Opinion

I. Overview

In 1994, a lawsuit was brought against a surgery center and its owner, an oral surgeon, by a patient of the center for medical malpractice when, the year before, her surgery went horribly wrong, resulting in, among other things, vaginal bleeding. The patient’s bleeding was the result of the center’s having supplied the wrong pump and the wrong fluids for an operation to remove a polyp from her uterus. The center’s medical malpractice insurer defended that claim under a medical malpractice policy issued in 1993 and eventually paid the policy limits after litigation of the claim.

In 1996, another lawsuit was filed by the patient against the same surgery center and its owner for sexual battery arising out of the fact that the owner of the center had attempted to tamponade the patient’s vaginal bleeding during the operation that went awry. The medical malpractice insurer defended that claim too, but under the policy issued in 1993 and under which it had also defended the medical malpractice claim. The insurer continued to defend the claim until the limits of the 1993 policy were exhausted. However, it did not continue to defend the claim under the limits of a later policy, issued in 1996. The insurer asserted that the claim was “related” to the claim made in 1993, and, under the terms of both the 1993 and 1996 policies, the second claim stemmed from the same occurrence as the 1993 claim, and therefore was covered by the limits of only the 1993 policy.

The center and its oral surgeon owner then settled the 1996 suit with the patient out of the their own pockets, paying $250,000 for the battery claims, another $250,000 to settle a “potential” wrongful death action, though the patient remains alive to this day. They then turned around and sued the insurer for reimbursement of those sums, their attorney fees, and for bad faith, alleging that its decision to discontinue defending the second suit was in bad faith. The trial court entered a judgment in their favor for about $890,000.

On appeal, the primary issue before us is thus whether allegations of sexual battery and invasion of privacy arising out of the fact that the owner of a surgery center tried to stop a patient’s vaginal bleeding are “related” to the prior allegation of medical malpractice against the center itself for having supplied the wrong pump and fluids which caused the bleeding in the first place.

The question must be answered in the affirmative. The test for relatedness as the word “related” is used in malpractice insurance policies is, as the Supreme Court explained in Bay Cities Paving & Grading, Inc. v. Lawyers’ Mutual Ins. Co. (1993) 5 Cal.4th 854 [ 21 Cal.Rptr.2d 691 , 855 P.2d 1263 ], whether a claim is either logically or causally related to another claim.

In the case before us there is an undeniable causal relation between the particular sexual battery and invasion of privacy claims which the patient brought against the owner of the surgery center and her earlier malpractice claim. But for the earlier malpractice causing the bleeding in the first place, the owner of the center would never have touched her vaginal area in the effort to save her life, precipitating both claims. We therefore must reverse the judgment against the defendant insurer, which is predicated on the idea that the 1994 claims and the 1996 claims were not “related” under two successive medical malpractice policies.

II. The Underlying Cases and Claims

A. The 1994 Medical Malpractice Claim

In May 1993 Jennifer Louise Hamel went in for surgery, specifically the removal of a polyp from her uterus, at an outpatient surgery center owned by oral surgeon, Neil Friedman. Friedman provided to the surgeon in charge of Hamel’s surgery a particular kind of pump for the operation that was not appropriate for hysteroscopies, because its pressure is too high. Nor did he have the appropriate liquid on hand for the uterine distention. Instead of sorbitol and glycerine, there was only sterile water. Because the uterus has open blood vessels, the combination of the wrong pump and the wrong fluid meant that large amounts of water were sucked into Hamel’s vascular system, in turn causing disintegration of blood cells, electrolyte imbalance, and ultimately a massive pulmonary edema and cardiac arrest.

A malpractice case filed in 1994 resulted in a $9 million judgment (and that was after a $24 million judgment was reduced to present value), in which the jury apportioned fault 55 percent to the surgeon in charge and 45 percent to the outpatient center owned by Friedman. That judgment gave rise to an earlier appeal, in which the outpatient center argued that the nurse on hand who supplied the water (and who did not call out the nature of the fluid being supplied) was really the “borrowed servant” of the surgeon, and the jury should have been instructed to that effect.

This court, in an unpublished opinion, rejected that argument, because the outpatient center had proffered an altogether different theory at trial—that Hamel had suffered an allergic or toxic reaction. In fact, the outpatient center had twice acknowledged that it was responsible for the nurse’s errors. So, in January 1998, we affirmed the judgment entered against the outpatient center in May 1995. The very next month the insurer paid out its policy limits on its 1993 policy.

