42 Cal. 3d 437
Cal.
1986

Lead Opinion

Opinion

BROUSSARD, J.

This сase concerns the standards and procedures to be followed by the trial court when the Board of Prison Terms notifies it that its sentence is disparate when compared to the sentences imposed in other cases.

*441On September 19, 1979, defendant was found guilty of five counts of robbery. The jury also found that defendant had used a knife in two of the robberies. Before sentencing, defendant was rearrested and charged with two additional counts of robbery with a firearm enhancement for each robbery. Defendant pled guilty to the additional charges, and admitted the enhancements.

The matters were consolidated for sentencing on February 8, 1980. The trial court sentenced defendant to twelve years imprisonment: the upper term of five years for the princiрal robbery, a one-year enhancement for the use of a knife in that robbery (Pen. Code, § 12022, subd. (b)),1 and six consecutive terms of one year (one-third of the middle term) for the remaining six robberies. Defendant did not appeal.

Section 1170, subdivision (f) of the determinate sentence law requires the Board of Prison Terms to review every sentence “to determine whether the sentence is disparate in comparison with the sentences imposed in similar cases.”2 There is no statutory definition of disparity. The Attorney General, *442however, has advised the board that a sentence is disparate if “there is a substantial difference between the subject sentence and the sentences imposed on other offenders committing the same offense under similar circumstances.” (60 Ops.Cal.Atty.Gen. 143, 147 (1977).) Review is not designed to uncover legal error in sentencing (People v. Herrera (1982) 127 Cal.App.3d 590, 597 [179 Cal.Rptr. 694]), although the board may call such error to the attention of the trial court, but instead to promote uniformity in the exercise of sentencing discretion.

Justice Kline, dissenting in the Court of Appeal in the present case, described more fully the background and purpose of section 1170. “[T]he determinate sentencing law,” he said, expresses “a rather definitive legislative effort to diminish inequitable disparities in punishment to the extent practicable. (See Parnas & Salerno, The Influence Behind, Substance and Impact of the New Determinate Sentencing Law in California (1978) 11 U.C. Davis L.Rev. 29, 39.) This effort was motivated in part by judicial decisions disapproving administration of the Indeterminate Sentence Law in certain particulars (see, e.g., In re Rodriguez (1975) 14 Cal.3d 639 [122 Cal.Rptr. 552, 537 P.2d 384], and In re Stanley (1976) 54 Cal.App.3d 1030 [126 Cal.Rptr. 524, 128 Cal.Rptr. 829]), but it was at least equally the result of ‘an accumulation of evidence that [under the indeterminate sentencing system] there were wide sentencing disparities in terms of both who went to prison and their lengths of stay for similar offenses.’ (Determinate and Indeterminate Sentence Law Comparisons Study, Rep. to Cal. Legis., Joint Com. on Rules, Arthur D. Little, Inc. (May 1980) p. II-1; see also Crowther, Crime, Penalties and Legislatures (1969) 381 Annals 147, 153.) This evidence was punctuated by the developing national consensus that significant disparities in the punishment imposed upon like individuals committing like offenses was a pernicious evil endangering the very integrity of the criminal justice system. (See, e.g., Hoffman & Stover, Reform in the Determination of Prison Terms: Equity, Determinacy and the Parole Release Function (1978) 7 Hofstra L.Rev. 89; von Hirsch, Doing Justice: The Choice of Punishments (1976); Gaylin, Partial Justice: A Study of Bias in Sentencing (1974); Hagan, Extra-Legal Attributes and Criminal Sentencing: An Assessment of a Sociological Viewpoint (1974) 8 Law & Soc’y. Rev. 357; Frankel, Criminal Sentences: Law Without Order (1973); President’s Commission on Law Enforcement and the Administration of Justice, Task Force Report: The Courts (1967); and Shepard v. United States (6th Cir. 1958) 257 F.2d 293, 294 (opn. of Potter Stewart, J.).)

“Numerous empirical studies revealed that the chief explanation for sentencing disparity was not the differences in defendants but the differences in judges. (See, e.g., Rubin, Disparity and Equality of Sentences—A Constitutional Challenge (1967) 40 F.R.D. 55, 58; McGuire & Holtzoff, The *443Problem of Sentence in the Criminal Law (1940) 20 B.U.L. Rev. 423; Glueck, Crime and Justice (1936) p. 127; Gaudet et al., Individual Differences in the Sentencing Tendencies of Judges (1933) 23 J.Crim. L. C. & P. S. 811; and Everson, The Human Element in Justice (1919-20) 10 J. Crim. L. 90.) Thus the movement to promote uniformity in sentencing . . . was in no small part a movement to diminish judicial discretion. (See, e.g., Countdown for Judicial Sentencing in Of Prisons and Justice, Sen. Doc. No. 70, 88th Cong., 2d Sess., pp. 331-332 (1964).)”

