State v. Van Kirk , 306 Mont. 215 ( 2001 )

32 P.3d 735 (2001) 2001 MT 184

STATE of Montana, Plaintiff/Respondent,
v.
Marvin VAN KIRK, Defendant/Appellant.

No. 99-413.

Supreme Court of Montana.

Submitted on Briefs February 23, 2001. Decided September 6, 2001.

Edward Yelsa, Anaconda, MT, for Appellant.

Joseph P. Mazurek, Montana Attorney General, Tammy K. Plubell, Assistant Montana Attorney General; Michael B. Grayson, Deer Lodge County Attorney, Anaconda, MT, for Respondent.

Justice PATRICIA O. COTTER delivered the Opinion of the Court.

¶ 1 On January 19, 1999, a jury convicted Marvin Van Kirk (Van Kirk) of four criminal offenses: driving under the influence of alcohol, failing to carry proof of vehicle registration, operating a motor vehicle without liability insurance, and driving a vehicle with a suspended or revoked driver's license. Van Kirk appeals his convictions to this Court. We affirm.

¶ 2 Van Kirk raises three issues on appeal:

1. Did the District Court correctly conclude that the arresting officer had a particularized suspicion to initially stop Van Kirk's vehicle?

2. Did the District Court properly conclude that Van Kirk did not have a right to counsel prior to submitting to a breath test or performing field sobriety maneuvers?

3. Did the District Court commit reversible error by allowing the arresting officer to testify about Van Kirk's performance on the horizontal gaze nystagmus test?

FACTUAL BACKGROUND

¶ 3 On October 3, 1997, at approximately 8:00 p.m., Officer John Kelly (Officer Kelly) was giving a person a ride from the police station to his vehicle parked at the First Quarter Restaurant in Anaconda, Montana. While in the First Quarter Restaurant's parking lot, Officer Kelly observed a truck pull into the parking lot to his right. The driver looked in Officer Kelly's general direction, and then immediately pulled back out of the parking area. It appeared to Officer Kelly that the driver of the truck wanted to get away from him, so he decided to investigate by following the vehicle.

¶ 4 While following the truck, Officer Kelly observed that its speed was only seven to ten mph in a twenty-five mph zone. Officer Kelly also noticed that the driver was continuously watching him in the rearview mirror. Officer Kelly watched the vehicle shift from the edge of the roadway to the mid-point and across it, several times. On at least one occasion, the vehicle would have impeded any oncoming traffic. Based on his observations, Officer Kelly decided to initiate a traffic stop.

¶ 5 Officer Kelly approached the vehicle on foot after initiating the stop. When Van Kirk rolled down his window, Officer Kelly detected the moderate odor of an alcoholic beverage and noticed that Van Kirk's eyes were bloodshot. Officer Kelly also observed that there was a screwdriver, rather than a key, in the ignition of the truck. Officer Kelly asked to see Van Kirk's driver's license, but Van Kirk replied he did not have one. Officer Kelly later confirmed through the dispatcher that Van Kirk's license was revoked. Officer Kelly also asked to see the vehicle registration and proof of insurance. Van Kirk responded to this by handing him a map. Officer Kelly then requested that Van Kirk step out of his truck and perform some field sobriety tests, but Van Kirk declined because it was dark and the road surface was rough. Officer Kelly responded that Van Kirk could perform the tests at the police station under better conditions.

¶ 6 Officer Kelly decided to place Van Kirk under arrest and transport him to the police station. Upon reaching the police station, Officer Kelly read Van Kirk Montana's implied consent advisory form and asked him to submit to a breath test. Van Kirk agreed to the test and he registered a blood alcohol level of .115. Upon completion of the breath test, Officer Kelly requested that Van Kirk perform some field sobriety tests. Another officer videotaped Van Kirk's performance of the field sobriety maneuvers. Officer Kelly scored Van Kirk four points out of eight on the walk-and-turn test because Van Kirk missed touching his heel to toe twice on each direction, and took too many steps prior to and after his turn. Van Kirk performed well on the one-leg-stand test. Finally, Officer Kelly conducted the horizontal gaze nystagmus (HGN) test upon Van Kirk. Officer Kelly scored Van Kirk four points out of six on this test.

