People v. Williams (2002) 28 Cal. 4th 408 (121 Cal.Rptr.2d 854; 49 P.3d 203)
[No. S098552. July 15, 2002.]
THE PEOPLE, Plaintiff and Respondent, v. STEVEN VAUGHN WILLIAMS, Defendant and Appellant.
(Superior Court of Shasta County, No. 97F9335, William Gallagher, Judge.)
(The Court of Appeal, Third Dist., No. C031921, 89 Cal. App. 4th 85.)
(Opinion by Brown, J., expressing the unanimous view of the court.)
Linda J. Zachritz, under appointment by the Supreme Court, for Defendant and Appellant.
Bill Lockyer, Attorney General, David P. Druliner and Robert R. Anderson, Chief Assistant Attorneys General, Jo Graves, Assistant Attorney General, John A. O'Sullivan, J. Robert Jibson and Charles A. French, Deputy Attorneys General, for Plaintiff and Respondent.
Rockard J. Delgadillo, City Attorney (Los Angeles), Debbie Lew, Assistant City Attorney, and Sunnie Lee Daniels, Deputy City Attorney, for Office of the Los Angeles City Attorney as Amicus Curiae on behalf of Plaintiff and Respondent. [28 Cal. 4th 410]
A jury convicted defendant of driving while under the influence. The People's case included the result of a preliminary alcohol screening (PAS) test showing that defendant's breath registered a blood-alcohol [28 Cal. 4th 411] content of .181 percent. fn. 1 The Court of Appeal held the PAS test result should have been excluded because the California Highway Patrol failed to comply substantially with title 17 of the California Code of Regulations (all references to title 17 are to title 17 of the California Code of Regulations), but deemed the error harmless and affirmed the conviction. We granted review to determine whether the absence of substantial compliance with the regulations justifies a blanket exclusion of PAS results or goes merely to the weight of the evidence.
FACTUAL AND PROCEDURAL BACKGROUND
On December 27, 1997, Gary Pickle's neighbor, Russell Bailey, hosted a party. While the party was in progress, someone drove a pickup truck into Pickle's driveway, spun a tire-squealing "brodie," and sped away. After the driver repeated this maneuver several times at half-hour intervals, Pickle confronted Bailey and a few of his guests, including defendant, whom Pickle had seen driving the truck on other occasions. Pickle's request that he not be disturbed by late-night driving antics appeared to upset defendant, who was agitated and belligerent. Fifteen minutes later, at approximately 1:45 a.m., Pickle heard the truck in his driveway again and called the police.
At approximately 2:02 a.m., Shasta County Deputy Sheriff Scocca arrived and saw a truck with the lights on idling at the side of the road. When she parked near the truck, the driver--defendant--initially ducked under a blanket but emerged moments later disheveled, smelling of alcohol, and slurring his words. No alcohol containers were found in the truck.
California Highway Patrol (CHP) Officers D'Arcy and Barrett arrived at 2:05 a.m. After defendant performed poorly on two field sobriety tests, he agreed to provide a breath sample by blowing into an Alco Sensor IV, which produced a reading of .181 percent. Officer D'Arcy arrested defendant for driving under the influence of alcohol. Defendant refused Officer D'Arcy's request that he submit to a chemical test as required by California's implied consent law.
The Shasta County District Attorney filed an information charging defendant with driving while under the influence (Veh. Code, § 23152, subd. (a)) and driving with a suspended license (id., § 14601.5, subd. (a)). The information further alleged defendant had suffered three prior driving under the influence violations (see id., § 23175) and two prior convictions within five years for driving with a suspended license (id., § 14601). [28 Cal. 4th 412]
At trial, defendant moved to exclude the PAS result, as the testing procedures did not conform to the requirements of title 17 and the evidence was more prejudicial than probative. (Evid. Code, § 352.) Testimony at an in limine hearing established the following:
The CHP had formally approved the Alco Sensor IV as a device to perform PAS tests. The Alco Sensor IV is readied for use by inserting a clean mouthpiece. The device then shows the date, time, and temperature, and indicates it is either unable (NOGO) or ready to test (TEST). If ready, an air blank then records a .000 percent result. Officer D'Arcy followed this procedure; after he observed defendant for approximately 13 minutes, during which time defendant did not burp, vomit, spit up, or drink any liquids, defendant blew into the mouthpiece.