Carrying out its statutory mission to promote uniformity in sentencing, the board undertakes а statistical analysis of discretionary sentences imposed under the Determinate Sentencing ‍‌‌​​​‌​‌​‌​‌​‌‌​‌​​​​​​‌‌‌​‌​​‌‌‌‌‌‌​‌‌​‌​​​​‌​​‍Act. The board’s methodology is described in the introduction to its Report on Sentencing Practices—Determinate Sentencing Law (hereafter Report), issued February 28, 1985.3 The board has acquired data on over 75,000 cases, including, in each case, the prisoner’s criminal history, his social background, the circumstances of each offense, and the actual sentence. When it receives a new case, the board conducts a preliminary screening (an Automated Sentence Review) to determine the range of possible sentences and the relative likelihood of each. A computerized simulation generates 10,000 theoretical sentences for the cases. The computer then determines the percentage of simulated sentences which are equal to or higher than the actual sentence imposed, and provides a statistical measure of the difference between the expected sentence and the actual sentence. If both criteria exceed an established threshold, the case is identified as one requiring further analysis.

In the secondary screening process, the board’s staff obtains a list of cases comparable to the subject case. The staff examines the file in the subject case and each comparable case to determine if there are facts which would justify a statistically disproportionate sentence. If the staff finds none, it refers thе matter to a panel of two board members and a hearing representative. The panel reexamines the subject cases and comparable cases. If it also finds no justification for a disproportionate sentence, it directs the board’s counsel to notify the court that, in its opinion, the sentence is disparate.

In the present matter the board first determined that the trial court erred in imposing six consecutive one-year terms for the subordinate offenses. Relying on People v. Harvey (1979) 25 Cal.3d 754, 761 [159 Cal.Rptr. 696, 602 P.2d 396] and People v. Childs (1980) 112 Cal.App.3d 374, 388-389 [169 Cal.Rptr. 183], the board decided that section 1170.1 prohibited *444consecutive sentences for subordinate terms in excess of five years. Defendant’s sentence, the board concluded, should be reduced by 1 year from 12 years to 11 years.4

The board then set out to determine whether an 11-year sentence would be disparate. The claimed disparity in defendant’s sentence arose from the fact that he not only received the upper term for the principal offense, but that all of the possible subordinate terms were imposed consecutively. A computer simulation indicated only a small probability that a sentence as severe as 11 years for someone convicted of the same crimes and enhancements would be imposed. This computation identified defendant’s sentence as one requiring further study.

The Sentencing Review Unit then examined the record in defendant’s case in detail but was unable to discover any unusual factors justifying the severe sentence. The Case Analysis Committee compared defendant’s sentence with 29 other factually similar cases. Each of these сases had the following common factors: (1) the principal offense was robbery, (2) the defendant was not under supervision at the time of the earliest offense, (3) he had no prior juvenile criminal history, (4) he had no prior adult convictions, (5) no victim was injured, and (6) there were three or more unstayed subordinate counts. Of the twenty-nine defendants, only six received the upper term plus two or more consecutive terms, and only two received the maximum possible number of consecutive terms. The average sentence for the subgroup was 77 months, and the most likely sentence either 48 or 60 months; defendant’s corrected sentence of 132 months was substantially longer. The committee concluded that defendant’s sentence was in fact disparate. A panel of the board independently reviewed the comparison, and concurred in that conclusion.

On August 20, 1981, the board filed a motion for resentencing pursuant to section 1170, subdivision (f). The motion asserted not only that defendant’s sentence was disparate, but also that it was legally in error because section 1170.1, as then interpreted, limited subordinate terms to a total of five years. (See fn. 4, ante.) The board noted, however, that a corrected term of 11 years would still be disparate, and recommended resentencing to a term between 5 and 10 years.

*445At argument on the board’s motion, the judge questioned the board’s methodology. He was particularly concerned with the failure of the board’s statistical analysis to take into account certain subjective factors such as defendant’s “рersonality, his particular conduct, his attitude, his demeanor during the time of the offenses.” There is a difference, the court said, between a robber who stands at a distance from the victim and points a weapon and one who puts a knife to the victim’s throat, and the board’s statistical analysis did not appear to take account of such differences.5

Counsel for the board explained that the statistical tables depend on objective, quantifiable matters and thus cannot include subjective factors. Before making its recommendation, however, the board also reviews the case record and the probation report, looking for other, nonquantifiable factors which might justify an especially severe sentence. Thus the board does consider subjective, personal facts to the extent that those factors appear in the record of the case.