B. Friedman’s First Bad Faith Claim Against His Malpractice Insurer, for Its Handling of the Medical Malpractice Claim

Hamel’s 1994 suit against the outpatient center, interesting enough, might have been settled for much less money, a fact which was the genesis of a second lawsuit and appeal. Friedman and the outpatient center had a $1 million medical malpractice policy for 1993 with Norcal Mutual Insurance Company which required Dr. Friedman’s consent for any settlement. He took an active role in the litigation and was, as we noted in the unpublished opinion, “unquestionably the decision maker.” In late March and April he was still maintaining his innocence in the Hamel matter and wanted to fight to his exoneration. He yelled at the claims adjuster for even considering settlement. Thus when, in early April 1995, Hamel’s attorney submitted a settlement limits demand ($1 million, which was the limit for Norcal’s 1993 malpractice policy), Friedman “said without hesitation that he would refuse to settle.” He ignored his attorney’s advice to take the offer.

In May 1995, after the chance to settle for policy limits expired, it turned out that Friedman had withheld information about the pump which the outpatient center had supplied. The pump actually had a warning label saying “for laparoscopy procedures only.”

That month other disturbing new information came to light. Hamel’s EKG strips and arterial blood gas records were discovered in a desk drawer, separate from her medical file. Those records had not been produced during discovery, even though Dr. Friedman and the outpatient center had claimed to have turned over all records in their possession.

Even worse, it became known that the nurse who supplied the water instead of the proper fluid was alleging that Friedman had tried to get her to alter Hamel’s medical records. She disclosed the fact that Friedman himself had operated the pump in which he used the water that ultimately led to Hamel’s pulmonary edema.

When these facts were revealed, it became impossible to settle the case for policy limits. Thus when Friedman sued his malpractice carrier for bad faith, this court readily affirmed a judgment in the insurer’s favor. No matter what else the insurer had done, it was Friedman, not Norcal, who had thrown away the opportunity to settle for policy limits.

C. The 1996 Sexual Battery Claim

During the operation in May 1993, Friedman noticed that Hamel was bleeding from her vagina and attempted to tamponade the vaginal bleeding. That entailed putting surgical gauze into her vagina and applying pressure to her abdomen for two or three minutes. In May 1996, Hamel sued Friedman and the outpatient center again, this time for battery, sexual battery and invasion of privacy.

Hamel’s complaint was not based on any purported sexual assault by Friedman. He was not alleged, for example, to have touched her “inappropriately” in the course of, say, an examination prior to the polyp removal, or to have touched her for any sexual purpose.

Rather, the theory of the new suit was that Friedman had no business being in the operating room and touching her at all. Thus when he touched Hamel during and after the operation that went awry, it was ipso facto battery, regardless of whether he had a good reason for doing so at the time. And, because he had touched Hamel in the vaginal area in order to stop the bleeding, Hamel’s attorney simply added causes of action for sexual battery and invasion of privacy along with the one for ordinary battery.

This new lawsuit lavished attention on the medical malpractice which had been the subject of the earlier litigation and Friedman’s failed attempt to cover it up: The first several pages reiterated the basic facts that Hamel suffered a variety of injuries as the result of Friedman’s basic stinginess in trying to scrimp on fluids and in not supplying the right pump. Moreover, by way of a cause of action for spoliation of evidence, the complaint elaborated on Friedman’s attempts to cover up his role in the May 1993 disaster by hiding medical records and instructing his nurse to alter them. In that context the complaint included allegations of battery, sexual battery, and invasion of privacy based on the fact that Friedman himself simply should not have been present in the operating room at all, and had no right to touch her.

D. Friedman’s Second Bad Faith Claim Against His Malpractice Insurer, for Its Handling of the Sexual Battery Claim

When Hamel’s attorney filed the second action, the judgment from the first action was still in the process of being appealed. Our decision on the case would not be filed until, as mentioned above, January 1998. The policy limits on the 1993 policy had not been paid yet, so Norcal agreed to continue paying for Friedman’s and the outpatient center’s defense of the second action under a reservation of rights. However, when the limits on the 1993 policy were paid by September 1998, Norcal withdrew from the defense of the second lawsuit against Friedman.

It was that decision which led to the present appeal. After the withdrawal, Friedman and the outpatient center settled the second lawsuit for $250,000, and settled a “potential” wrongful death action (though Hamel remained alive as of the time this case was briefed and presumably remains alive to this day) for another $250,000.

In May 2000, Friedman and the outpatient center sued Norcal for bad faith. Their theory was that the 1996 policy which Norcal had issued the outpatient center applied to the second lawsuit. Thus the insurer acted in bad faith by withdrawing when it had paid the policy limits on its earlier policy, and when it refused to indemnify the outpatient center for the $250,000 settlements.

