE denied coverage on the grounds the policy excluded loss from “cracking,” “latent defect” and “inherent vice.” Homeowners sued for breach of contract, breach of the covenant of good faith and fair dealing, and additional tort causes of action. The *1115superior court granted summary judgment for FIE, finding as a matter of law the loss was predominantly caused by latent defect or inherent vice. We reverse.
Facts
Summary judgment is proper if there is no triable issue of material fact and the moving party is entitled to judgment as a matter of law. (Code Civ. Proc., § 437c, subd. (c).) In reviewing that determination, we construe the moving party’s papers strictly and those of the opposing party liberally. (Howell v. State Farm Fire & Casualty Co. (1990) 218 Cal.App.3d 1446, 1448 [267 Cal.Rptr. 708].) Viewed in that light, the record discloses the following facts.
Homeowners noticed cracks in the interior sheetrock of their house, which according to the complaint was “new.” After the builder made cosmetic repairs, the cracks reappeared and homeowners were advised to obtain a professional investigation of the cause. They reported the loss to FIE.
FIE retained McGill-Martin-Self, a civil engineering firm, which in turn commissioned an investigation by Robert C. Kendall and Associates, consulting engineers. In addition to observing cracking and deflection visible on the surface, these engineers removed small portions of the sheetrock in homeowners’ house and larger areas in a neighboring home with the same design and similar damage.
The engineers retained by FIE found numerous deficiencies in the design and construction of the house’s floor and frame, including: lack of proper contact or connection between structural members; deviations from the framing plan, e.g., the use of a heavier header than called for; and deviation from the flooring plans, in particular the substitution of “gang-nailed floor trusses for 2 x 12 floor joists” creating an “unusual connection” with “no solid framing between the bearing wall above and the support beam below.” The engineers concluded these deficiencies provided insufficient support for the house’s floor and walls, allowing deflections and cracking of the sheet-rock.
Homeowners also retained a consulting engineer, Lee Mattis. Mattis agreed that many of the house’s problems were due to substitution of “Truswal floor trusses for 2 by 12 joists . . . .” His declaration continues: “My site investigation found additional construction deficiencies related to the substitution of the Truswal trusses: [<][] a) Proper provisions for shear transfer and blocking were not designed or installed. . . . Truswal addressed vertical loads and provided drawings with truss sizes and locations. *1116Fine print on these drawings states that lateral force bracing is to be designed by others.” Mattis also noted five “[p]oor framing practices” that were “plainly visible from the crawl space.” He concluded the contractor and developer had performed below the construction industry standard of care and homeowners’ loss was “primarily due to poor framing practices” associated with the use of floor trusses.
In their complaint, homeowners alleged they learned of the framing deficiencies only through expert inspection: “At no time prior to Robert C. Kendall and Associates and McGill-Martin-Self inspections had the visual cracking in the sheetrock at the home been such as to place a reasonable person on notice of such deficiencies.” In depositions, homeowners similarly stated they would not have been able to detect or recognize the framing deficiencies without expert assistance.
The policy alleged to have been in force at the time the loss manifested was the third edition of FIE’s “Protector Plus Homeowners Package Policy —California.” Under this policy, FIE insured against “accidental direct physical loss” to the dwelling, subject to stated exclusions. Exclusion 6 disavowed coverage for loss from, inter alia: “Wear and tear; marring; deterioration; inherent vice; latent defect; . . . cracking . . . of . . . walls, floors, roofs or ceiling; . . .”
The superior court found, “as a matter of law, that plaintiffs’ loss was predominantly caused by latent defect or inherent vice, and is therefore excluded by the terms of the insurance policy issued by defendant.”
Discussion
I. Efficient Proximate Cause
Homeowners’ primary contention on appeal is that the defective framing, even if deemed a latent defect or inherent vice, was merely the result of another, nonexcluded peril, to wit, “negligent construction,” which they argue was the efficient proximate cause of the loss. (Garvey v. State Farm Fire & Casualty Co. (1989) 48 Cal.3d 395, 403-404 [257 Cal.Rptr. 292, 770 P.2d 704].) We reject this argument because builder negligence, under the facts of this case, is not a peril distinct from the creation of the defective framing. (See Finn v. Continental Ins. Co. (1990) 218 Cal.App.3d 69, 72 [267 Cal.Rptr. 22].)