¶ 7 Upon completing the field sobriety test, Officer Kelly read Van Kirk his Miranda rights. Van Kirk informed Officer Kelly that he wanted to speak with an attorney, and Officer Kelly did not thereafter question Van Kirk. Van Kirk was charged with four offenses: (1) driving under the influence of alcohol, a misdemeanor; (2) failing to carry proof of vehicle registration, a misdemeanor; (3) operating a motor vehicle without liability insurance in effect; and (4) driving a vehicle with a suspended or revoked driver's license. On July 13, 1998, a justice court jury convicted Van Kirk of all four offenses, and he appealed to the District Court.

¶ 8 Van Kirk filed several pretrial motions and supporting briefs in District Court, including a motion to suppress evidence and a motion to dismiss the charges against him. Van Kirk contended that Officer Kelly lacked particularized suspicion to initiate an investigative stop. In addition, Van Kirk asserted that when Officer Kelly did not allow him to speak to an attorney prior to giving a breath sample or performing field sobriety maneuvers, Van Kirk was denied his right to counsel. The District Court denied these motions.

¶ 9 On January 19, 1999, the District Court held a jury trial on this matter. During trial, when the State asked Officer Kelly how Van Kirk scored on the HGN test, Van Kirk objected on the grounds that the State had not laid sufficient foundation for the testimony. Officer Kelly testified that he was certified to administer the HGN test, and that he properly administered the test. Officer Kelly also testified that he has a bachelor's degree in medical technology and had worked as a lab supervisor and technician at Montana State Hospital. The District Court overruled Van Kirk's objection, and allowed Officer Kelly to testify from his training and experience that there is a correlation between alcohol consumption and nystagmus. Officer Kelly then explained how Van Kirk scored on the HGN test. The jury convicted Van Kirk of all four offenses. He now appeals.

STANDARD OF REVIEW

¶ 10 We review a district court's denial of a motion to suppress to determine whether its findings of fact are clearly erroneous and whether its conclusions of law are correct. State v. Hatler, 2001 MT 38, ¶ 5, 304 Mont. 211, ¶ 5, 19 P.3d 822, ¶ 5. We review a district court's evidentiary rulings for an abuse of discretion. State v. Gustafson, 2000 MT 364, ¶ 14, 303 Mont. 386, ¶ 14, 15 P.3d 944, ¶ 14.

ISSUE 1

¶ 11 Did the District Court correctly conclude the arresting officer had a particularized suspicion to initially stop Van Kirk's vehicle?

¶ 12 Van Kirk argues the arresting officer did not have adequate particularized suspicion to initially stop his truck, that all the resulting evidence against him should be suppressed as fruit of an illegal seizure, and therefore the charges against him must be dismissed. The State counters that Van Kirk's apparent attempts to avoid the arresting officer, coupled with his driving at a slow rate of speed and crossing from the edge of the road across the middle of the road multiple times, combined to provide sufficient particularized suspicion to make an investigative stop of Van Kirk's vehicle. We agree.

¶ 13 The Fourth Amendment to the United States Constitution and Article II, Section 11, of the Montana State Constitution protect against unreasonable searches and seizures, including brief investigatory stops of vehicles. State v. Elison, 2000 MT 323, ¶ 15, 302 Mont. 228, ¶ 15, 14 P.3d 456, ¶ 15. Montana peace officers, however, are authorized to initiate an investigatory stop of any vehicle observed in circumstances that create a particularized suspicion that an occupant of the vehicle has committed or is committing a criminal offense. State v. Halvorson, 2000 MT 56, ¶ 8, 299 Mont. 1, ¶ 8, 997 P.2d 751, ¶ 8. The requirement that a particularized suspicion be found in order to justify an investigatory stop of a vehicle is codified in a Montana statute:

Investigative stop. In order to obtain or verify an account of the person's presence or conduct or to determine whether to arrest the person, a peace officer may stop any person or vehicle that is observed in circumstances that create a particularized suspicion that the person or occupant of the vehicle has committed, is committing, or is about to commit an offense.

Section 46-5-401, MCA.

