I respectfully dissent. The facts clearly indicate that this lawsuit acted as a catalyst by speeding up the process of issuance of the final regulations.
Sections 11135 and 11139.5 of the Government Code were enacted in September of 1977; yet, two years later, when this lawsuit was instituted, only proposed regulations had been drafted. No dates had been- set for public hearings; no date had been set for issuance of final regulations; indeed, the proposed regulations themselves had not yet been issued. There were plans to send questionnaires to various agencies in January for the purpose of further prolonged study of the fiscal cost of the regulations that would have taken a minimum of 12 weeks to complete, after which at least 30 days would have been necessary to notice public hearings; in addition, respondent planned to seek the concurrence of the State Building Standards Commission—a complex process which would have taken a long, if not inordinate, period of time.
In contrast, only one month after the initiation of this lawsuit, the trial court ordered respondent to advise it by January 30, 1980, of a date when the final *356regulations would be issued. Just prior to the January court hearing, respondent issued the proposed regulations and notified the court that public hearings would be completed by the end of May.1 On February 13, 1980, respondent published a Notice of Proposed Adoption of Regulations, which listed dates in April for public hearings to be held in Los Angeles, Fresno, San Francisco and Sacramento. The final paragraph of the notice read: “The Secretary of the Health and Welfare Agency has determined that the above regulations will have no direct costs or savings to state agencies or increased or new costs to local government pursuant to Section 2231 of the Revenue and Taxation Code. The building standards contained in the above regulations are hereby noticed for public hearing with the concurrence of the State Building Standards Commission.” This is in direct contrast to respondent’s original plan, under which the cost evaluation would not have been completed until at least late March or early April, after which public hearings would have been noticed and the concurrence of the commission obtained. Thus, it is clear that respondent indeed had accelerated the promulgation process.
At the January 30 trial court hearing, respondent advised the court that final regulations would be issued by June 2, 1980. The trial court decided to continue the matter until such time. When it became apparent that respondent was not going to meet the June 2 deadline, the trial judge issued an alternative writ of mandate to either issue the final regulations by June 30, or show cause why the regulations should not issue. Respondent issued the final regulations on June 24, 1980.2
As the majority recognizes, “ ‘voluntary’ corrective action, induced by litigation, may properly be considered a ‘benefit’ of the litigation in determining the *357propriety of an attorney fee award. [Citations.] ’’(Northington v. Davis (1979) 23 Cal.3d 955, 960, fn. 2 [154 Cal.Rptr. 524, 93 P.2d 221]; see maj. opn. at p. 352 above.) Petitioners are entitled to attorneys’ fees if there is a causal connection between their suit and respondent’s actions. (See, e.g., Sullivan v. Com. of Pa. Dept. of Labor (3d Cir. 1981) 663 F.2d 443; Morrison v. Ayoob (3d Cir. 1980) 627 F.2d 669, cert. den. (1981) 449 U.S. 1102 [66 L.Ed.2d 828, 101 S.Ct. 898]; Ross v. Horn (3d Cir. 1979) 598 F.2d 1312; Young v. Kenley (4th Cir. 1981) 641 F.2d 192; Smith v. University of North Carolina (4th Cir. 1980) 632 F.2d 316; Robinson v. Kimbrough (5th Cir. 1981) 652 F.2d 458; Harrington v. DeVito (7th Cir. 1981) 656 F.2d 264; United Handicapped Federation v. Andre (8th Cir. 1980) 622 F.2d 342; Chicano Police Officer’s Assn. v. Stover (10th Cir. 1980) 624 F.2d 127; Foster v. Boorstin (D.C. Cir. 1977) 561 F.2d 340; Cuneo v. Rumsfeld (D.C. Cir. 1977) 553 F.2d 1360.)