This lawsuit went more favorably for Friedman and the outpatient center. The trial was bifurcated, with the issue of coverage under the 1996 policy being tried to the court first. In that phase the trial judge determined that the word “occurrence” in the 1996 policy (as in a “claim” is “any occurrence . . . which may result in a . . . lawsuit”) did not include the sexual battery claim, so that it wasn’t within the purview of the 1993 policy and its $1 million limit. The trial judge thus ruled that Norcal had a duty to defend under the 1996 policy. The bad faith case then went to the jury, and the result was a judgment for $889,327.89. (The sum was made up of $500,000 for the two $250,000 settlements, attorney fees to defend the second suit, attorney fees to bring the bad faith suit, and prejudgment interest.) From that judgment Norcal now appeals. Friedman himself is no longer in the case for reasons not relevant to this opinion.

III. The Insurance Policy Language

For our purposes, the salient language is the identical definition of “occurrence” found in both the 1993 and 1996 policies. An occurrence is a “single act or omission or series of related acts or omissions involving direct patient treatment. ” (Italics added.)

The significance of the definition of occurrence is this: Both the 1993 and 1996 policies were “claims made” policies, which defined “claim” to mean either an “actual claim or suit” or a “potential claim or suit,” and a potential claim was defined as any “occurrence . . . which may result in a . . . lawsuit.”

Thus if the 1996 sexual battery suit arose out of the same occurrence as the 1994 medical malpractice suit, then only the 1993 policy could cover it.

There is no dispute that on May 24, 1993, i.e., during the 1993 policy, Friedman gave Norcal notice of a potential claim made, namely that Hamel would likely bring a lawsuit based on the surgery that went awry on May 21, 1993. The telephone report notes of the conversation show that Friedman told Norcal that Hamel had gone into “arrest.” For its part Norcal recognized that the event would give rise to a lawsuit, and quickly advised Friedman not to say anything to anyone other than a Norcal claims representative or place its correspondence in the patient’s file.

IV. Discussion

A. Claims Can Be “Related” Without Being Subject to a Res Judicata Defense (at Least in California)

To avoid confusion at the outset, we must address a topic which, at first, may seem off-point, but which turns out (as will become clear in part IV.B. of the opinion, below) to go to the heart of the case: Why, if Hamel’s 1996 sexual battery suit was supposedly “related” to her 1994 medical malpractice suit, wasn’t the 1996 suit subject to a res judicata defense? If the 1996 suit was not precluded by res judicata, then perhaps it can indeed be said to constitute a different “claim” than the one made in 1994, and the insurer erred in not ascribing the 1996 suit to be a “claim” made on the 1996, as distinct from a claim made on the 1993 policy. In fact, one of Friedman’s primary arguments in this appeal is that because the trial court, in rejecting the insurer’s summary judgment motion based on res judicata, “found” that the second lawsuit was “different” and “separate” from the first lawsuit, the idea that both are related to the same occurrence reported on May 24, 1993 is “factually incorrect.”

Not so at all. Actually, in the majority of jurisdictions, including the federal courts, the 1996 suit would have been subject to a successful res judicata defense, and this appeal would probably not be before us now, since the defense still might have been successfully asserted while there was still some money on the 1993 policy meter. That is because the majority use the “same transaction” test for res judicata.

Federal courts, and the majority of jurisdictions in the United States, use a test for what a claim is under the doctrine of res judicata articulated in the Restatement Second of Judgments, section 24. The Restatement first defines res judicata, in section 18, from the point of view of a plaintiff: The rule is that a valid and final personal judgment in the plaintiff’s favor means the plaintiff cannot thereafter maintain an action on “the original claim or any part thereof.” Then, in section 24, subdivision (1) the Restatement defines “claim” to mean “any part of the transaction, or series of connected transactions, out of which the action arose.” In section 24, subdivision (2), a “transaction” is defined as something “determined pragmatically, giving with to such considerations as whether the facts are related in time, space, origin, or motivation, whether they form a convenient trial unit, and whether their treatment as a unit conforms to the parties’ expectations or business understanding or usage.”

Under the Restatement “same transaction” criteria, there can be no doubt that both the 1994 medical malpractice suit and the 1996 sexual battery were the same claim. The malpractice and batteries were literally part of the same series of interconnected events that took place in the surgery room on a particular day in May 1993. They were related in time and space: The sexual touching happened either during the botched polyp removal—or, if one wants to split hairs—in the immediate aftermath of the botched polyp removal. They were related in origin: The origin of the touching was the need to stop Hamel’s bleeding, which had itself been caused by the use of the wrong pump and solution. They were related in motivation: The touching was prompted by Friedman’s attempts to ameliorate the immediate and life-threatening effects of the malpractice, which had been the result of Friedman’s (alleged) decision to scrimp on the fluids and use an existing pump rather than obtain a proper one. And did the malpractice and battery form a convenient unit? Hamel entered the surgery expecting the operation to go well and not to be touched by anyone she had not already consented to touching her. Friedman supplied the wrong solution and equipment—indeed, he was in the room operating the pump—and he only touched her because (a) he was already in the room (as the 1996 complaint asserted, he was there “in his capacity as owner, operator . . . .”); (b) she was bleeding; and (c) he, as an oral surgeon present at the site because he was operating the pump (ironically the very pump which, because its pressure was too high, had caused the bleeding in the first place), presumably knew something about how to stop bleeding. (And, indeed, at the coverage trial here, Friedman would later testify that was precisely why he touched her.) It would be hard to imagine a given set of facts which didn’t form a more “convenient trial unit.”