¶ 14 Particularized suspicion does not require that an officer possess proof beyond a reasonable doubt that a crime has been, is, or is about to be committed before initiating an investigative stop. Hatler, ¶ 11. Particularized suspicion is quantitatively different and less stringent than probable cause to arrest or conduct a search. State v. Williamson, 1998 MT 199, ¶ 12, 290 Mont. 321, ¶ 12, 965 P.2d 231, ¶ 12. In order to show sufficient cause to stop a vehicle, the burden is on the State to show (1) objective data from which an experienced police officer can make certain inferences, and (2) a resulting suspicion that an occupant of a vehicle is or has been engaged in wrongdoing or was a witness to criminal activity. State v. Gilder, 1999 MT 207, ¶ 10, 295 Mont. 483, ¶ 10, 985 P.2d 147, ¶ 10.

¶ 15 The gravamen of the particularized suspicion standard is that the totality of the circumstances confronting the officer at the time of the investigative stop must give rise to a particularized and objective basis for suspecting criminal activity. Halvorson, ¶ 9. In evaluating the totality of the circumstances, this Court considers the quantity, or content, and quality, or degree of reliability, of the information available to the officer at the time of the investigatory stop. Gilder, ¶ 11.

¶ 16 Here, Van Kirk first drew suspicion by pulling into and then immediately pulling back out of a parking lot after spotting Officer Kelly's patrol car. It appeared to Officer Kelly that the driver of the vehicle wanted to get quickly away from him, so he decided to follow and observe this vehicle. While following Van Kirk's vehicle, Officer Kelly observed that, even though the speed limit was twenty-five mph, Van Kirk was going only seven to ten mph. Officer Kelly also observed Van Kirk's vehicle meandering from the edge of the roadway to the center of the roadway and across it several times. On at least one occasion, Van Kirk's vehicle would have impeded any oncoming traffic. Based on this erratic driving and slow speed, Officer Kelly decided to initiate a traffic stop.

¶ 17 Van Kirk argues that these facts are analogous to those in State v. Lafferty, 1998 MT 247, 291 Mont. 157, 967 P.2d 363, where this Court concluded there were insufficient facts to support a particularized suspicion. We disagree. In Lafferty, the officer observed Lafferty driving across the fog line twice, and upon it once. Lafferty, ¶ 13. Here, Van Kirk's crossing over the edge of the road was only one of several facts giving rise to Officer Kelly's particularized suspicion. Officer Kelly also observed Van Kirk's apparent desire to avoid him, Van Kirk's unusually slow speed, and Van Kirk's meandering from the edge of the road to across the center of the road. This constitutes substantially more suspicious behavior than that observed in Lafferty.

¶ 18 We conclude that, based on the totality of the circumstances, Officer Kelly had particularized suspicion to initiate an investigatory stop of Van Kirk's vehicle. Therefore, the District Court correctly denied Van Kirk's motion to suppress evidence based on his claim that Officer Kelly did not have sufficient grounds to stop his vehicle. The District Court is affirmed on this issue.

ISSUE 2

¶ 19 Did the District Court properly conclude that Van Kirk did not have a right to counsel prior to submitting to a breath test or performing field sobriety maneuvers?

¶ 20 Van Kirk argues that the District Court's failure to suppress evidence obtained after he demanded assistance of counsel warrants reversal of his convictions. The State counters that Van Kirk did not have a right to counsel before submitting to the breath test and field sobriety tests. The State argues further that Van Kirk's rights were not violated when Officer Kelly videotaped the breath test and field sobriety tests prior to giving his Miranda warnings.

¶ 21 The Fifth Amendment to the United States Constitution and Article II, Section 25 of the Montana Constitution provide that no person may be compelled to testify against himself in a criminal proceeding. The results of a breath test, however, are not self-incriminating communications. State v. Armfield (1984), 214 Mont. 229, 235, 693 P.2d 1226, 1229. Moreover, a person has no right to consult with counsel prior to deciding to take an alcohol breath test, because consent is deemed given as a matter of law. Section 61-8-402, MCA; State v. Strand (1997), 286 Mont. 122, 126, 951 P.2d 552, 553.

¶ 22 The privilege against self-incrimination does not extend to real or objective evidence. State v. Thompson (1989), 237 Mont. 384, 386, 773 P.2d 722, 723. In Thompson, this Court rejected the argument that an officer must read a DUI suspect his Miranda rights before audio-videotaping the suspect performing field sobriety tests. Thompson, 237 Mont. at 387, 773 P.2d at 724. A mere request that the suspect perform a series of sobriety tests, done without any interrogation of the suspect, does not constitute a custodial interrogation. Id.