Following the hearing on the board’s motion, the trial judge issued his order granting the motion only insofar it asserted legal error. Defendant’s sentence was reduced to 11 years by making one of the subordinate terms run concurrently with the principal term.

Defendant appealed, contending that the trial court abused its discretion by not resentencing him to a term within the range recommended by the board. The Court of Appeal affirmed the sentence. We granted a hearing to consider the steps to be followed by a trial court when it is notified that the board considers its sentence disparate and the need for a statement of the reasons for thе court’s decision.

We endorse the framework established by People v. Herrera, supra, 127 Cal.App.3d 590, the first appellate case to construe section 1170, subdivision (f). Under that framework trial courts must undertake a two-step analysis. “First, [they] must determine whether the sentence imposed was indeed disparate.” (P. 601.) Second, if they find the sentence disparate, the courts must then decide whether to recall the sentence. (Ibid.)

*446The trial court begins its analysis knowing that the board has already found the sentence disparate. The court recognizes that the board, in performing its duties under the determinate sentencing law, has acquired extensive experience and expertise in the comparative review of sentences. It has gathered data, established statistical methods of analysis, and devised computer simulations. Employing those techniques, it hаd reviewed and compared over 75,000 sentences. The trial judge, on the other hand, may have little or no experience in comparative review of sentences. His understanding of statistical methods and computer simulations may be fragmentary. His knowledge of sentences may be largely limited to those imposed by himself and other judges in his court. Accordingly, Herrera concluded, the determination of the board that a sentence is disparate is entitled to “great weight.” (127 Cal.App.3d 590, 600-601.) We agree with this conclusion.

The problem is one of defining what is meant, in each step of the analysis, by according “great weight” to the board’s determination. Herrera stated that “a trial court will have met its obligation of according the Board’s finding of disparity ‘great weight’ in the first part of the analysis ... if the record shows that the court seriously considered the information provided by the Board, and attempted to discern whether, when compared to sentences imposed by his colleagues, the sentence he imposed . . . was ‘disparate.’” (P. 601.) The court confessed greater difficulty in describing the second part of the analysis, but said that “[ejven though the judge may agree that the sentence he imposed does not conform to the ‘observed sentencing pattern’ relied on by the Board (and is therefore disparate), he may question the underlying rationale of the sentencing pattern practiced by the majority of his colleagues.” (P. 602.)

We conclude, however, that the Herrera formulation does not fully and accurately state the obligation of the trial court. That court must not only seriously consider the information provided by the board; it must give weight to th& finding of the board. “To give weight” implies that the faсt that the board, in its expert judgment, made a finding of disparity will itself exert an influence on the court’s decision—an influence beyond that which simply flows from the information furnished by the board. Moreover, in the second step of the analysis, it is not the function of the court to question the sentences imposed by other judges. Such an inquiry would permit a judge to impose disparate sentences as an expression of his personal view of sentencing policy, a result which would defeat the statutory goal of uniformity.

We think a better exposition of the weight to be given the board’s finding is that advanced by Justice Kline, drawing upon our decision in People v. *447Carl B. (1979) 24 Cal.3d 212 [155 Cal.Rptr. 189, 594 P.2d 14], reaffirmed in People v. Javier A. (1985) 38 Cal.3d 811 [215 Cal.Rptr. 242, 700 P.2d 1244], (See also People v. Fields (1984) 159 Cal.App.3d 555, 569 [205 Cal.Rptr. 888].) Both Carl B. and Javier A. involved trial court decisions rejecting a Youth Authority recommendation that a juvenile convicted of crime be committed to the authority instead of state prison. We said that such a recommendation was entitled to “great weight’’ (Carl B., supra, 24 Cal.3d 212, 214; Javier A., supra, 38 Cal.3d 811, 819), and went on to explain what that meant. Such a recommendation, we said, must be followed in the absence of “substantial evidence of countervailing considerations of sufficient weight to overcome the recommendation.” (Javier A., supra, 38 Cal.3d 811, 819; see Carl B., supra, 24 Cal.3d 212, 214-215.)

In the present context, giving “great weight” to a finding of disparity in the first step of the analysis means that the trial court must accept the board’s finding of disparity unless based upon substantial evidence it finds that the board erred in selecting the appropriate comparison group6 or in determining that defendant’s sentence differs significantly from that imposed upon most members of that group. If there are unique elements in the case which render it unsuitable for comparison with other ‍‌‌​​​‌​‌​‌​‌​‌‌​‌​​​​​​‌‌‌​‌​​‌‌‌‌‌‌​‌‌​‌​​​​‌​​‍cases, or subjective factors which distinguish it from other cases, such matters can be considered in the second part of the analysis when the court considers whether a disparate sentence is justified.