“As [the Supreme Court] explained in Garvey, supra, 48 Cal.3d at pages 406-407, the scope of coverage under an all-risk homeowner’s policy *1117includes all risks except those specifically excluded by the policy. When a loss is caused by a combination of a covered and specifically excluded risks, the loss is covered if the covered risk was the efficient proximate cause of the loss. (Id. at p. 402.) [The court] emphasized in Garvey, supra, 48 Cal.3d at page 408, that if third party negligence is not excluded under such a policy, it is a covered peril. As we stated, third party negligence under a homeowner’s policy is a ‘risk of physical loss’ under the policy. (Ibid.)” (State Farm Fire & Casualty Co. v. Von Der Lieth (1991) 54 Cal.3d 1123, 1131-1132 [2 Cal.Rptr.2d 183, 820 P.2d 285, A.L.R.5th 2129].)
California courts have consistently applied the efficient proximate cause analysis where two or more distinct actions, events or forces combined to create the damage. (E.g., Sabella v. Wisler (1963) 59 Cal.2d 21, 26, 31-32 [27 Cal.Rptr. 689, 377 P.2d 889] [negligent construction of sewer and inadequate compaction of fill (covered perils), both causing settling (excluded peril)]; Sauer v. General Ins. Co. (1964) 225 Cal.App.2d 275, 278 [37 Cal.Rptr. 303] [leakage of water from plumbing system (covered) and sinking of earth (excluded)]; Gillis v. Sun Ins. Office, Ltd. (1965) 238 Cal.App.2d 408, 419 [47 Cal.Rptr. 868, 25 A.L.R.3d 564] [windstorm (covered) and water (excluded)]; Premier Ins. Co. v. Welch (1983) 140 Cal.App.3d 720, 725 [189 Cal.Rptr. 657] [negligent installation of sewer (covered) and saturation of earth in heavy rain (excluded)]; Garvey v. State Farm Fire & Casualty Co., supra, 48 Cal.3d at pp. 412-413 [negligent construction (covered) and earth movement (excluded)]; Howell v. State Farm Fire & Casualty Co., supra, 218 Cal.App.3d at pp. 1459-1460 [fire (covered) and earth movement (excluded)]; State Farm Fire & Casualty Co. v. Von Der Lieth, supra, 54 Cal.3d at pp. 1127-1128, 1133 [third party negligence in failing to stabilize and dewater earth (covered) and earth movement and rising groundwater (excluded)]; Brian Chuchua’s Jeep, Inc. v. Farmers Ins. Group (1992) 10 Cal.App.4th 1579, 1580-1581, 1583 [13 Cal.Rptr .2d 444] [earthquake (covered) and leaking gasoline storage tank (excluded)].)
When, however, the evidence shows the loss was in fact occasioned by only a single cause, albeit one susceptible to various characterizations, the efficient proximate cause analysis has no application. An insured may not avoid a contractual exclusion merely by affixing an additional label or separate characterization to the act or event causing the loss. Thus, in Finn v. Continental Ins. Co., supra, a loss was caused by leakage from a broken sewer pipe and coverage had been denied under an exclusion for seepage or leakage. The insured did not show some additional force, act or event had broken the pipe, but simply argued the “break,” rather than the “leak,” was the efficient proximate cause. (218 Cal.App.3d at pp. 71-72.) The appellate *1118court rejected that argument, holding the hypothesized “break” was not conceptually distinct from the observed “leak”; rather, “ ‘leakage’ and ‘seepage’ necessarily imply some break or gap in the thing leaking.” (Id. at p. 72.) Consequently, the court held, the case “involved not multiple causes but only one, a leaking pipe.” (Ibid.)
In the case at bar, the undisputed evidence showed the loss (the cracking walls) resulted from the use of certain deficient framing techniques in construction of the house. As in Finn, supra, 218 Cal.App.3d 69, homeowners here point as an efficient proximate cause to no event, action or force other than the building of the house, but simply argue the defects were in turn the result of the builder’s negligence. Whether characterized as negligent, intentional or innocently inadvertent, the peril itself—the defective framing—is one and the same. To say builder negligence “caused” the defective framing is, in this context, to indulge in misleading wordplay, akin perhaps to saying a murderer’s malice aforethought “caused” the killing.