Hamel’s second lawsuit was not subject to a successful res judicata defense, however, because California uses a different test for claim preclusion, namely the primary right test. The test has its origin in 19 th century joinder rules. (See generally, Heiser, California’s Unpredictable Res Judicata (Claim Preclusion) Doctrine (1998) 35 San Diego L.Rev. 559, 571-577 (Heiser article); cf. McCarty v. Fremont (1863) 23 Cal. 196, 197 [ 23 Cal. 197 ] [observing that a complaint which united causes of action for personal injury and property damage into one count was a clear violation of old Practice Act’s joinder rules].) If one needs a quick handle on the test, the core concept is harm suffered. (See Agarwal v. Johnson (1979) 25 Cal.3d 932, 954 [ 160 Cal.Rptr. 141 , 603 P.2d 58 ] [“the significant factor is the harm suffered, and that the same facts are involved in both suits is not conclusive”].) While the test has come in for some academic criticism for being harder to apply and understand than the federal transaction test (see generally, Heiser article; see also Comment, Res Judicata: Should California Abandon Primary Rights (1989) 23 Loyola L.A. L.Rev. 351), as recently as two years ago our high court declined the invitation to abandon the primary right theory and adopt the Restatement approach (though there because the result in that particular case would have been the same under either test). (See Mycogen Corp. v. Monsanto Co. (2002) 28 Cal.4th 888, 909, fn. 13 [ 123 Cal.Rptr.2d 432 , 51 P.3d 297 ].)

Here, we have a virtually textbook case illustrating the different operations of the two approaches to res judicata. As shown above, clearly the malpractice and battery were part of the same transaction, but gave rise to different harms. The harm from the medical malpractice was the bodily injury resulting from an operation in which the wrong equipment and fluids were used. That harm stemmed from the violation of the right to be free of negligence in connection with the operation. (Cf. Bay Cities, supra, 5 Cal.4th at p. 860 [primary right of client in legal malpractice suit was to be free of attorney negligence in connection with particular debt collection matter].) By contrast, the harm from the battery, sexual battery, and invasion of privacy torts was a harm to Hamel’s dignitary and privacy interests in being touched by a person that she never consented to have touch her, even, as was the case here, for a good purpose. (Cf. Sanchez-Scott v. Alza Pharmaceuticals (2001) 86 Cal.App.4th 365, 377-378 [ 103 Cal.Rptr.2d 410 ] [complaint properly stated cause of action for invasion of privacy because patient never consented to an examination of her breasts in examination room by unidentified male drug salesperson].)

But just because the harms sought to be compensated in the 1994 and 1996 suits were different (and hence there was no res judicata under the California test) does not mean the suits did not arise from related claims. We must remember that it is the actual language of the insurance contract and not the common law doctrine of res judicata which governs this case. (See Waller v. Truck Ins. Exchange, Inc. (1995) 11 Cal.4th 1, 18 [ 44 Cal.Rptr.2d 370 , 900 P.2d 619 ] [“look first to the language of the contract”]; Montrose Chemical Corp. v. Admiral Ins. Co. (1995) 10 Cal.4th 645, 677 [ 42 Cal.Rptr.2d 324 , 913 P.2d 878 ] [“The precise question, of course, is what result follows under the language of the policies of insurance to which the parties agreed, including the standardized definitions that were incorporated into those policies.”].)

B. Under Bay Cities, the Test for the Word “Related” as Used in Standardized Claims Made Malpractice Policies Is Either Logic or Causation

Our task is made considerably easier because the Supreme Court has already examined the meaning of the word “related” in the context of a claims made malpractice policy. That examination was in Bay Cities, supra, 5 Cal.4th 854 .

It is important to realize that Bay Cities is an opinion with two distinct parts, and one cannot stop reading after finishing the first part. The case arose out of a lawyer’s two separate acts of negligence in the course of trying to collect money owed a client who was a general contractor. First, the lawyer didn’t serve a stop notice on the project. Second, he didn’t file a complaint to foreclose the mechanic’s lien that he did file. The question came to our Supreme Court in a context similar to the one at bar, i.e., whether the malpractice policy’s “per claim” limit applied to both acts of negligence.