In the second stage, the trial court must again give great weight to the board’s finding of disparity, a finding it upheld in the first stage of the analysis. That finding does not automatically require it to recall its sentence.7 Under the reasoning of People v. Carl B., supra, *44824 Cal.3d 212 and People v. Javier A., supra, 38 Cal.3d 811, however, giving great weight to the finding does require the court to recall its sentence unless there is substantial evidence of countervailing considerations which justify a disparate sentence. Such considerations can include subjective factors like those mentioned by the trial court—such as defendant’s attitude and demeanor at the time of the crime, and the manner in which he threatened the victim.8

Requiring the trial judge merely to “consider’’ the finding of disparity, as the dissent proposes, gives no weight at all to that finding. The judge would remain free to disregard the finding for any reason, or no reason at all. And without findings, which again the dissent would not require, his decision would not be subject to effective appellate scrutiny.

Under a regime of essentially unlimited and unreviewable discretion, some judges would still revise sentences to avoid disparity but others, in factually indistinguishable cases, would decline to do so: disparate sentences, disparately reconsidered. The uniformity sought to be promoted by the board’s statewide sentence review would be defeated by such individual divergence.

In the present case the court rejected the board’s finding of disparity, and thus did not determine whether a disparate sentence was justified. That ruling, however, appears erroneous, since there is little evidence that the board erred in selecting the comparison group or in finding that defendant’s sentence was significantly longer than the sentences imposed on most members of that group.9 Moreover, although the judge made no findings or statement of reasons, his comments during argument on the motion suggest that he misunderstood the board’s methodology: he did not realize that the board’s procedures do take into account “subjective” factors to the extent they are reflected in the record; his assumption that the subjective factors were unique to the case at hand lacks support;10 his comment that a five-to ten-year sentence would also be disparate suggests that he did not understand the meaning of that term as used by the board. (See fn. 5, ante, p. 445.) We conclude that the court’s ruling that its sentence was not disparate should be reconsidered in light of the analysis in this opinion.

*449We recognize that the judge referred to unique and subjective factors which might justify a disparate sentence in this case.11 Having found no disparity, however, the judge did not determine the significance of those factors. Upon remand, he is free to reaffirm the present sentence if he finds that there are countervailing considerations, not taken into account in the board’s analysis, which justify a disparate sentence.

This brings us to a question which has hampered appellate review of disparate sentences in this and other cases—the question of findings. Since the mode of analysis we have set out requires findings, we believe it appropriate to address that question by way of guidance to the court on remand, •and to other courts reviewing sentences under section 1170, subdivision (f).

Since the Legislature did not impose an explicit requirement for a statement of reasons, the Court of Appeal in People v. Mitchell (1984) 152 Cal.App.3d 433, 437 [199 Cal.Rptr. 507] and People v. Shepeard, supra, 169 Cal.App.3d 580, 585, fn. 5, held that such a statement was unnecessary. The courts, however, have the inherent power to require a statement of reasons as a judicially declared rule of criminal procedure. (See, e.g., In re Podesto (1976) 15 Cal.3d 921, 938 [127 Cal.Rptr. 97, 544 P.2d 1297]; People v. Vickers (1972) 8 Cal.3d 451, 461 [105 Cal.Rptr. 305, 503 P.2d 1313]; People v. Cohan (1955) 44 Cal.2d 434, 442 [282 P.2d 905, 50 A.L.R.2d 513].)12 In In re Podesto, supra, 15 Cal.3d 921, we emphasized that a requirement of articulated reasons to support a given decision serves a number of interests: it is frequently essential to meaningful review; it acts аs an inherent guard against careless decisions, insuring that the judge *450himself analyzes the problem and recognizes the grounds for his decision; and it aids in preserving public confidence in the decision-making process by helping to persuade the parties and the public that the decision-making is careful, reasoned and equitable. (15 Cal.3d at p. 937.)

The present case is also one where a statement of the trial judge’s reasons for reaffirming a disparate sentence appears essential to meaningful appellate review. The trial court’s decision to deny a motion to recall under this subdivision is an appealable order. (People v. Herrera, supra, 127 Cal.App.3d 590, 595-597.) That decision may be based either upon errors in the board’s determination or upon the presence of countervailing considеrations which justify a disparate sentence. If those errors or considerations are not identified by the trial judge, the appellate court cannot perform its reviewing function.