Officer D'Arcy had used the Alco Sensor IV in the field for more than two years. After participating in an eight-hour training session, he received certification by the state Department of Justice in the use of a similar breath-testing device that was used at the local jail. Officer D'Arcy also received approximately 20 minutes of practical training with the Alco Sensor IV; it produced results consistent with those recorded by the machine on which he was initially trained.
The accuracy of the Alco Sensor IV used to screen defendant was verified by periodic testing with a standard .10 percent alcohol solution. The device was deemed accurate if the test solution produced a reading ranging from .09 to .11 percent; if not, it had to be recalibrated. The machine's first calibration, on January 28, 1997, produced a .11 percent reading. Although within the permissible range, the machine was recalibrated to record an accurate .10 percent result. All 19 tests conducted between January 1997 and February 1998 produced readings within the accepted range of .09 to .11 percent. The eight calibrations immediately prior to defendant's test, including the last test on November 17, 1997, produced readings slightly below .10 percent.
After the in limine hearing, the trial court admitted the evidence. The court found the evidence complied with the foundational requirements described in People v. Adams (1976) 59 Cal. App. 3d 559, 561 (Adams), and People v. Bury (1996) 41 Cal. App. 4th 1194, 1202 (Bury). These cases upheld admission of breath tests measuring blood-alcohol content where (1) the testing device was in proper working order, (2) the test was properly administered, and (3) the operator was competent and qualified. (Bury, supra, at p. 1202; Adams, supra, at p. 561.)
The instant trial court questioned whether there was even substantial compliance with the title 17 regulations. Nevertheless, the testing satisfied [28 Cal. 4th 413] the Adams/Bury foundational prerequisites. The court observed (1) the device had been calibrated frequently; (2) it had worked properly on many occasions and would not produce a result if improperly administered; and (3) the officer was qualified and competent to administer the test. The trial court therefore concluded the PAS evidence was more probative than prejudicial. The court observed defendant's refusal to submit to the required chemical test rendered the preliminary test even more probative than otherwise. The court confirmed its ruling when defendant raised an Evidence Code section 352 objection after trial.
I. THE COURT OF APPEAL OPINION
The Court of Appeal found the admission of the PAS results erroneous, though harmless. The Court of Appeal, contrasting the administrative regulations governing PAS tests with the validation methods used in this case and the test administered to defendant, concluded these actions failed to substantially comply with the regulations and ruled the test results were inadmissible. The discrepancies cited by the Court of Appeal included (1) the failure to test the instrument every 10 days or after the testing of every 150 subjects, whichever came first; (2) the failure to comply with the regulations specifying these tests be performed by either employees of a forensic alcohol laboratory or persons who have completed the requisite training, who must then forward the results to a forensic alcohol laboratory; (3) the officers' failure to observe the suspect continuously for 15 minutes; (4) the failure to test two separate breath samples that reveal blood-alcohol concentration within .02 grams per 100 milliliters of each other; and (5) the failure to describe the result at trial in terms of alcohol concentration in the blood. There was also no evidence Officer D'Arcy had been trained by a forensic alcohol analyst or trainee or that his training included the subjects described in the regulations. (Tit. 17, § 1221.4, subd. (a)(5).)