¶ 23 Here, Officer Kelly arrested Van Kirk alongside the road and transported him to the police station in his patrol car. At the station, Officer Kelly read Van Kirk Montana's implied consent advisory form and asked Van Kirk to submit to a breath test. Van Kirk agreed to the breath test. Next, Officer Kelly requested that Van Kirk complete some field sobriety maneuvers while another officer videotaped his performance. Upon completing the field sobriety tests, Officer Kelly read Van Kirk his Miranda rights. After Van Kirk told Officer Kelly he wanted to speak with an attorney, Officer Kelly did not question Van Kirk.

¶ 24 We hold that Van Kirk was not denied his right to counsel, or his right against self-incrimination. Accordingly, the District Court is affirmed on this issue.

ISSUE 3

¶ 25 Did the District Court commit reversible error by allowing the arresting officer to testify about Van Kirk's performance on the horizontal gaze nystagmus (HGN) test?

¶ 26 Van Kirk argues the District Court abused its discretion by allowing Officer Kelly to testify about Van Kirk's performance on the HGN test, because Officer Kelly was unable to meet the foundation requirements set forth in Hulse v. State, Dept. of Justice, 1998 MT 108, 289 Mont. 1, 961 P.2d 75. The State counters that even if it was error for the District Court to allow the HGN testimony, it was harmless error because there was overwhelming evidence to convict Van Kirk of DUI.

¶ 27 In order to lay the proper foundation for admitting the results of the HGN test, the State, in addition to showing that the test was properly administered, must also establish a scientific basis for the reliability of the test results. Hulse, ¶ 72. Before HGN test results can be introduced into evidence, a qualified expert must explain the underlying scientific basis of the correlation between alcohol consumption and nystagmus. Id.

¶ 28 Officer Kelly has a bachelor's degree in medical technology and worked as a lab supervisor and technician at Montana State Hospital. Officer Kelly was certified to administer the HGN test and he testified that he properly administered the test to Van Kirk. However, nothing in the evidence establishes that Officer Kelly was specially trained or educated, or that he had adequate knowledge to qualify him as an expert able to explain the correlation between alcohol consumption and nystagmus, the underlying scientific basis of the HGN test. Pursuant to our holding in Hulse, we agree with Van Kirk that the District Court abused its discretion in permitting Officer Kelly to offer the HGN results into evidence without the State first establishing the requisite scientific basis for the test's reliability. We next examine whether the error was prejudicial.

¶ 29 Section 46-20-701(1), MCA, provides that not all error committed by a district court in a criminal proceeding is reversible error:

A cause may not be reversed by reason of any error committed by the trial court against the convicted person unless the record shows that the error was prejudicial.

We have held that error is prejudicial, and requires reversal, if a reasonable possibility exists that the inadmissible evidence might have contributed to a conviction. State v. Gray (1983), 207 Mont. 261, 268, 673 P.2d 1262, 1266. However, we have at times applied a slightly different standard when confronted with errors that arguably implicate constitutional rights, relying on Chapman v. California (1967), 386 U.S. 18, 87 S.Ct. 824, 17 L.Ed.2d 705.

¶ 30 In Chapman, the United States Supreme Court held that the test for harmless constitutional error is whether the court can declare its belief that the error was harmless beyond a reasonable doubt. Chapman, 386 U.S. at 24, 87 S.Ct. 824. The Court in Chapman rejected the notion that all federal constitutional errors, regardless of the facts and circumstances, are harmful and require automatic reversal of a conviction. Id. at 22, 87 S.Ct. 824. We have applied the Chapman harmless error rule to cases in which federal constitutional error is involved. State v. Rothacher (1995), 272 Mont. 303, 312-13, 901 P.2d 82, 87; State v. McKenzie (1980), 186 Mont. 481, 532-33, 608 P.2d 428, 458. However, over time it appears we have interchanged our harmless error rule expressed in State v. Gray, supra, with the Chapman harmless error rule, resulting in some inconsistency and confusion over the proper test to be applied in a harmless error analysis.