Interestingly enough, the first part of the opinion deals with whether there was one “claim” or two claims by reference to the primary right doctrine which we have already discussed. The intermediate appellate court had held that the two acts of negligence presented two different claims, but the high court said no: The client had but “one primary right”—the right, as we alluded to above, to be free of negligence in the collection of the debt. (See Bay Cities, supra, 5 Cal.4th at p. 860 .) He breached that right “in two ways”—by not serving a stop notice and by not filing a timely complaint to foreclose the mechanic’s lien—“but it nevertheless remained a single right.” (Ibid.) The court went on to say that the client had suffered but “a single injury” for which it cited the primary rights discussion in Slater v. Blackwood (1975) 15 Cal.3d 791 [ 126 Cal.Rptr. 225 , 543 P.2d 593 ], a case which is one of the pillars of the California common law on res judicata. (Indeed, most of the citations to Bay Cities by intermediate appellate courts since the opinion was handed down about 11 years ago are in the context of discussing issues related to res judicata.)

The Bay Cities court then had some fun with the implications of the insured attorney’s position that there had been two claims which had been presented. First, it was wholly anomalous that “the loss of or damage to a single right could give rise to more than one claim under an attorney’s professional liability policy.” (Bay Cities, supra, 5 Cal.4th at p. 861 .) And then the court noted that under the attorney’s view, “the greater the number of an attorney’s negligent acts, the greater the number of claims under the policy, even if all the acts cause only a single injury. Such a rule would have the plainly undesired result of providing the attorney who had made one error with an incentive to then make as many additional errors and omissions as possible, so as to increase the amount of insurance coverage.” {Ibid.)

If one were to stop reading Bay Cities with the conclusion of part I, one might be tempted to conclude in the case before us that where there are two different primary rights involved, there are two separate claims. But it does not follow that just because two acts give rise to only one claim because they involve only one primary right—which is what the first part of Bay Cities certainly does hold—that two acts which involve different primary rights necessarily give rise to two claims. The first part of Bay Cities cannot be read for that proposition. And, probably to ensure against such a misreading of the opinion, the Bay Cities court added a second part to its opinion, in which it specifically delineated the criteria for the relatedness of two separate claims. (See Bay Cities, supra, 5 Cal.4th at pp. 866-873.) Justice Kennard, in her concurring opinion, objected to the inclusion of this part of the opinion as unnecessary. (See id. at pp. 873-876 (conc. opn. of Kennard, J.)) She was also troubled by the substance of the majority’s discussion which had concluded that the language “related acts, errors or omissions” was “inherently unambiguous.” (See id. at p. 876.)

Be that as it may, six justices signed the opinion, including its second part, which directly tackled the problem of the meaning of the word “related.” In that part the court explicitly made the assumption that each separate omission (the failure to file a stop notice, the failure to file a complaint to foreclose the lien) did indeed give rise to a separate claim. Even so, the court would hold that the two claims were related to each other such that the per claim limit applied to both acts.

Much of the discussion centered on the insured’s assertion that the word “related” was ambiguous, and should be construed against the insurer. The argument for ambiguity seized on the idea that the word “related” denotes both causal and logical relationships. Because the two acts of negligence by the attorney in Bay Cities were not causally related, only logically related, the attorney thought he should get the benefit of the two choices. If there was no causal relationship between the two errors, then the two omissions were not “related,” even though there might be a logical relationship. (See Bay Cities, supra, 5 Cal.4th at p. 871 .)

The Supreme Court rejected that argument. (See Bay Cities, supra, 5 Cal.4th at pp. 866-873.) Most of the discussion from pages 866 through 872 of the opinion was devoted to demonstrating that, understood in its ordinary sense, the causal and logical aspects of the word “related” are not mutually exclusive. The court summarized its rule this way: “We agree with the court in Gregory [v. Home Ins. Co. (7th Cir. 1989) 876 F.2d 602 ] that the term ‘related’ as it is commonly understood and used encompasses both logical and causal connections.” (Bay Cities, supra, 5 Cal.4th at p. 873 .)

The bottom line was that, despite the lack of a causal relationship, the attorney’s two omissions were related, and in “multiple respects.” (Bay Cities, supra, 5 Cal.4th at p. 873 .) They arose out of the same transaction, involved the same client, were committed by the same attorney, and resulted in the same injury, i.e., loss of the debt. (Ibid.)

In the present case, there can be absolutely no doubt that the battery and invasion of privacy claims were causally related to the malpractice claim. Friedman was present in the room, operating the very pump which he had supplied, and which because he had supplied the wrong pump and the wrong fluids, there was an intake of water which caused the bleeding that he tried to stop, and that attempt to stop the bleeding itself constituted the battery claims. It is hard to imagine separate acts which are more clearly causally related.

The only thing that differentiates this case from Bay Cities is that the two acts gave rise to different injuries, which is explainable because Bay Cities involved a case of clear logical relationship whereas here we deal with a case of clear causal relationship. However, as we noted above, this is not a case involving California’s doctrine of res judicata but a case involving the language of an insurance contract.