If every possible characterization of an action or event were counted an additional peril, the exclusions in all-risk insurance contracts would be largely meaningless. An earthquake, it could be said, was merely the immediate cause of loss and was itself the result of “changing tectonic forces,” a nonexcluded peril. Wear and tear on floorboards would be covered as the result of nonexcluded “friction.” An exclusion for freezing plumbing could be avoided by the simple observation the pipes would not have frozen absent “very low temperature,” a nonexcluded and, hence, covered peril.1
Homeowners’ reliance on the Supreme Court’s statement in State Farm Fire & Casualty Co. v. Von Der Lieth, supra, 54 Cal.3d at page 1132, that “third party negligence under a homeowner’s policy is a ‘risk of physical loss’ under the policy,” is unavailing. Von Der Lieth does not stand for the proposition that third party negligence is in every case a distinct cause of loss. Rather, the Von Der Lieth court discussed Finn v. Continental Ins. Co., supra, 218 Cal.App.3d 69 and distinguished it as involving “one cause of a loss” (54 Cal.3d at p. 1133), whereas Von Der Lieth itself involved at least two separate causes (one dependent on the other) that were “distinct in origin” (ibid.).
*1119Certainly evidence here suggested that use of the described framing materials and techniques in homeowners’ house was negligent. But if these deficiencies of construction are properly characterized as latent defects or inherent vices, homeowners may not escape the policy exclusion simply by showing their use was negligent. We turn, therefore, to the question whether undisputed facts show the defects were, as a matter of law, within the cited exclusions.
II. The Latent Defect and Inherent Vice Exclusions
Coverage was denied on the basis of exclusions for “latent defect” and “inherent vice.” The policy did not define these terms. As will appear, we conclude that when latency is defined by a “reasonable inspection” test, a triable issue of fact exists whether the deficient framing in homeowners’ house falls within these exclusions.
Historically, the latent defect and inherent vice exclusions were closely related to the restriction of property insurance to coverage of risks, i.e., fortuities. Under the “objective” theory of fortuity, a risk was thought to be something that happened to the property from an external cause, not something inherent or latent in the property; damage from such an inherent cause was considered, like wear and tear, a certainty and, hence, uninsurable. (Hecker & Goode, Wear and Tear, Inherent Vice, Deterioration, etc.: the Multi-Faceted All-Risk Exclusions (1986) 21 Tort & Ins. L. J. 634-635, 638-640.)
Thus, in Mellon v. Federal Ins. Co. (S.D.N.Y. 1926) 14 F.2d 997, 1002, the court, speaking of risks, quoted approvingly from an English case: “ ‘The expression does not cover inherent vice or mere wear and tear. ... It covers a risk, not a certainty; it is something which happens to the subject-matter from without, not the natural behavior of that subject-matter, being what it is, in the circumstances under which it is carried. . . .’” Later in the same decision, the court adopted the reasoning of other English cases that coverage of latent defects likewise was inconsistent with the concept of insurance, as lacking a fortuitous event—a casualty—causing the loss. (Id. at pp. 1003-1004; see also Goodman v. Fireman’s Fund Ins. Co. (4th Cir. 1979) 600 F.2d 1040, 1042 [“A loss is not considered fortuitous if it results from an inherent defect in the object damaged, from ordinary wear and tear, or from the intentional misconduct of the insured.”].)
The “objective” view of fortuity expressed in Mellon, supra, 14 F.2d 997 has largely been abandoned in favor of a “subjective” view, under which an event is fortuitous if it was contingent, i.e., unknown to the parties, at the *1120inception of the policy. (See Sabella v. Wisler, supra, 59 Cal.2d at p. 34; Compagnie des Bauxites v. Insurance Co. of N. Am. (3d Cir. 1983) 724 F.2d 369, 372.) Nevertheless, the original association of latent defects and inherent vices with the concept of objective fortuity led to a significant line of decisions defining latent defect and inherent vice narrowly, applying the terms to flaws in the substance of the insured property not only undetected, but undetectable.