A statement of reasons will also help to guide the trial judge. Disparate sentence review is not a common and familiar procedure, and both the record in this case and the frequent trial court decisions rejecting board findings of disparity suggest that an inarticulated misunderstanding of the process may underlie many decisions. Requiring a written statement of reasons imposes an intellectual discipline that may lead to better reasoned decisions.

Finally, a statement of reasons serves to preserve public confidence in the sentencing process. As noted by the Court of Aрpeals for the Ninth Circuit, “[wjhen there is a substantial disparity in sentences imposed upon different individuals for engaging in the same criminal activity, the preservation of the appearance of judicial integrity and impartiality requires that the sentencing judge record an explanation.” (United States v. Capriola (9th Cir. 1976) 537 F.2d 319, 321.) We conclude that the trial court should state on the record its reasons for finding that a sentence is or is not disparate; if it finds disparity but nevertheless declines to reduce the sentence to a nondisparate term, it should explain the reasons which justify a disparate sentence.

The Court of Appeal below expressed reluctance to require a trial judge to state his reasons on the record “because the trial court is required to state its reasons for an aggravated, enhanced or consecutive term at the initial sentencing hearing. ... We can think of no authority or rationale to require the sentencing court to set forth an entirely different set of factors to justify a perceived disparity by the Board.”

*451When the trial judge originally imposed his sentence, however, he was presumably unaware of its disparate character; in stating the reasons for the sentence he would perceive no need to explain why he was imposing one substantially more severe than those imposed in similar cases. When the board calls the disparity to the trial court’s attention, however, the court must face this issue and determine if its sentence was justified. In explaining that justification, the court is not required to set out a new and different sеt of factors. The reasons which led it to impose an aggravated, enhanced or consecutive term at the original hearing may still be good reasons for such a term, disparity notwithstanding. But as the record here suggests, the trial court may also rely on factors which did not appear in the original sentencing. In either event, the appellate court needs to know the basis for the decision. As the court stated in People v. Mitchell, supra, 152 Cal.App.3d 433, 438, the absence of such findings makes it “virtually impossible for the losing party to attack the judgment of the court on specific grounds and somewhat equally impossible to determine on appeal whether the trial court abused its discretion.”

For the foregoing reasons the order imposing sentence is reversed, and the case remanded for further proceedings in accordance with this opinion.13

Reynoso, J., Grodin, J., and Lui (Elwood), J.,* concurred.

Notes

All statutory citations are to the Penal Code.

Section 1170, subdivision (f), currently reads as follows: “(f)(1) Within one year after the commencement of the term of imprisonment, the Board of Prison Terms shall review the sentence to determine whether the sentence is disparate in comparison with the sentences imposed in similar cases. ‍‌‌​​​‌​‌​‌​‌​‌‌​‌​​​​​​‌‌‌​‌​​‌‌‌‌‌‌​‌‌​‌​​​​‌​​‍If the Board of Prison Terms determines that the sentence is disparate, the board shall notify the judge, the district attorney, the defense attorney, the defendant, and the Judicial Council. The notification shall include a statement of the reasons for finding the sentence disparate.

“Within 120 days of receipt of this information, the sentencing court shall schedule a hearing and may recall the sentence and commitment previously ordered and resentence the defendant in the same manner as if the defendant had not been sentenced previously, provided the new sentence is no greater than the initial sentence. In resentencing under this subdivision the court shall apply the sentencing rules of the Judicial Council and shall consider the information provided by the Board of Prison Terms.

“(2) The review under this section shall concern the decision to deny probation and the sentencing decisions enumerated in paragraphs (2), (3), (4), and (5) of subdivision (a) of Section 1170.3 and apply the sentencing rules of the Judicial Council and the information regarding the sentences in this state of other persons convicted of similar crimes so as to eliminate disparity of sentences and to promote uniformity of sentencing.”

On Decеmber 11,1981, the date of the hearing on the board’s recommendation to reconsider defendant’s sentence, a shorter version of section 1170, subdivision (f) was in effect. It provided that “[i]n all cases the Board of Prison Terms shall, not later than one year after the commencement of the term of imprisonment, review the sentence and shall by motion recommend that the court recall the sentence and commitment previously ordered and resentence the defendant in the same manner as if he had not been previously sentenced if the board determines the sentence is disparate. The review under this section shall concern the decision to deny probation and the sentencing decisions enumerated in subdivisions (b), (c), (d), and (e) of Section 1170.3 and apрly the sentencing rules of the Judicial Council and the information regarding the sentences in this state of other persons convicted of similar crimes so as to eliminate disparity of sentences and to promote uniformity of sentencing.” (Stats. 1980, ch. 1117, § 7, p. 3597.)