Homestead Insurance Company v. American Empire Surplus Lines Insurance Company (1996) 44 Cal.App.4th 1297 [ 52 Cal.Rptr.2d 268 ] is inapposite. Because the case involved a dispute between two insurers on successive years of risk, the court expressly found it unnecessary to analyze whether the first lawsuit and the second lawsuit arose from “ ‘interrelated’ ” acts. (Id. at p. 1306.)

To the degree that Homestead can be read for the blanket proposition that no claim made after the expiration of a claims made policy can ever be ascribed to that policy because the definition of claim is “subordinate” to the insuring clause promising to pay any claim made during the policy period (see Homestead, supra, 44 Cal.App.4th at p. 1305 [theme that to be covered a claim must be made during the policy period] and p. 1306 [dicta that “insured must still receive notice” of claims during policy period]) we must respectfully part company with it.

Courts must interpret insurance policies, like all contracts, to try to give effect to every clause and harmonize the various parts with each other. (See ACL Technologies, Inc. v. Northbrook Property & Casualty Co. (1993) 17 Cal.App.4th 1773, 1785 [22 Ca. Rptr. 2d 206] [collecting authorities to the effect that terms in insurance policies should not be rendered either redundant or surplus].) Courts must try to give effect to “potential claim” clauses, harmonizing them with other parts of the policy, not just ignore them by saying that if an insured does not receive notice of a particular claim during a policy period it is automatically not covered by that policy. The Supreme Court did not ignore or read out the “interrelated” language in Bay Cities on the theory that the policy was a “claims made” policy and we are not about to do it now in this case.

Here, the “potential claim” language serves the insured by keeping all related claims within the coverage of an existing policy (the first one against which the first claim is made); a contrary rule (i.e., “forget the potential claim” language, if a claim is not presented to the insured within the policy period then it can never be covered by that policy) would ill-serve both insureds and insurers. If “related” claims that come after a policy period ends can still be covered by an earlier policy (the way that Norcal did here until the limits ran out), then insurers can confidently write renewal policies even in cases like Friedman’s where the insured has been hit with a claim—even, as in this case, a very bad one. Ignore the language, and malpractice insurers have a strong incentive to drop every insured who is unfortunate enough to have any claim made against him or her during the policy period. That incentive effect is of particular significance in California where our primary right approach to res judicata makes it (as the instant case well shows) relatively easy for third party claimants to file new (but related) lawsuits alleging new harms after a malpractice policy is renewed.

C. There Was No Possibility of Coverage

The “potentiality” rule in insurance coverage law is extremely well covered territory indeed. We will simply refer the reader to Buss v. Superior Court (1997) 16 Cal.4th 35, 45-49 [ 65 Cal.Rptr.2d 366 , 939 P.2d 766 ], Montrose Chemical Corp. v. Superior Court (1993) 6 Cal.4th 287, 295-296 [ 24 Cal.Rptr.2d 467 , 861 P.2d 1153 ], and Horace Mann Ins. v. Barbara B. (1993) 4 Cal.4th 1076, 1081 [ 17 Cal.Rptr.2d 210 , 846 P.2d 792 ] as authoritative compilations of the relevant rules.

The determination of whether there is such a potentiality is made by looking at “facts alleged or otherwise disclosed.” ( Buss, supra, 16 Cal.4th at p.46.) Those facts are thus taken from the complaint and facts extrinsic to the complaint “ ‘known by the insurer at the inception of a third party lawsuit.’ ” (See Montrose Chemical Corp. v. Superior Corp., supra, 6 Cal.4th at p. 295 , quoting Gray v. Zurich Insurance Co. (1966) 65 Cal.2d 263, 276 [ 54 Cal.Rptr. 104 , 419 P.2d 168 ].)

On the other hand, while the universe of facts bearing on whether a claim is potentially covered includes extrinsic facts known to the insurer at the inception of the suit as well as the facts in the complaint, it does not include made up facts, just because those facts might naturally be supposed to exist along with the known facts. An insured is not entitled to a defense just because one can imagine some additional facts which would create the potential for coverage. Thus in Westoil Terminals Co., Inc. v. Industrial Indemnity Co. (2003) 110 Cal.App.4th 139, 154 [ 1 Cal.Rptr.3d 516 ], the court refused to consider the scenario of what would have happened if the insured had not leased out a facility to the third party complainant, but instead would have kept operating it (which would have assumed that pollution spills might have occurred during a time when such spills would trigger a duty to defend). And in Hurley Construction Co. v. State Farm Fire & Casualty Co. (1992) 10 Cal.App.4th 533, 538-539 [ 12 Cal.Rptr.2d 629 ], the court refused to consider the idea that a third party complaint charging a repair services company with conspiracy to fraudulently bill (ironically, it was another insurer who said it had been fraudulently billed) might be amended to also allege property damage or bodily injury. As the court summed its rule: “The insured may not speculate about unpled third party claims to manufacture coverage.” (id. at p. 538; see also Gunderson v. Fire Ins. Exchange (1995) 37 Cal.App.4th 1106, 1114 [ 44 Cal.Rptr.2d 272 ] [quoting Hurley]; Croskey, et al., Cal. Practice Guide: Insurance Litigation (The Rutter Group 2003) ¶ 7:577, p. 7B-21.)