¶ 31 In State v. McKenzie, supra, we addressed whether the error in giving jury instructions on rebuttable presumptions, which effectively shifted the burden of proving each element of the crime from the State to the defendant, was harmless. We weighed what appeared to us to be three definable approaches to harmless error under various United States Supreme Court cases, and concluded that the proper test for determining federal constitutional errors under Chapman was the "overwhelming weight of the evidence" test. We expressed our belief that confining our review to one component of the case "[i]n artificial isolation," would result in a lopsided view of the case on appeal. McKenzie, 186 Mont. at 533-34, 608 P.2d at 458.

¶ 32 We reaffirmed our commitment to the "overwhelming weight of the evidence" test in Brodniak v. State (1989), 239 Mont. 110, 115, 779 P.2d 71, 74, referring to that test as "[o]ne method used by this Court to ascertain whether there is a reasonable possibility that the inadmissible evidence contributed to the verdict." In Brodniak, we applied a mixed test which borrowed language from both McKenzie and Gray, concluding that we must look both to the overwhelming evidence and the totality of the circumstances in assessing whether there is a reasonable possibility that the inadmissible evidence contributed to the verdict. Brodniak, 239 Mont. at 115, 779 P.2d at 74.

¶ 33 In State v. Fuhrmann (1996), 278 Mont. 396, 925 P.2d 1162, we expressed the belief that the federal harmless error rule and Montana's harmless error rule were "essentially the same," and concluded:

[t]hat in either case overwhelming evidence of a defendant's guilt can render harmless a district court's error. McKenzie, 608 P.2d at 458; Brodniak, 779 P.2d at 74.

Fuhrmann, 278 Mont. at 407, 925 P.2d at 1169.

¶ 34 Thus, it is clear that, over time, the "overwhelming evidence" analysis eclipsed the more substantive inquiry of whether the erroneously admitted evidence might have contributed to the conviction. The above cases demonstrate a recent inclination to simply tally the quantity of the admissible evidence of guilt, instead of evaluating the qualitative impact the inadmissible evidence might have had on the finder of fact.

¶ 35 There are two problems with this type of "scorecard" analysis. First, the test always involves "adding up" the admissible evidence on a case-by-case basis, to determine if it is overwhelming enough to transcend the effect of the inadmissible evidence, thereby ignoring the weight of the tainted evidence in favor of the sum of the other evidence. This ad hoc type of analysis results in an unpredictable and subjective framework which provides little guidance to parties in deciding what evidence to offer at trial. Moreover, the "overwhelming evidence" test may also have the unintended consequence of inviting the State to offer inadmissible yet damaging evidence in a strong case, even if by happenstance, since the worst that can happen is that the error is noted but deemed harmless. Neither result is desirable.

¶ 36 If we are to preserve the substantial rights of the accused and the fundamental fairness of the criminal trial, and provide consistent guidance to practitioners in the criminal practice, we must do more than simply quantify the admissible evidence offered by the State on a case-by-case basis. We therefore deem it appropriate to formally adopt an approach to harmless error issues which, first, defines for the trial court and the parties the type of error in question, and second, depending on the type of error defined, sets forth a uniform standard against which the error must be analyzed to determine if the error is sufficiently prejudicial to compel reversal of a conviction. We began this process last year when we distinguished between trial and structural errors, in State v. LaMere, 2000 MT 45, ¶¶ 39-50, 298 Mont. 358, ¶¶ 39-50, 2 P.3d 204, ¶¶ 39-50. We now complete the analysis.

¶ 37 Today, we adopt a two-step analysis to determine whether the alleged error prejudiced the criminal defendant's right to a fair trial and is therefore reversible. The first step in the analysis is an inquiry of whether the claimed error is categorized as "structural" error or "trial" error.

¶ 38 "Structural" error is that type of error that "[a]ffects the framework within which the trial proceeds, rather than simply an error in the trial process itself." Arizona v. Fulminante (1991), 499 U.S. 279, 310, 111 S.Ct. 1246, 1265, 113 L.Ed.2d 302, 331; LaMere, ¶ 48. Such error is typically of constitutional dimensions, precedes the trial, and undermines the fairness of the entire trial proceeding. Because of its nature, it cannot be qualitatively or quantitatively weighed against the admissible evidence introduced at trial. Structural error is presumptively prejudicial and is not subject to harmless error review jurisprudentially or under our harmless error statute found at § 46-20-701, MCA. LaMere, ¶ 50.

