I concur in everything that my colleagues say, but I would say more. And what I would say implicates a concept apparently overlooked by the court below: the concept of the finality of judgments.
A bit of the procedural history of this case needs to be revisited: in 1996, Judge Carol Yaggy, then a probate court commissioner, entered an order in a conservatorship proceeding requested by the then conservator, the conservatee’s wife. It required, as the majority notes, that conservatee Paul J. Winters (recently hospitalized and diagnosed with, among other things, Alzheimer-type dementia) would “not have the power either to amend or revoke [his trust] without the prior approval of this Court.” This order was subsequently reapproved and reissued by Judge Yaggy when, upon the death of the wife, a professional conservator was appointed to succeed her as conservator.
As the majority opinion notes, a week or so before the conservatee died in December 1997, an amendment was prepared by respondent, signed by the conservatee and then notarized by respondent, but all without seeking court approval. About a month after Winters’s death, in January 1998, respondent sought approval of this amendment from Judge Yaggy, only to be met with opposition from the conservatee’s stepdaughter, who would take everything under an unaltered trust.
No one, least of all respondent, should have been surprised when, on April 8, 1998, Judge Yaggy denied the petition, essentially saying that the orders *676she had entered twice before meant what they said, i.e., that an amendment to the subject trust had to be approved by the probate court before it could be effective.
Appellants, who stood to be the beneficiaries of the putative amendment, appealed this order to this district (Conservatorship of Winters, A082566), but later dismissed their appeal after effecting a settlement with the stepdaughter. Thereupon, of course, Judge Yaggy’s order of April 8, 1998, became final.
Except, apparently, to respondent and the court below. Evidently encouraged by respondent’s counsel via their motions in limine, the court below engaged in what may charitably be called substantial second-guessing of Judge Yaggy’s order of April 8, 1998, as well as—albeit implicitly—the two orders underlying it. I regard this exercise and the unmistakable implication of the ruling under review that Judge Yaggy’s order was “unreasonable,” to be contrary to the spirit, if not the letter, of the concept of judicial finality.1
This concept has many subparts, e.g., res judicata, collateral estoppel, the rule prohibiting collateral attacks on judgments, etc. Underlying all of them is the basic principle that: “If a judgment, no matter how erroneous, is within the jurisdiction of the court, it can only be reviewed and corrected by one of the established methods of direct attack.” (8 Witkin, Cal. Procedure (4th ed. 1997) Attack on Judgment in Trial Court, § 1, p. 507.) Our appellate courts have repeatedly quoted this basic rule. (See, e.g., Estate of Buck (1994) 29 Cal.App.4th 1846, 1854 [35 Cal.Rptr.2d 442]; People v. $6,500 U.S. Currency (1989) 215 Cal.App.3d 1542, 1548 [264 Cal.Rptr. 294].)
Now it is technically correct that neither respondent nor the court below specifically sought to “correct” Judge Yaggy’s various orders. To that extent, I will concede that what transpired below may not have been, strictly speaking, a collateral attack on those orders. But “review” those orders it certainly did, with many, many explicit references effectively labeling them as “unreasonable.” It seems to me that this is fundamentally contrary to both the respect due from one superior court department to another and, more importantly, to the overall concept of the finality of judgments.
*677And, finally, it was and is also contrary to the principle laid down by this court a decade ago that trial courts ought not to second-guess probate courts on matters within the latter’s peculiar jurisdiction. Thus, in Estate of Gump (1991) 1 Cal.App.4th 582, 607 [2 Cal.Rptr.2d 269], we said (quoting from an earlier, but by then depublished, opinion in the same controversy): “ ‘The law disfavors the avoidance of probate proceedings and the litigation on the law side of the court of matters properly heard and resolved in probate.’ ” This controversy provides an appropriate “amen” to that statement.
A petition for a rehearing was denied September 12, 2001, and the opinion was modified to read as printed above.
Nor, contrary to the contention of respondent’s counsel at oral argument, is such an exercise mandated or permitted by the “trial within a trial” mechanism discussed in, e.g., Mattco Forge, Inc. v. Arthur Young & Co. (1997) 52 Cal.App.4th 820, 831-840 [60 Cal.Rptr.2d 780]. That process may well involve a determination of what, hypothetically, a “reasonable” judge or fact finder would have done had the alleged malpractice not have precluded or prevented the necessary finding or determination. It most certainly does not permit, much less mandate, legal malpractice litigation being effectively determined on the premise that an existing final judgment is “unreasonable.”