Here, the facts shown in the third party complaint in 1996 established from day one that the allegations of battery, sexual battery, and invasion of privacy were predicated on the idea that Friedman should not have been present at all in the operating room during the surgery in May 1993, and his touching of Hamel was a battery and invasive of her privacy because it was done without her consent. There have been no allegations, or facts ever presented to or learned by the insurer, which even hint that Friedman touched Hamel for his own sexual gratification. (A fact corroborated by Friedman’s own testimony at the coverage trial.) There thus was never any potential that the 1994 and 1996 suits were unrelated, even assuming (for sake of argument) that if Friedman had touched Hamel for purposes of sexual gratification that the claims would not be related. There never was even the potential that the 1996 suit would have been “covered” under the 1996 policy.

Friedman’s argument that merely because he can make a plausible argument for coverage (the argument is that because a “claim” was made during a “claims” made policy there was a potential for coverage) is frivolous. The potentiality rule for the duty to defend is pegged to the possibility of actual indemnity coverage, not the mere existence of a plausible argument. As Buss makes clear, “Both the duty to indemnify and the duty to defend are in fact dependent on coverage—the former on actual coverage, the latter on at least potential coverage.” (Buss, supra, 16 Cal.4th at p. 46, fn. 10 .) Thus “in an action wherein none of the claims is even potentially covered, the insurer does not have a duty to defend.” (Id. at p. 47.) You don’t prove an insurer has a duty to defend merely by making a good argument for potential coverage, you show it by demonstrating a potential for coverage under the terms of the actual policy.

V. Disposition

Because there was no possibility of coverage under the second suit, Norcal’s decision to discontinue its defense when it had paid the limits on its 1993 policy was within its rights under its two insurance policies. It was thus error for the trial court to rule that there had been a breach of contract in the liability phase of the trial. The judgment is reversed with directions to enter a new judgment in Norcal’s favor.

Appellant Norcal shall recover its costs on appeal.

Moore, J., and Aronson, J., concurred.

A petition for a rehearing was denied July 21, 2004, and respondent’s petition for review by the Supreme Court was denied October 13, 2004.