In Employers Casualty Company v. Holm (Tex.Civ.App. 1965) 393 S.W.2d 363, 367, for example, the court stated the inherent vice exclusion “does not relate to an extraneous cause but to a loss entirely from internal decomposition or some quality which brings about its own injury or destruction.” In Mattis v. State Farm Fire & Cas. Co. (1983) 118 Ill.App.3d 612 [73 Ill.Dec. 907, 454 N.E.2d 1156, 1162, 41 A.L.R.4th 1082] the latent defect exclusion was held to refer to “some inherent defect in the materials used in construction which could not be discovered by any known or customary test »
Consistent with Mattis, supra, 454 N.E.2d 1156, numerous cases have defined a latent defect simply as one “which could not be discovered by any known and customary test.” (Plaza Equities Corp. v. Aetna Casualty and Surety Co. (S.D.N.Y. 1974) 372 F.Supp. 1325, 1331; Essex House v. St. Paul Fire & Marine Insurance Co. (S.D.Ohio 1975) 404 F.Supp. 978, 992; General American Transp. Corp. v. Sun Insurance Office, Ltd. (E.D.Tenn. 1965) 239 F.Supp. 844, 846, affd. per curiam (6th Cir. 1966) 369 F.2d 906; General Motors Corp. v. The Olancho (S.D.N.Y. 1953) 115 F.Supp. 107, 118, affd. per curiam (2d Cir. 1955) 220 F.2d 278.) Employing a narrow definition of latent defect, these decisions have held construction or design defects to be outside the exclusion because they could have been detected or recognized by proper testing or analysis.
In General American Transp. Corp. v. Sun Insurance Office, Ltd., supra, the collapse of a concrete structure under construction was found to be proximately caused by negligent welding in a steel supporting system. (239 F.Supp. at p. 845.) Employing the “known and customary test” definition of latent defect, the district court held the defect was not latent because it “could have been discovered with proper inspection.” {Id. at p. 846.) In its per curiam affirmance, the court of appeals agreed the faulty welding was not a latent defect or inherent vice because it “could have been discovered with proper inspection, such as radiography.” (369 F.2d at p. 908.)
Plaza Equities Corp. v. Aetna Casualty and Surety Co., supra, reached a similar result. A large metal sculpture of a phoenix fell through the concrete *1121of a multilevel shopping plaza.2 Although the collapse was attributed to the insureds’ failure to design and build sufficient structural support, the loss was held not to come within an exclusion for “ ‘inherent or latent defect.’ ” The loss was caused by a misjudgment in design, “not by any defect in the materials used nor in their installation. In any event, [the insurer] has not presented any evidence that this misjudgment could not have been discovered through the use of normal weight distribution and stress calculations.” (372 F.Supp. at p. 1331.)
In Essex House v. St. Paul Fire & Marine Insurance Co., supra, the court similarly held faulty design and construction of a brick wall were not “inherent or latent defect[s].” (404 F.Supp. at pp. 991-992, italics omitted.) The wall’s failure was attributed to negligent design and construction in several respects, including the placement of the windows and the use of undersized bolts. (Id. at pp. 985-986.) Following General American Transp. Corp. v. Sun Insurance Office, Ltd., supra, 239 F.Supp. 844, the Essex House court found the exclusion inapplicable because “[n]othing more than a simple visual inspection of the Essex House and the construction plans would have been sufficient to detect the shoddy building techniques employed therein.” (404 F.Supp. at p. 992.)
A deficiently designed and constructed retaining wall was likewise held not to be a latent defect in Mattis v. State Farm Fire & Cas. Co., supra: “Here there was no defect in the materials used, but in the design or construction of the north wall to withstand the force exerted by the consolidation of the backfill placed against it. [¶] Furthermore, the pleading failed to allege that the defect could not have been discovered by known scientific testing.” (454 N.E.2d at p. 1162; see also Ariston Air. & Cater’g Sup. Co., Inc. v. Forbes (1986) 211 N.J.Super. 472 [511 A.2d 1278, 1283] [“inherent or latent defect” to be interpreted as similar to accompanying terms “deterioration” and “wear and tear”; design and construction defects not within exclusion].)