The Court of Appeal's response to noncompliance was exclusion. "Exclusion of PAS test results in every drunk driving case involving an Alco Sensor IV will deter intentional reliance upon a flawed system that, despite the best intentions and sincere efforts of the vast majority of members of the [CHP], will continue to deliver untrustworthy test results in every drunk driving case until it is appropriately corrected. . . . [T]he [CHP] will be able to produce . . . scientifically valid evidence once it brings its training and maintenance program into compliance with Title 17." The Court of Appeal nevertheless affirmed defendant's conviction, finding the admission of the test results to be harmless error as defined by People v. Watson (1956) 46 Cal. 2d 818, 836. [28 Cal. 4th 414]
[1a] The Adams court authorized admitting breath test evidence after a showing of (1) the reliability of the instrument, (2) the proper administration of the test, and (3) the competence of the operator. (Adams, supra, 59 Cal. App. 3d at p. 567.) To meet these requirements, the evidence would be admitted upon either a showing of compliance with the title 17 regulations or independent proof of the three elements. (Adams, at p. 567.) fn. 2 Subsequent courts have uniformly followed Adams. (See Bury, supra, 41 Cal. App. 4th at p. 1201; Coniglio, supra, 39 Cal. App. 4th at p. 681; People v. Sangani (1994) 22 Cal. App. 4th 1120, 1137; Davenport v. Department of Motor Vehicles (1992) 6 Cal. App. 4th 133, 140; People v. Perkins (1981) 126 Cal. App. 3d Supp. 12, 18; People v. French (1978) 77 Cal. App. 3d 511, 522 (French).)
We have formulated "generally accepted rules by which the reliability and thus the relevance of scientific evidence is determined." (People v. Harris (1989) 47 Cal. 3d 1047, 1094, citing People v. Kelly (1976) 17 Cal. 3d 24.) In Kelly, we explained a party seeking to introduce evidence based on a new scientific technique needed to qualify the technique as scientifically valid. (Kelly, at p. 30.) Even for techniques thus established, the proponent must "demonstrate that correct scientific procedures were used in the particular case." (Ibid.) fn. 3 Adams applied this latter principle in the specific context of breath tests, and we therefore consider it a proper application of Kelly.
Essential to Adams was the principle that admissibility depends on the reliability and consequent relevance of the evidence, not the precise manner in which it was collected. Compliance with regulations is sufficient to support admission, but not necessary. Noncompliance goes only to the weight of the evidence, not its admissibility. (Adams, supra, 59 Cal. App. 3d at p. 567.)
 There are two ways to construe the Court of Appeal opinion below: (1) the court imposed an exclusionary rule to encourage the police to comply with the administrative regulations, or (2) the court found the regulations to be coextensive with the Adams foundational requirements. As we explain, either construction is problematic. [28 Cal. 4th 415]
The more likely interpretation is that the court created a new exclusionary rule, excluding the evidence without regard to its relevance. The Court of Appeal candidly noted at the beginning of its analysis that "[its] concern focuse[d] less on creating evidentiary contests and more on a government agency's intentional failure to comply with mandatory duty." The Court of Appeal objected to the agency's apparent indifference to the regulations. "[T]he [CHP] has designed and implemented training and maintenance programs and procedures . . . which do not satisfy the requirements of Title 17. . . . Yet the [CHP] is under a mandatory duty to comply with Title 17." The opinion invoked the "exclusionary rule" as "a judicially created remedy designed to deter such [police carelessness]."
The exclusion of relevant evidence, however, is barred by the California Constitution's Right to Truth-in-Evidence provision, unless otherwise compelled by the federal Constitution. (Cal. Const., art. I, § 28, subd. (d).) By adopting this provision in 1982, the voters indicated that excluding evidence is not an acceptable means of deterring police misconduct. (In re Lance W. (1985) 37 Cal. 3d 873, 887.) Even prior to the enactment, courts recognized that absent a constitutional error the failure to follow the regulations in obtaining evidence was not grounds for exclusion. (People v. Perkins, supra, 126 Cal. App. 3d at Supp. 18; French, supra, 77 Cal. App. 3d at p. 522; People v. Rawlings (1974) 42 Cal. App. 3d 952, 956.)
The Court of Appeal's analysis improperly focused not on the propriety of admitting the evidence, but on Officer D'Arcy's carelessness in collecting it. The court then warned that "peace officers and their agencies would be mistaken to assume they may seek haven in the good faith exception to the exclusionary rule, for example, and there find a license to be casual or, worse, careless." The defect in this analysis is that the evidence was not admitted under a good faith exception to the exclusionary rule, but due to its relevance. Insofar as the Court of Appeal applied an evidentiary touchstone other than the relevance of the evidence, its analysis was incorrect.