We see no difference between the two versions material to this appeal.

The board follows the same procedure whether the sentence is disproportionately high or low, but the trial court has authority only to correct disparately high sentences.

In People v. Hernandez (1981) 30 Cal.3d 462, 466-468 [179 Cal.Rptr. 239, 637 P.2d 706], we held that the rationale of Harvey did not apply to limit subordinate terms, and expressly disapproved Childs. (P. 468, fn. 6.) Thus under Hernandez, defendant’s original 12-year sentence was lawful. At the time of the board proceeding and of the trial court resentencing hearing, however, controlling authority indicated that the five-yеar limit on consecutive terms required a reduction in defendant’s sentence. The trial court accordingly reduced the consecutive subordinate terms to five years. The People do not challenge that ruling.

The judge also said that the board’s recommended sentence was, in his opinion, itself disparate since it involved a range of five to ten years. This concern, however, suggests that the judge did not understand the concept of disparity as used by the board. A sentence is disparate if it falls outside the range of sentences ordinarily imposed for crimes of that character. In the present case, the board found the normal range of sentences was between five and ten years. Thus a sentence within that range would not be disparate; one of 11 or 12 yeаrs would be.

Compare People v. Shepeard (1985) 169 Cal.App.3d 580, 585, footnote 5 [215 Cal.Rptr. 401], where the board may have erred in failing to consider defendant’s use of a firearm when selecting the comparison group.

Defendant suggests that the imposition of a disparate sentence would constitute cruel or unusual punishment in violation of the California Constitution, citing the analysis in In re Lynch (1972) 8 Cal.3d 410 [105 Cal.Rptr. 217, 503 P.2d 921]. The Lynch analysis described techniques for determining whether a statute imposed disproportionate punishment; it does not describe a method for determining whether a particular sentence is disproportionate when compared to other sentences imposed under the same statute. Although cases subsequent to Lynch have held that a punishment which is not disproportionate in the abstract is nevertheless constitutionally impermissible if disproportionate to the defеndant’s individual culpability (People v. Dillon (1983) 34 Cal.3d 441, 480 [194 Cal.Rptr. 390, 668 P.2d 697]), defendant does not claim that an 11-year sentence is disproportionate to his individual culpability.

We agree with defendant that disparate sentence review serves to protect constitutional values, and that in an extreme case a disparate sentence might be unconstitutional, either as “unusual” punishment for the offense in question, or as a denial of the equal protection of the laws. Defendant has not attempted to show that the disparity in his case reaches that level, but appears content to make us aware of the constitutional underpinnings of his statutory argument.

They do not, however, include hostility to the concept of uniform sentencing (cf. People v. Javier A., supra, 38 Cal.3d 811, 818) or to the sentencing practices of his colleagues.

The board may have made one factual error; it said that nоne of defendant’s robbery victims lost more than $500, while in fact two victims lost $800. It does not appear that this error would affect the selection of the comparison groups.

Unlike the board, the trial judge did not examine the records in 29 comparable cases to determine the factors present in those cases.

The judge said there was a difference between a robber who stands at a distance from the victim and points a weapon and one who holds a knife to the victim’s throat (see p. 445, ante), implying, we presume, that defendant was the knifeholder. The court raised the point not to justify a disparate sentence, but to inquire whether the board’s statistical tables could take account of such a difference.

See also People v. Belmontes (1983) 34 Cal.3d 335, 348 [193 Cal.Rptr. 882, 667 P.2d 686] (statement of reasons for sentencing under section 667.6); People v. Ramirez (1979) 25 Cal.3d 260, 276 [158 Cal.Rptr. 316, 599 P.2d 622] (inmate excluded from California Rehabilitation Center entitled to a statement of reasons); In re Bye (1974) 12 Cal.3d 96, 110 [115 Cal.Rptr. 382, 524 P.2d 854] (Narcotic Addict Evaluation Authority must provide statement of reasons supporting revocation of outpatient status); In re Sturm (1974) 11 Cal. 3d 258, 272-273 [113 Cal.Rptr. 361, 521 P.2d 97] (Adult Authority required to support denial of an application for parole release with a written statement of reasons); cf. Topanga Association for a Scenic Community v. County of Los Angeles (1974) 11 Cal.3d 506, 513-518 [113 Cal.Rptr. 836, 522 P.2d 12] (administrative grant of a zoning variance ‍‌‌​​​‌​‌​‌​‌​‌‌​‌​​​​​​‌‌‌​‌​​‌‌‌‌‌‌​‌‌​‌​​​​‌​​‍must be accompanied by findings); but see Van Atta v. Scott (1980) 27 Cal.3d 424, 444-446 [166 Cal.Rptr. 149, 613 P.2d 210] (trial judge not required to issue a written statement of reasons when own-recognizance release is denied); In re John H. (1978) 21 Cal.3d 18, 22-24 [145 Cal.Rptr. 357, 577 P.2d 177] (juvenile court not required to explain decision to commit a minor to the Youth Authority); People v. Edwards (1976) 18 Cal.3d 796, 802-805 [135 Cal.Rptr. 411, 557 P.2d 995] (trial judge not required to explain his reasons for denying a grant of probation).