At least two features of the approach taken in the foregoing line of decisions are material to the case at bar. First, the latent defect and inherent vice exclusions are restricted primarily to flaws or deficiencies in the materials used; faulty design or construction is deemed not to come within the exclusions. Second, under the “known and customary test” standard, the hypothetical inspection may include expert and scientific analysis. Thus, courts have held defects nonlatent because they could have been discovered by “known scientific testing" (Mattis v. State Farm Fire & Cas. Co., supra, *1122454 N.E.2d at p. 1162), “weight distribution and stress calculations” (Plaza Equities Corp. v. Aetna Casualty and Surety Co., supra, 372 F.Supp. at p. 1331) and “radiography” (General American Transp. Corp. v. Sun Insurance Office, Ltd., supra, 369 F.2d at p. 908).
FIE relies heavily on Acme Galvanizing Co. v. Fireman’s Fund Ins. Co. (1990) 221 Cal.App.3d 170 [270 Cal.Rptr. 405] (hereafter Acme Galvanizing), a decision of this division that was also cited in the trial court’s summary judgment ruling. In Acme Galvanizing, we noted a paucity of California case law on the latent defect exclusion and adopted a definition from Code of Civil Procedure section 337.15, subdivision (b), which sets out the limitations period for actions arising from latent construction deficiency: a defect is latent if it “is not apparent by reasonable inspection.”3 Applying this definition, we held a defective weld in an industrial kettle, which had been detected only by “ ‘destructive testing’ ” of the kettle, was latent because it was “not apparent upon reasonable inspection but only after a post-failure examination by an expert.” (221 Cal.App.3d at pp. 176, 178.)
FIE contends that when applied to homeowner policy exclusions, as here, the “reasonable inspection” formula adopted in Acme Galvanizing must be interpreted in the same manner as in the statute of limitations case law. Pursuant to sections 337.1 and 337.15, different time limitations apply for bringing an action based on a construction defect, depending on whether the defect is “latent” (10 years from completion) or “patent” (4 years). The distinction turns on whether the defect is “apparent by reasonable inspection,” in which case the defect is patent and subject to the four-year statute. (§ 337.1, subd. (e).) Case law holds the hypothetical inspection is that which the “average consumer” of the property would undertake “ ‘in the exercise of ordinary care and prudence.’ ” (Geertz v. Ausonio (1992) 4 Cal.App.4th 1363, 1367, 1370 [6 Cal.Rptr.2d 318].) Because an average homeowner would not have observed or recognized the structural defects in homeowners’ house without expert assistance, the defects, FIE argues, were latent as a matter of law.
FIE’s factual premise—that structural construction flaws are not apparent upon the inspection of an average homeowner acting without expert analysis or assistance—may be granted. Several considerations, however, persuade us the formula for determining what constitutes a latent defect (“not apparent by reasonable inspection”), when applied to homeowner policies, should not be interpreted so as to confine the postulated reasonable inspection to that of the average homeowner without expert assistance.
First, FIE’s interpretation ignores the significant line of decisions discussed above in which latent defects and inherent vices are restricted to a *1123much narrower class of deficiencies, those that could not have been discovered even by reasonable expert or scientific examination. At the least, these cases show the “reasonable inspection” component of latent defect and inherent vice has no precise definition in the legal language.
Second, Acme Galvanizing does not itself require the broad interpretation FIE suggests. Although the plaintiffs’ expert testified as to nondestructive methods for failure testing of welds, the welding defect in that case was in fact “ ‘not readily detectable upon reasonable inspection,’ ” but, rather, was discoverable only by “ ‘destructive testing.’ ” (221 Cal.App.3d at p. 176.) Consequently, Acme Galvanizing does not stand for the proposition that a defect discoverable by reasonable expert examination at the time a policy issues is necessarily latent. Moreover, the policy in Acme Galvanizing was for insurance of commercial property. Like the court in Opsal v. United Services Auto. Assn. (1991) 2 Cal.App.4th 1197, 1204 [10 Cal.Rptr.2d 352], “we question whether the arcane terms ‘inherent vice’ and ‘latent defect’ are sufficient to apprise a reasonable homeowner that the insurer is excluding coverage for property damage caused by third party negligence which results in a defect or condition not discoverable except by postdamage expert investigation.” (Italics added.)