Defendant offers a different construction of the court's analysis. He equates the administrative regulations and the Adams foundational requirements, explaining "[t]itle 17, essentially, specifies and regularizes the Adams foundational requirements concerning the competency of PAS tests." There is some basis for finding the Court of Appeal so held. The opinion emphasized the Adams conclusion: " '[T]he validity of the test itself is to be determined in accordance with general scientific standards . . . .' (People v. Adams, supra, 59 Cal. App. 3d at p. 567, italics added.) Those scientific [28 Cal. 4th 416] standards are embodied in Title 17, and the [CHP] will be able to produce such scientifically valid evidence once it brings its training and maintenance program into compliance with Title 17." According to this interpretation, noncompliance renders PAS results irrelevant as a matter of law.
There are several problems with thus construing the Court of Appeal's opinion. For one thing, the Adams court rejected the defendant's claim "that the calibration requirement [of title 17] goes to the essence of the substantive value of the tests, and that evidence taken in the absence of statutorily mandated safeguards is incompetent . . . ." (Adams, supra, 59 Cal. App. 3d at p. 563.) Rather than equate the regulatory standards with the minimum showing of reliability, Adams expressly held that title 17 compliance and the tripartite foundational requirements were distinct and independent means to support the admission of test results. Compliance with the regulations was sufficient: "The regulations are an expressed standard for competency of the test results; in effect, they are a simplified method of admitting the results into evidence." (Adams, at p. 567.) On the other hand, although the regulations are a standard of competency, they are not the only standard. Even absent compliance with the regulations, the People could obtain admission of the evidence through the general foundational requirements: "[T]he prosecution then must qualify the personnel involved in the test, the accuracy of the equipment used and the reliability of the method followed before the results can be admitted." (Ibid.)
The Court of Appeal held that the requisite level of compliance with title 17 was substantial rather than total, in accord with Adams's conclusion that evidence could be admissible despite noncompliance. "While Adams may authorize the admission of test results where substantial compliance with Title 17 is shown, it does not authorize the negation of a mandatory duty, where, as here, substantial compliance is not shown. To hold otherwise would render Title 17 a nullity . . . ." But Adams did not deem the test results admissible due to substantial compliance with title 17; the word "substantial" does not even appear in the majority opinion. fn. 4 The Adams court upheld the admission of the results due to their reliability, demonstrated by their meeting the foundational requirements, notwithstanding noncompliance with the regulations. (Adams, supra, 59 Cal. App. 3d at p. 567.) [28 Cal. 4th 417]
Furthermore, Adams did not provide any indication there was even substantial compliance with the calibration requirements. fn. 5 After noting the regulation ordered testing by the usual operator either weekly or following every 100 subjects, whichever came first, and the consequent duty to report the results to a licensed laboratory, the Court of Appeal merely observed that the operator had not complied with this regulation. (Adams, supra, 59 Cal. App. 3d at pp. 562-563.) There was no indication of how often the device was tested, and therefore no evidence of how substantial was the compliance with the calibration regulation. If substantial compliance had been a basis for the Adams court's decision, it would have cited the substantial nature of the compliance.