The majority errs, unfortunately, in its vain search for specific evidence in the record to identify the “causal” connection, such as an affidavit or statement by the respondent that the lawsuit induced him to speed up the process. Such an affidavit, however, is simply unnecessary here. The trial judge, after considering the totality of the circumstances, was quite able to make a determination that a causal connection in fact existed. This judge sat through the entire proceedings from their very beginning; he listened to both parties testify, and read the various pleadings. He had the opportunity to observe the demeanor of the respondent at the hearings. He asked for a date of final issuance which had not been set, and the respondent responded by setting the date. Thereafter, at least two months earlier than respondent had planned, respondent noticed public hearings to be held in April. Thereafter, when further delay was evident, the trial judge ordered that the final regulations be issued by June 30, and the respondent responded by issuing them. It was thus quite reasonable for the trial judge to conclude that respondent had felt compelled to hasten the process of issuing final regulations.
Furthermore, there is no evidence to the contrary. Respondent has not shown that he would have set a June date for final issuance had there been no hearing in January, nor that he would have noticed public hearings as early as February, nor that he would have actually issued the final regulations in June had no alternative writ of mandate been issued. He showed no evidence of other forces as his motivation for acting so swiftly. He alone had control over information explaining his actions—he, rather than petitioners, should have borne the burden of going forward with evidence to prove that the lawsuit did not act as a catalytic factor. (Cf. Nadeau v. Helgemoe (1st Cir. 1978) 581 F.2d 275, 281 (the inference drawn from the chronological sequence of events is stronger when evidence of defendant’s behavior is under defendant’s control).) Therefore, the only reasonable conclusion to be drawn is that the lawsuit, and *358the subsequent orders of the trial judge, compelled respondent to act. Thus, the lawsuit indeed acted as a catalyst in speeding up the promulgation process.
At issue today is the future of public interest litigation such as that at bar which has benefited handicapped people throughout the State of California. Our private attorney general statute was passed to ensure that attorneys would remain motivated to prosecute cases in the public interest. “[Privately initiated lawsuits are often essential to the effectuation of the fundamental public policies embodied in constitutional or statutory provisions, and . . . without some mechanism authorizing the award of attorney fees, private actions to enforce such important public policies will as a practical matter frequently be infeasible.” (Woodland Hills Residents Assn., Inc. v. City Council (1979) 23 Cal.3d 917, 933 [154 Cal.Rptr. 503, 593 P.2d 200].) Our policy therefore should be to encourage, rather than discourage, public litigation suits. To do so, we must allow a trial judge to make reasonable inferences from his or her perception of the totality of the proceedings before him or her, and to independently assess the catalytic effect of the proceedings on defendants. Requiring some kind of documental or testimonial proof of a causal connection only serves to deeply and impermissibly penetrate into the realm of discretion of the trial judge. I would therefore give great deference to a trial judge’s finding of a causal connection, and would find no abuse of discretion absent concrete evidence by respondent that other factors, rather than the lawsuit, prodded him into speeding up the process. I would therefore uphold the award of attorneys’ fees in this case.
Mosk, J., and Reynoso, J., concurred.
The declaration of William Wilder, which respondent filed immediately prior to the December trial court hearing, did not indicate that respondent planned to publicly issue the proposed regulations in January. A close reading of the declaration reveals only that a draft of the proposed regulations was to be included in the questionnaire package that was to be sent to various agencies for purposes of fiscal analysis of the regulations. Obviously, such an analysis could not be accomplished without a reading of the proposed regulations by the agencies.
The majority argues that since respondent informed the trial court that final regulations would be released by the end of June prior to the issuance of the alternative writ, the writ did not prod him into speeding up the process. (See maj. opn., p. 354, fn. 6, ante.) Not only has the majority misstated the facts (see text accompanying this footnote), it has also ignored the realities of this case. The facts remain that respondent did not set any date for the issuance of final regulations until ordered to do so by the trial court; before an alternative writ was issued, respondent acknowledged that there would be a delay in issuing the regulations; respondent did not actually issue the regulations until the trial court ordered him to do so in the alternative writ. These facts, coupled with the history of previous delays in the entire promulgation process, clearly indicate that respondent would not have issued the final regulations had petitioners not initiated this lawsuit and successfully convinced the trial court to issue an alternative writ of mandate.