People v. Herrera, supra, 127 Cal.App.3d 590, People v. Mitchell, supra, 152 Cal.App.3d 433, and People v. Shepeard, supra, 169 Cal.App.3d 580, are disapproved to the extent that they are inсonsistent with the views stated herein.

Associate Justice, Court of Appeal, Second District, assigned by the Chairperson of the Judicial Council.



Dissenting Opinion

MOSK, J.

I dissent.

I agree that the two-step analysis outlined in People v. Herrera (1982) 127 Cal.App.3d 590, 601 [179 Cal.Rptr. 694], sets out a workable and proper procedure for a trial court notified by the Board of Prison Terms (Board) that it has imposed a disparately high sentence. I also agree that at the first step, in which the court simply reviews the Board’s finding that a sentence is statistically disparate, the court should defer to the expertise of the Board entrusted by the Legislature with the task of analyzing data and reviewing comparable cases.

In the present case, the court rejected out of hand the Board’s finding of disparity without offering any concrete evidence to support a contrary conclusion and on the basis оf major misconceptions about the Board’s methodology. Yet the question whether a sentence is disparate is a factual rather *452than a discretionary one, and the Board’s careful review under statutory mandate should be accepted as correct in the absence of a strong showing to the contrary. As the majority conclude, the trial court should accept the Board’s finding unless on clear evidence it finds that the latter erred in selecting the appropriate comparison group or in determining that the sentence differs significantly from that imposed on most members of the group.

I part company with the majority, however, when it comes to outlining the trial court’s task in the second stage of its review of a Board finding of disparity. While the finding is entitled to great weight, that means only that the court should carefully reassess its sentencing order in light of the notification that it is at variance with the typical range of sentences in similar cases. The court retains discretion to deviate from that range if it deems appropriate in the particular case.

The majority hold that the trial court must recall a disparate sentence unless it is justified by substantial evidence of countervailing considerations. In effect, they infer from Penal Code section 1170 a presumption in favor of resentencing. Their interpretation of the provision, however, finds no support in either the language or policy enunciated by the Legislature.

The statute requires, in relevant part, “Within 120 days of receipt [of notice that a sentenсe is disparate], the sentencing court shall schedule a hearing and may recall the sentence and commitment previously ordered and resentence the defendant in the same manner as if the defendant had not been sentenced previously, provided the new sentence is no greater than the initial sentence. In resentencing under this subdivision the court shall apply the sentencing rules of the Judicial Council and shall consider the information provided by the Board of Prison Terms.” (§ 1170, subd. (f)(1).)

In interpreting statutory language, we must look to the words themselves, giving effect to their usual and ordinary import. (People v. Aston (1985) 39 Cal.3d 481, 489 [216 Cal.Rptr. 771, 703 P.2d 111]; Moyer v. Workmen’s Comp. Appeals Bd. (1973) 10 Cal.3d 222, 230 [110 Cal.Rptr. 144, 514 P.2d 1224].) On its face, the statute explicitly leaves recall of a disparate sentence within the trial court’s discretion: it “may” recall the sentence. Moreover, even if the court defers to the Board’s finding, in resentencing it must do so “in the same manner as if the defendant had not been sentenced previously” and is bound only to “consider” the information provided by the Board. Judicial discretion is preserved: the Judicial Council Sentencing Rules, which the court must follow in resentencing, permit the court to *453evaluate such factors as demeanor, personality, and attitude of the defendant in committing the crime. By contrast, the Legislature explicitly incorporated a presumption in Penal Code section 1170, subdivision (b), concerning the terms of imprisonment, directing, “When a judgment of imprisonment is to be imposed and the statute specifies three possible terms, the court shall order imposition of the middle term, unless there are circumstances in aggravation or mitigation of the crime.” (Italics added.) As we explained in City of Hueneme v. City of Oxnard (1959) 52 Cal.2d 385, 395 [341 P.2d 318], when a statute, with reference to one subject contains a given provision, the omission of such provision from a similar statute concerning a related subject is significant to show that a different intention existed.

