I respectfully dissent. With the exception of the majority opinion, I know of no case which suggests that an attorney whose advice is correct may be held liable for malpractice.
*423Relying on the standard developed in Smith v. Lewis (1975) 13 Cal.3d 349 [118 Cal.Rptr. 621, 530 P.2d 589, 78 A.L.R.3d 231] and its progeny,1 the majority concludes that an attorney may face malpractice liability despite the fact that the law is ultimately resolved in accordance with the advice given. Although this application of the Smith standard follows logically from its emphasis on the duty of care owed a client, it nonetheless raises a troubling anomaly: where the law is unsettled, the attorney who gives advice later determined to be correct may well have committed malpractice, while the attorney whose advice turns out to be erroneous may avoid liability entirely.
The law cannot tolerate such incongruous results. As Justice Holmes so aptly observed long ago, “[t]he life of the law has not been logic: it has been experience. ” (Holmes, Common Law (1881) p. 1.) Experience now tells us that the Smith standard, however rational and well-suited to its original purpose, no longer makes sense. We must therefore formulate a new standard that draws a fair and reasonable distinction between culpable and nonculpable practitioners.
The defect inherent in the Smith standard, made ever clearer by today’s majority opinion, is that the concept of legal error is confused with that of fault, converting a question of law into one of fact. Malpractice consists of four elements: duty arising out of the attorney-client relationship, breach of that duty, causation and damages. The second element breaks down further into two components: legal error and failure to use “such skill, prudence and diligence as lawyers of ordinary skill and capacity commonly possess and exercise in the performance of the tasks which they undertake.” (Lucas v. Hamm (1961) 56 Cal.2d 583, 591 [15 Cal.Rptr. 821, 364 P.2d 685].) The first is a question of law, the second a question of fact.
The question of whether an attorney erred necessarily must be resolved before any issue of negligence arises. An attorney who renders erroneous advice may not be negligent in doing so. (See Davis v. Damrell (1981) 119 Cal.App.3d 883 [174 Cal.Rptr. 257].) A second attorney may fail to perform adequate research but somehow give his client accurate advice. Neither of these attorneys has committed malpractice. (See Mallen & Levit, Legal Malpractice (2d ed. 1981) § 250, p. 317.)
*424Where the law is settled, it is relatively easy to determine whether the attorney’s advice was erroneous. Problems arise only with respect to issues of law that are unresolved or in a state of flux at the time the advice is given. In either instance, however, the question of whether the advice was wrong is a question of law.
Ironically, Smith itself reflects this basic approach. At the outset of the analysis the court stressed: “the crucial inquiry is whether his advice was so legally deficient when it was given that he may be found to have failed to use ‘such skill, prudence, and diligence as lawyers of ordinary skill and capacity commonly possess and exercise in the performance of the tasks which they undertake. ’ [Citation.] We must, therefore, examine the indicia of the law which were readily available to defendant at the time he performed the legal services in question.” {Id., at p. 356.) (Italics added.)
Thus, Smith initially proposed a two-step test for determining whether an attorney has been negligent. As noted, the threshold inquiry is a legal one, whether adequate legal authority existed at the time to support the advice given. Only when this question is answered in the negative is it necessary to move to the second part of the test, the factual inquiry as to whether the attorney breached the standard of care in rendering the erroneous advice.
Applying this test to the case at bar reveals that Attorney Mash did not err in advising his client in 1971 that her husband’s federal military pension was not community property. As the majority notes, “[i]n 1971, the California view regarding the characterization of vested federal military retirement pensions as community or separate property was unsettled.” (Ante, p. 416.) In fact, Mash relied on an opinion of this court, French v. French (1941) 17 Cal.2d 775 [112 P.2d 235, 134 A.L.R. 366], in concluding that the pension was not divisible. As French remained good law, this reliance was neither unreasonable nor erroneous. Because Mash committed no error, the malpractice claim must fail.
It is imperative that a lawyer remain free to choose one of a number of reasonable and legally supportable solutions to an otherwise unsettled legal question and advise the client accordingly without facing a malpractice suit.
Bird, C. J., and Taber, J.,* concurred.
Prior to Smith attorneys in California were not liable “for lack of knowledge as to the true state of the law where a doubtful or debatable point [was] involved.” (Sprague v. Morgan (1960) 185 Cal.App.2d 519, 523 [8 Cal.Rptr. 347].) Smith modified that rule so that even with regard to an unsettled area of the law an attorney is obligated to “undertake reasonable research in an effort to ascertain relevant legal principles and to make an informed decision . . . .” {Smith, supra, 13 Cal.3d at p. 359.)
Assigned by the Chairperson of the Judicial Council.