¶ 39 Structural error is automatically reversible and requires no additional analysis or review. Examples of structural error include errors in the jury selection process (LaMere); total deprivation of the right to counsel (Gideon v. Wainwright (1963), 372 U.S. 335, 83 S.Ct. 792, 9 L.Ed.2d 799); and lack of an impartial trial judge (Tumey v. Ohio (1927), 273 U.S. 510, 47 S.Ct. 437, 71 L.Ed. 749). See, generally, the discussion in LaMere, ¶ 23.

¶ 40 As opposed to "structural" error, "trial" error is that type of error that typically occurs during the presentation of a case to the jury. Such error is amenable to qualitative assessment by a reviewing court for prejudicial impact relative to the other evidence introduced at trial. Trial error is not presumptively prejudicial and therefore not automatically reversible, and is subject to review under our harmless error statute, § 46-20-701(1), MCA.

¶ 41 Thus, the first step in the inquiry is whether the error at issue is structural error or trial error. If structural, the inquiry ends and the verdict is reversed. If the error is the more typical "trial" error, then the analysis proceeds to the second step, which involves the determination of whether the error was harmless under the circumstances.

¶ 42 As a threshold matter, once a convicted person raises and establishes that the evidence in question was erroneously admitted and has alleged prejudice under the "reasonable possibility" test, it then becomes incumbent on the State to demonstrate that the error at issue was not prejudicial. We have previously held that the test of prejudicial error is whether there is a reasonable possibility that the inadmissible evidence might have contributed to a conviction. Gray, 207 Mont. at 268, 673 P.2d at 1266. It is how the State demonstrates that the error was not prejudicial that we address next.

¶ 43 Because our focus returns to the question of whether there was a reasonable possibility that the erroneously admitted evidence might have contributed to the conviction, it is time to abandon the use of the "overwhelming evidence" test in answering the question of whether an error is harmless. Such a subjective analysis of the volume of the evidence, geared as it is to quantity, simply does not adequately assess the qualitative impact of tainted evidence. Therefore, we overrule our previous decisions in McKenzie, Brodniak, and Fuhrmann to the extent that they applied the "overwhelming weight of the evidence" test in analyzing harmless error. We choose instead to utilize the "cumulative evidence" test in answering this inquiry. This test looks not to the quantitative effect of other admissible evidence, but rather to whether the fact-finder was presented with admissible evidence that proved the same facts as the tainted evidence proved. This test, applied by the United States Supreme Court in Harrington v. California (1969), 395 U.S. 250, 89 S.Ct. 1726, 23 L.Ed.2d 284, is inarguably more restrictive than the "overwhelming evidence" test. However, we believe its application, modified to incorporate the qualitative standards we set forth below, will bring us closer to gauging the potentially prejudicial impact of the tainted evidence than the "overwhelming evidence" test has done in the past.

¶ 44 In making its proof under the "cumulative evidence" standard, the State must direct us to admissible evidence that proved the same facts as the tainted evidence. Moreover, the State must also demonstrate that the quality of the tainted evidence was such that there was no reasonable possibility that it might have contributed to the defendant's conviction. By way of example only, assume a case where the defendant is charged with armed robbery of a liquor store. At trial, the police officer repeats the statements of an unidentified, non-testifying store patron to the effect that the defendant was present in the liquor store at the time of the offense. The State may be able to demonstrate that there is no reasonable possibility that this inadmissible hearsay evidence might have contributed to the defendant's conviction, where the defendant's admission to investigators that he was, in fact, present in the store at the time of the offense merely to purchase alcohol, was properly admitted. On the other hand, if the tainted evidence is the admission of the defendant's involuntary confession to the robbery, the State would be hard-pressed to demonstrate that, qualitatively, there is no reasonable possibility that this evidence might have contributed to the defendant's conviction, even though there was other evidence tending to prove that the defendant committed the crime.

¶ 45 We readily acknowledge that there will be cases in which there was no other admissible evidence proving the same facts that the tainted evidence proved, making the burden of producing cumulative evidence of the fact impossible. Clearly, if the only evidence tending to prove an element of the crime is tainted, then reversal will be compelled. Assume, for example, that the accused is charged with robbery under § 45-5-401(1)(b), MCA. The only evidence of the defendant's threat to inflict bodily injury on the liquor store clerk (the threat being an essential element of the crime charged) is the repetition by the investigating officer of the hearsay statement of a non-testifying witness. Under this circumstance, the threat element was proven only by inadmissible evidence. That element having not been proved by other admissible evidence, the error in admitting the tainted evidence cannot be harmless as a matter of law, and reversal is compelled.