Third, the considerations that have led to a broader definition of latent defect for statute of limitations purposes are inapplicable to determining the scope of a homeowners insurance exclusion. Under section 337.1, subdivision (a), a contractor is generally given complete protection from liability for a patent deficiency—even for personal injuries or death resulting from such a deficiency—four years after substantial completion of the improvement. As befits such a statutory scheme, in which the recovery of injured parties against the party who created the cause of the injury is completely cut off after a relatively short period, the cases have given “patent” a narrow application and its antonym, “latent,” a broad one.
Thus, under sections 337.1 and 337.15, even actual observation of a deficiency or its immediate effects will not make it patent if the average consumer would not have fully understood its physical cause or known it to be a deficiency. A design deficiency, for example, resulting in inadequate drainage of a deck was held arguably latent because the nonexpert owners and occupants could have reasonably thought better maintenance would prevent the periodic flooding (Geertz v. Ausonio, supra, 4 Cal.App.4th at pp. 1371-1372); a defect in an office building’s heating and air conditioning system was likewise held latent because the precise mechanical cause of the malfunction could not be determined (Baker v. Walker & Walker, Inc. (1982) 133 Cal.App.3d 746, 762-763 [184 Cal.Rptr. 245]). However sensible such a *1124broad view of latency may be in the statute of limitations context, it is inappropriate for interpreting an exclusion in an all-risk homeowners insurance policy.
Finally, the expansive view of latency urged by FIE would be inconsistent with the insured’s reasonable expectation that progressive damage manifesting in loss during a policy period is normally covered under that policy. (Prudential-LMI Com. Insurance v. Superior Court (1990) 51 Cal.3d 674, 699 [274 Cal.Rptr. 387, 798 P.2d 1230].) FIE contends, in effect, that no damage resulting from a preexisting defect is covered unless the defect was open and obvious to even untrained eyes. The average homeowner, inexpert in the details of home construction and structural engineering, would not recognize improper framing techniques, inadequate compaction of fill, or other serious structural deficiencies if he or she saw them. If for that reason these are necessarily considered latent defects, the exclusion would result in a lack of coverage for virtually all preexisting flaws in a home.4
FIE argues its interpretation is necessary so the insurer may “avoid concern about becoming the guarantor of the quality of construction of all insured property." If an insurer, however, were never to be responsible for losses caused by faults created before the policy issued, the “manifestation rule” (Prudential-LMI Com. Insurance v. Superior Court, supra, 51 Cal.3d at pp. 694-699) would make no sense; that rule holds the insurer responsible for claims on loss manifesting during the policy period, even though the cause may have been present, and the damage begun, before the inception of the policy.
Nor does a narrow definition of latent defect and inherent vice mean an insurer has no choice but to cover unknown preexisting defects. The insurer could, first, actually conduct the “reasonable inspection” upon which latency depends, including expert investigation and analysis where necessary. If this is thought too burdensome, the insurer could write into its homeowners policies an appropriate exclusion phrased in “ “clear and unmistakable language.” ’ ” (Reserve Insurance Co. v. Pisciotta (1982) 30 Cal.3d 800, 808 [180 Cal.Rptr. 628, 640 P.2d 764].) FIE has apparently chosen to pursue the latter approach. A subsequent edition of its “Protector Plus" policy eliminates the latent defect and inherent vice exclusions, but adds an exclusion for “[f]aulty, inadequate or defective planning, zoning, development surveying, *1125siting, design, specifications, workmanship, construction, grading compaction, maintenance repair materials, construction, remodeling, or maintenance of part or all of any property . . . .”