Furthermore, as a matter of policy, trial courts should be guided but not placed in a straightjacket by the Board’s findings. The Board itself explained that it presents only a dry statistical study of what other judges have done in comparable cases; it makes no judgment regarding the appropriateness of any particular sentence, “dealing strictly with the benchmark established by other judges, and not any theoretical standard.” The Legislature’s decision to rest discretion in the trial court under section 1170 properly gives play to the court’s unique ability to take account of the personality, particular conduct, attitude, and demeanor of the defendant in committing the crime— factors the majority mislabel “subjective.” Although sentencing patterns are entitled to great weight, the trial judge remains in the best position to proportion the individual sentence to the particular defendant, within the limits of the Determinate Sentencing Act.

The majority’s formulation elevates statistical tendencies to the level of presumptions, as the present case illustrates: although there is no statutory presumption in favor of concurrent rather than consecutive sentences (People v. Reeder (1984) 152 Cal.App.3d 900, 923 [200 Cal.Rptr. 479]), because the statewide pattern favors concurrent sentences for robberies perpetrated on multiple victims in the same incident the majority require the trial court to justify its sentence after the fact just as though there were a presumption in the first instance. Presumptions of this kind should be established by the Legislature or the courts on the basis of substantive policy, not by numerical curves generated by a computer.

As the court emphasized in Herrera, “a finding by the Board that a sentence is ‘disparate’ does not mean that the sentence was imposed in violation of the law. . . . [A] motion brought by the Board pursuant to section 1170, subdivision (f) does not create a presumption that the sentence is ‘incorrect’ (unlawful) or that the prisoner is entitled to have his sentence *454recalled.” (127 Cal.App.3d at pp. 599-600.) The Board’s conclusions should function as a measuring stick, not a whipping rod for judges; the information that а sentence is disparate demands serious, careful consideration by the court, but only as an aid to its own exercise of discretion.

I am unpersuaded by the analogy the majority draw to our decisions in People v. Carl B. (1979) 24 Cal.3d 212 [155 Cal.Rptr. 189, 594 P.2d 14], and People v. Javier A. (1985) 38 Cal.3d 811 [215 Cal.Rptr. 242, 700 P.2d 1244], in which we held that a trial court must follow the recommendation of the Youth Authority in sentencing convicted juveniles absent “substantial evidence of countervailing considerations of sufficient weight to overcome the recommendation.” (Javier A., at p. 819; accord Carl B., supra, 24 Cal.3d at pp. 214-215.)

First, in those cases we applied the standard of review against the background of statutory language entirely silent as to the weight to be accorded Youth Authority recommendations. (See Welf. & Inst. Code, § 707.2.) By contrast, section 1170 addresses the matter and expressly vests ‍‌‌​​​‌​‌​‌​‌​‌‌​‌​​​​​​‌‌‌​‌​​‌‌‌‌‌‌​‌‌​‌​​​​‌​​‍the court with discretion to follow the recommendation or not, as it deems appropriate.

Second, although both the Board of Prison Terms and the Youth Authority are mandated to present findings concerning sentencing, their aims and procedures remain distinguishable in critical respects. The Youth Authority is a rehabilitative penal system, in which juveniles are sentenced indeterminately, and the authority confines its analysis to the amenability of a particular minor to treatment and rehabilitation. By contrast, the Board operates within a punitive system under the Determinate Sentencing Act, and analyzes the mathematical relationship of sentences rather than focusing on the particular qualities of individual defendants. Moreover, the Youth Authority presents its evaluation to the court before sentencing, whereas the Board review is initiated only after a defendant has been sentenced and committed to prison. The Legislature’s decision to structure review under section 1170 as retrospective strongly suggests it did not intend to accord the same weight to the Board’s findings as to those of the Youth Authority.

Finally, it is noteworthy that although the statute requires the trial court to state its reason at the time of sentencing for imposing an aggravated, enhanced, or consecutive term (§ 1170, subd. (c)), a similar requirement is conspicuously absent from subdivision (f). So long as the record shows that the court seriously considered the information provided by the Board, I would not impose a mandatory rule that a trial court rejecting the Board’s sentencing recommendation must in every case articulate its reason for doing so.

*455I would affirm the judgment.

Bird, C. J., and Lucas, J., concurred.

Respondent’s petition for a rehearing was denied September 18, 1986. Lucas, J., and Panelli, J., were of the opinion that the petition should be granted.