¶ 46 However, in those cases where the tainted evidence does not go to the proof of an element of the crime charged, and there is no other admissible evidence tending to prove the particular fact at issue, the admission of the tainted evidence will be deemed harmless only if the State demonstrates that no reasonable possibility exists that the admission of the tainted evidence might have contributed to the defendant's conviction. Again, by way of example, suppose the State offers evidence in a DUI case that the defendant is a convicted child molester. As there would be no other admissible evidence which would prove this fact, and since the tainted evidence does not go to the proof of an element of DUI, the State would have to demonstrate that, qualitatively, there was no reasonable possibility that the tainted evidence might have contributed to the defendant's conviction—a virtually impossible burden to carry, given the highly inflammatory nature of child molestation evidence. If, on the other hand, the tainted evidence consisted of a passing reference to the fact that the defendant went through bankruptcy years earlier, the State could well demonstrate that, qualitatively, there was no reasonable possibility that the tainted evidence might have contributed to the defendant's conviction.

¶ 47 Thus, restated, we hold that, in order to prove that trial error was harmless, the State must demonstrate that there is no reasonable possibility that the inadmissible evidence might have contributed to the conviction. To do this, the State must demonstrate that the fact-finder was presented with admissible evidence that proved the same facts as the tainted evidence and, qualitatively, by comparison, the tainted evidence would not have contributed to the conviction. In those cases in which no such cumulative admissible evidence was presented, then we must determine whether the tainted evidence went to the proof of an element of the crime charged or, by contrast, to some fact not involving an element of the crime. If there was no cumulative evidence presented as to a fact proving an element of the crime charged, then the error in admitting the tainted evidence which proved that element cannot be considered harmless, the qualitative assessment is never reached, and the court's decision will be reversed. If the evidence in question did not prove an element of the crime, then the State must demonstrate that, qualitatively, there is no reasonable possibility that the tainted evidence might have contributed to the defendant's conviction.

¶ 48 Applying this test in the present case, the improper admission of the HGN evidence would be categorized as "trial" error under the above definition. Moreover, the HGN evidence was offered to prove that the defendant was "under the influence," which is an element of the DUI offense. The State therefore must demonstrate that (a) there was other admissible (cumulative) evidence demonstrating the "under the influence" element of the crime charged; and (b) that, qualitatively, no reasonable possibility exists that the tainted evidence might have contributed to the defendant's conviction. We conclude that the State demonstrated both elements here.

¶ 49 The fact-finder was presented with other admissible (cumulative) evidence that proved the same fact as the tainted (HGN) evidence, which was Van Kirk's intoxication. Officer Kelly testified to Van Kirk's slow, erratic, and suspicious driving. Once the ensuing stop was initiated, Officer Kelly observed that Van Kirk had bloodshot eyes and smelled of alcohol, and when asked to produce his vehicle registration and proof of insurance, Van Kirk handed over a map. At the police station, Van Kirk did not perform well on the walk-and-turn sobriety test, and he submitted to a breath test which indicated his blood/alcohol content was .115. On the basis of this other admissible (cumulative) evidence of intoxication, the State met its burden of proving that Van Kirk was under the influence. Moreover, we conclude that the introduction of the inadmissible HGN evidence was harmless because, qualitatively, and in comparison to the admissible evidence presented demonstrating that Van Kirk was under the influence, there is no reasonable possibility that it contributed to the conviction. Accordingly, Van Kirk's right to a fair trial was not prejudiced.

¶ 50 We will never be able to completely eliminate from judicial review the factual context surrounding each claimed error. However, we conclude that the harmless error test announced here, to be applied to the review of all cases after the date of this decision, will help to insure that the substantial rights of the defendant are not prejudiced by the admission of tainted evidence, and will bring the consistency to our future harmless error analyses that has been lacking in the past.

¶ 51 The District Court is affirmed.

We Concur: KARLA M. GRAY, Chief Justice, and JAMES C. NELSON, JIM REGNIER, TERRY N. TRIEWEILER, W. WILLIAM LEAPHART, and JIM RICE, JJ.