In addition to Acme Galvanizing, supra, 221 Cal.App.3d 170, FIE places its principal reliance on Carty v. American States Ins. Co. (1992) 7 Cal.App.4th 399 [9 Cal.Rptr.2d 1], Winans v. State Farm Fire and Cas. Co. (9th Cir. 1992) 968 F.2d 884, and Tzung v. State Farm Fire and Cas. Co. (9th Cir. 1989) 873 F.2d 1338.5 In Tzung, the court first held damage due in part to inadequate protection against soil expansion was excluded under a policy exclusion for “Faulty Materials or Workmanship.” (Tzung, supra, at pp. 1339-1342.) As an alternative basis for excluding the plaintiffs’ loss, the court then went on to consider briefly the “Inherent or Latent Defects” exclusion. Contrasting Essex House v. St. Paul Fire & Marine Insurance Co., supra, 404 F.Supp. at page 992 (visual inspection of building and plans would have revealed construction defects) with Merz v. Allstate Ins. Co. (W.D.Pa 1988) 677 F.Supp. 388, 389 (construction defects were all “unknowable, concealed by the ground”), the court held the two cases, read together, “suggest a general principle that defects in construction may constitute inherent or latent defects if the problems thus created are not readily discoverable.” (Tzung, supra, at p. 1342.) Because the design and construction defects at issue in Tzung—described as “imbedded in the ground”—were discoverable only through expert examination of the apartment building “and the soils beneath it,” they were not “readily discoverable.” (Ibid.)6
In light of Tzung's evident approval of Essex House, supra, 404 F.Supp. 978, we doubt it stands for the general principle that defects discoverable only by expert examination are per se latent. Insofar as the case may be so read, we reject the conclusion. As explained above, a standard that classes as *1126latent or inherent all defects whose discovery requires expert examination or analysis sweeps too broadly.
In Winans v. State Farm Fire and Cas. Co., supra, 968 F.2d at pages 886-887, the Ninth Circuit adopted Acme Galvanizing's formulation of the latency test (“not apparent upon reasonable inspection”) and held that pursuant to Acme Galvanizing and its own holding in Tzung, a defect is latent if it is discoverable only through an “intensive post-failure expert examination.” (Id. at p. 887.) After the plaintiffs had noticed cracks and separations in footings, slabs, walls and ceilings in their home, State Farm hired subsurface exploration experts to investigate the cause of the damage. In rejecting the plaintiffs’ assertion the experts had discovered the contractor’s negligence after their preliminary inspection, which consisted of a visual inspection and the digging of two shallow test holes, the court impliedly found that such a preliminary inspection would in fact meet the Acme Galvanizing and Tzung tests. (Id. at pp. 885, 887.)
Finally, in Carty v. American States Ins. Co., supra, 7 Cal.App.4th at pages 403-404, the court relied on both Tzung, supra, 873 F.2d 1338 and Acme Galvanizing, supra, 221 Cal.App.3d 170 in holding construction defects in the plaintiffs’ home, consisting of inadequate compaction of fill, inadequate reinforcement of the concrete slab and lack of anchoring of the foundation to bedrock, were latent because they were discovered only by “expert examination, including soil testing” and, hence, were “neither readily observable nor apparent on reasonable inspection.” (7 Cal.App.4th at p. 404.)
Unlike FIE, we do not read the foregoing cases as precluding any expert examination in determining what is a reasonable inspection; rather, for the reasons already set out, we believe a “reasonable inspection” for purposes of the latent or inherent defect exclusion may, under some circumstances, include appropriate expert assistance or analysis.
The undisputed evidence in this case was that much of the loss was due to the substitution of floor trusses for joists. Homeowners’ expert declared the construction plans reflected inadequate provision for shear transfer, blocking, and lateral force bracing in the use of these trusses. The same declarant noted five additional “[p]oor framing practices” visible from the crawl space. This testimony creates a triable issue of fact as to whether a reasonable expert inspection, undertaken at the time the policy issued, would have revealed the defects causing homeowners’ loss and, hence, as to whether the defects came within the exclusion for latent defect or inherent vice. The superior court therefore erred in granting summary judgment for FEE.
*1127Disposition
The judgment of the superior court in favor of FIE is reversed. Costs on appeal are awarded to appellants.
White, P. J., and Merrill, J., concurred.
Respondent’s petition for review by the Supreme Court was denied December 1, 1993. Panelli, J., Kennard, J., and Baxter, J., were of the opinion that the petition should be granted.