The invasion of privacy claim was functionally identical in its basis to the battery and sexual battery claims. That claim was not based, for example, on the contention that Friedman was just some stranger wandering around a medical area who happened to blunder into an intimate medical procedure. The 1996 complaint cannot be read that way. Rather, the 1996 complaint set the context for the invasion of privacy claim as follows: First, in the general allegations, the complaint asserted that Hamel had only consented to a diagnostic procedure, and had not consented to a “resectoscope” procedure. The “resectoscope” procedure was thereafter defined in the complaint as the “Surgical Procedure.” Next, still in the general allegations, the complaint made the point that “Friedman, in his capacity as owner, operator, landlord and representative of the owner of The Out-Patient Surgery Center, instructed staff of The Out-Patient Surgery Center to use sterile water in said Nezhat Pump rather than sorbitol” and gave the instruction because he was too cheap to use sorbitol: “Defendant Friedman gave this instruction for use of sterile water when operating said Nezhat Pump for cost-cutting purposes and as an economic measure without the knowledge or approval of Plaintiff’s treating physician.” It was then “during and after performance of the [unconsented to] Surgical Procedure” that “defendant Friedman placed his hand inside Plaintiff’s vagina and placed his other hand on or about other parts of Plaintiff’s body.” Finally, under the specific heading for the invasion of privacy cause of action, the complaint’s key charging paragraph simply repeated the words just quoted about putting his hand inside Hamel’s vagina and his other hand on other parts, and asserted that in performing those acts Friedman was doing something he was not licensed to do and Hamel had a reasonable expectation of being free of those acts. Accordingly, unless there is some special reason to specifically mention the invasion of privacy claim, from here on out when we mention Hamel’s “battery” or “sexual battery” claim or lawsuit, we necessarily include within those terms her invasion of privacy claim. This case does not involve any issues concerning the thorny problem of an insurer’s obligations under so-called wasting or cannibalizing insurance policies, where defense costs are charged against the policy limits. (See generally Munro, Defense Within Limits: The Conflicts of ‘Wasting’ or ‘Cannibalizing’ Insurance Policies (2001) 62 Mont. L. Rev. 131 .) We are also spared issues that might arise in general liability policies (which are not cannibalizing policies) as to precisely when in the underlying litigation an insurer, once having an obligation to defend fastened on it, may properly withdraw consonant with the insured’s rights. (See, e.g., Jenkins v. Insurance Co. of North America (1990) 220 Cal.App.3d 1481 [ 272 Cal.Rptr. 7 ] [problem of whether a liability insurer had duty to fund an appeal of underlying case].) However, because of the close identity between Friedman and the center and the fact that it was the actions of Friedman himself that got the center sued in 1996, we shall refer to the center as Friedman for purposes of our discussion. Here is Friedman’s testimony on the point: “Q. This bleeding was going on in the outpatient surgery center? “A. Yes, sir. “Q. And the bleeding was coming from the vagina where the operative procedure had been performed? “A. Yes, sir. “Q. In an effort to try to save Ms. Hamel’s life, you made this tampon and you tamponaded the bleeding? “A. Yes, sir.” The court also cited to Big Boy Drilling Corp. v. Rankin (1933) 213 Cal. 646, 649 [ 3 P.2d 13 ], a case also involving efforts to recover money for work done on real property, to confirm its conclusion that the two acts of negligence really gave rise to only a single injury. In a slight ironic twist, though, Big Boy Drilling had reasoned that since the same purpose (collection of money) might be accomplished either by mechanics’ lien or personal judgment, since “both demands” arose “out of the same transaction, there is but one cause of action with two forms of relief.” (See Bay Cities, supra, 5 Cal.4th at p. 860 .) Since the rule in Gregory was adopted by our high court by six votes (with one considered abstention), it is instructive to review the case. There, the client was a broker selling investments in a videotape series. The broker hired a lawyer to draft a tax and security opinion letter regarding the investment. The letter said that the videotapes were not securities—and thus did not need to be registered with the Securities Exchange Commission (SEC). It also said that there were certain tax advantages to the investments. (See Gregory, supra, 876 F.2d at p. 603 .) The lawyer was wrong on both counts. A federal district court held that the videotapes sales were, as “investment contracts,” securities which had to be registered with the SEC. The IRS disallowed the income tax deductions. The broker consequently sued the lawyer for negligence in the preparation of the opinion letter. (See Gregory, supra, 876 F.2d at p. 603 .) Later, the lawyer brought a declaratory relief action to ascertain the limit of his malpractice policy— would it be $1 million if there were two claims (the error as to the securities registration and the error as to the tax advantages), or $500,000 if there were only one. (Ibid.) Like the present case the policy had a provision that two or more claims arising out of a “series of related acts” would be treated as a single claim. The Seventh Circuit ultimately held that the claims were related “in any meaningful sense of the term,” agreeing with the federal district court that they were “ ‘interdependent components of a single plan.’ ” (See Gregory, supra, 876 F.2d at pp. 605-606.) The court ended the opinion with the observation that the “common understanding of the word ‘related’ covers a very broad range of connections, both causal and logical.” (Ibid.) The invasion of privacy tort may also be said to be logically related to the 1994 malpractice suit. The 1996 complaint emphasized that Hamel had not even consented to the resectoscope procedure and that Friedman was present in his capacity of person-in-charge of the pump used in that procedure. Friedman’s (invasive) presence in the room, the subject of the 1996 claim, was thus directly tied to the very procedure which, under the 1994 claim, had led to Hamel’s pulmonary edema. We can easily imagine a more difficult scenario: Suppose a woman goes in for a routine pap smear and her male doctor negligently performs the procedure by not preserving the specimen according to the standard of care. Additionally, while performing the pap smear, the doctor touches the woman for sexual gratification. There is clearly no causal relationship between the acts. Are they “logically” related, perhaps on some theory that the doctor was taking advantage of an opportunity that presented itself because of the occasion for the malpractice and the occasion for the molestation were the same? We certainly do not speculate on this point, except to note that it might be a situation where Justice Kennard’s point in her Bay Cities concurrence that there are some different claims which, though they might seem related temporarily or thematically, still would not reasonably be considered related. (See Bay Cities, supra, 5 Cal.4th at p. 875 (cone. opn. of Kennard, J.).) In our hypothetical, the molestation is logically and causally ««related to the malpractice. (They might, under the federal transactional test for res judicata, be considered the same claim, but that is another story.) Not only do we have different injuries (the injury to the patient’s ultimate health inherent in the doctor’s breach of the standard of care and the injury to the patient’s personal integrity inherent in the molestation), but those injuries exist quite independently of each other. Hence the doctor might have molested the patient even if he done a perfect job on preserving the pap smear specimen. Conversely the doctor might have committed the malpractice without ever having touched the patient for sexual gratification. Whether this case (save for, perhaps, the invasion of privacy claim) also involves a logical relationship is another question, and under Bay Cities there is no need to reach that issue here. Again, we do not decide that hypothetical, which would involve the problem of whether a logical, if not causal, relationship existed between the two acts.