Finally, Adams expressly rejected the notion that the noncompliance undermined the reliability of the results. "Nor did [defendants] attempt any showing that the noncompliance affected the test results in any way, let alone rendered the results inaccurate." (Adams, supra, 59 Cal. App. 3d at p. 567.) Accordingly, the standards of reliability described by the Adams foundational requirements are not coextensive with the title 17 regulations. [1b] We conclude the holding of Adams, that the breath test results are admissible upon a showing of either compliance with title 17 or the foundational elements of (1) properly functioning equipment, (2) a properly administered test, and (3) a qualified operator, is the better approach. fn. 6
 In light of these standards, we find the trial court properly exercised its discretion in admitting the test results. (People v. Ashmus (1991) 54 Cal. 3d 932, 971.) Although the Alco Sensor IV was not tested with the frequency demanded by the regulations, the machine always performed within the acceptable range, and the slight inaccuracies usually underreported the amount of alcohol present. These facts recall French, where the test results were .19, .16, and .17 percent, with a procedural irregularity causing the second result to be .01 percent lower than it should have been. (French, supra, 77 Cal. App. 3d at p. 517.) French noted that not only did the error inure to the defendant's benefit by lowering the defendant's second result by .01 percent, but that the actual results of .19, .16 and .17 percent "were well above the .10 level" of presumptive intoxication. (Id. at p. 520.) The level of presumptive intoxication is now .08 percent, and the Alco Sensor IV here produced results with an error margin [28 Cal. 4th 418] of .01 or less in all 19 checks. From this record of reliability, the trial court could reasonably conclude the results of defendant's test, which exceeded the level of presumptive intoxication by .101, more than 10 times the machine's greatest inaccuracy, tended to establish defendant's intoxication.
The court could likewise conclude Officer D'Arcy's observation of defendant for 13 instead of 15 minutes, and his decision to take one instead of two tests, did not deprive the results of the reliability required for them to be relevant. fn. 7 Similarly, because the officer operated the machine as instructed and his misadministration could not produce false positive results, his unfamiliarity with the machine's operating theory did not compel the exclusion of the results.  (See fn. 8.) The trial court properly exercised its discretion in admitting the results. fn. 8
Although we reject the Court of Appeal's legal conclusions, we share its concern that laxity in complying with the regulations may undermine the reliability of the test. The trial court said the challenged evidence "push[ed] the outside of the envelope on the admissibility of PAS tests." Compliance with the regulations, by contrast, guarantees the People quick and certain admission of evidence, eliminating laborious qualification, critical cross-examination, and the risk of exclusion. Furthermore, compliance will ensure that the tests retain their reliability, and thus their relevance and admissibility, in the future.
We therefore affirm the judgment of the Court of Appeal.
George, C. J., Kennard, J., Baxter, J., Werdegar, J., Chin, J., and Moreno, J., concurred.
FN 1. The figure reflects the percent of alcohol in a person's blood based upon grams of alcohol per 210 liters of breath. (Veh. Code, § 23152.)
FN 2. As the Court of Appeal explained in Bury, title 17 regulations apply to PAS tests that determine the concentration of alcohol on the blood but not those that determine only its presence. (Bury, supra, 41 Cal. App. 4th at p. 1202.) The Bury court properly rejected as dicta the finding in Coniglio v. Department of Motor Vehicles (1995) 39 Cal. App. 4th 666, 677-681 (Coniglio) that title 17 never applies to PAS tests.
FN 3. The parties have not disputed the general scientific validity of breath testing. (See Bury, supra, 41 Cal. App. 4th at p. 1202.)
FN 4. The Adams dissent, which favored the imposition of an exclusionary rule to deter noncompliance, used the term in other contexts. (Adams, supra, 59 Cal. App. 3d 559, 572-573 (dis. opn. of Rattigan, J.).)
FN 5. Although there was no challenge in Adams to the administration of the test or the qualifications of the operator, all three foundational elements must be established for the evidence to be admissible; the chain is no stronger than its weakest link.
FN 6. The decision in Coombs v. Pierce (1991) 1 Cal. App. 4th 568, on which defendant relies, is not to the contrary. The People there did not establish the three foundational requirements (or compliance with title 17), and thus the test results were inadmissible. (Coombs, at p. 580.)
FN 7. Nothing found in the truck supports the inference that defendant drank, smoked, or vomited in the two minutes prior to the officers' observation.
FN 8. Although it is usually the People who seek to introduce breath test results, defendants also will be able to introduce relevant evidence in accordance with today's decision. "Whether . . . evidence is favorable to the People or to the defendant, the clear import of the 'Truth-in-Evidence' provision is that the trier of fact be given the opportunity to credit or discredit relevant evidence." (People v. Wood (1989) 207 Cal. App. 3d. Supp. 11, 17.)