State v. Korsakov , 2000 Tenn. Crim. App. LEXIS 553 ( 2000 )

34 S.W.3d 534 (2000)

STATE of Tennessee
v.
Richard KORSAKOV.

Court of Criminal Appeals of Tennessee, at Knoxville.

July 13, 2000.

Lloyd A. Levitt, Chattanooga, TN, for appellant, Richard Korsakov.

Paul G. Summers, Attorney General & Reporter; Elizabeth B. Marney, Assistant Attorney General; William H. Cox, III, District Attorney General, and C. Parke Masterson, Jr., Assistant District Attorney General, for appellee, State of Tennessee.

No Permission to Appeal Applied for to the Supreme Court

OPINION

TIPTON, J., delivered the opinion of the court, in which WOODALL and GLENN, JJ., joined.

The defendant, Richard Korsakov, appeals his conviction for driving under the influence of an intoxicant, third offense. He contends that (1) the breath test results were inadmissible because the officer administering the test failed to observe him for twenty minutes; (2) the photocopies of certification and maintenance records were inadmissible for lack of compliance with the rules of evidence; (3) he should have been allowed to cross-examine the officer on the Horizontal Gaze Nystagmus (HGN) test for the purpose of impeachment; (4) a sealed, miniature bottle of cognac was irrelevant and, therefore, inadmissible; (5) the trial court improperly commented upon the evidence by instructing the jury to disregard the address on an exhibit's tag; and (6) the cumulative effect of the errors at trial prejudiced him. Because we hold that the results of the breath test are inadmissible, we reverse the judgment of conviction and remand the case to the trial court for further proceedings consistent with this opinion.

The defendant, Richard Korsakov, appeals as of right from his conviction by a jury for driving under the influence of an intoxicant (DUI), third offense, a Class A misdemeanor. The trial court sentenced the defendant to serve eleven months, twenty-nine days in the workhouse to be suspended after one hundred twenty days. It imposed a $1110 fine. The defendant contends that:

(1) the breath test results are inadmissible because the officer administering the test failed to observe him for the required twenty-minute time period;

(2) the photocopies of the intoximeter's certification and maintenance records are inadmissible because the introducing officer is not the custodian of those records and they are not the best evidence;

(3) the trial court erred in not allowing him to cross-examine the officer on the Horizontal Gaze Nystagmus (HGN) test for the purpose of impeachment;

(4) the sealed, miniature bottle of cognac was irrelevant and, therefore, inadmissible;

(5) the trial court improperly commented upon the evidence by instructing the jury to disregard the address on an exhibit's tag; and

(6) the cumulative effect of the errors at trial prejudiced him.

Because we hold that the trial court erred in admitting the breath test results, we reverse the judgment of conviction and remand the case to the trial court for further proceedings consistent with this opinion.

Officer James Hostetter of the Chattanooga Police Department testified that around 2:00 a.m. on July 31, 1997, he was patrolling on South Holtzclaw. He said that the road narrowed and that the car in front of him drifted to the right, hit the curb beside a telephone pole, and jerked back into the road. He said that at first, he thought that the car had hit the pole. He said that he began a check on the license tags to see if the car was stolen. He stated that the car began drifting back and forth across the center line and stopped in the crosswalk at the intersection with East 23rd Street. He said that the car turned left onto East 23rd and remained in the left lane. He said that he was still waiting on the results of the license tag check when the car almost passed Lisa's Country Kitchen, then turned back into the parking lot. He said that the car drove into a parking space, then its reverse lights came on. He said that he activated his blue lights and parked perpendicular to the car.

Officer Hostetter testified that he approached the defendant as he got out of the car and asked for his driver license. He said that he could smell alcohol on the defendant's breath and asked the defendant if he had been drinking. He said that the defendant admitted that he had been drinking. The officer said that the defendant was difficult to understand. He said that he asked the defendant if he had any medical problems and that the defendant reported none. He said that he asked the defendant to perform the walk-and-turn test and the one-leg stand, but the defendant replied that he would not do "acrobatics for any policeman." He said that he did not remember the defendant saying that he was fifty-seven years old and could not do the tests. He said that based upon what he had seen and the odor of alcohol, he arrested the defendant for DUI. He said that he inventoried the defendant's car and found a miniature, unopened bottle of cognac in the console between the front seats.

Officer Hostetter testified that he and the defendant arrived at the jail at 2:45 a.m. He said that the defendant consented to a breath test to determine his blood alcohol content. He said that he observed the defendant from 2:45 a.m. until 3:19 a.m. to make sure that the defendant did not eat, drink, smoke, vomit, or belch. He said that during this time, he stood across the counter from the defendant and filled out paperwork. He admitted that he was not looking at the defendant the entire time but stated that if anything had happened, he would have seen, heard, or smelled it. He said that the defendant did not eat, drink, smoke, vomit, or belch in the time before he administered the test at 3:19 a.m.

Officer Hostetter testified that he performed the breath test with an Intoximeter EC-IR, number 1799. He presented a photocopy of a certificate from the Tennessee Bureau of Investigation (TBI), which stated that intoximeter number 1799 had been tested and was certified on February 5, 1996. He also presented photocopies of the intoximeter's maintenance records, which reveal that it was tested on May 7, 1997, and on August 6, 1997. He said that he was trained and certified to use the intoximeter while at the police academy in 1996. He said that the machine appeared to be working properly on July 31, 1997, and that he performed the test according to the TBI's requirements. He said that he did not recall commenting that the machine was adjusting its temperature and that he had never seen it do that. He said that the test revealed that the defendant's blood alcohol content was .12 percent. He admitted that people react differently to alcohol but said that in his opinion, the defendant was under the influence of alcohol at the scene.

Special Agent Lanny Allen Wilder testified as follows: He is the supervisor of the Nashville TBI Crime Laboratory and oversees the statewide breath alcohol program. His duties include certifying instruments. He has testified as an expert in the operation of the Intoximeter EC-IR in Tennessee courts. Before an intoximeter is put into service, it comes to the TBI to be tested for accuracy. He tests the intoximeter with live subjects and compares the results to blood tests performed at the same time. He also blows known concentrations of different types of alcohol into the machine. The machine must be within the greater of five percent or .005 percent weight by volume of the actual alcohol concentration to be certified by the TBI. Intoximeter number 1799 had undergone this testing and was certified. Once an intoximeter goes into service, the TBI tests it with a known standard of alcohol every ninety days.

Agent Wilder testified that once consumed, alcohol is present in an individual's blood. The blood exchanges carbon dioxide for oxygen in the lungs, and then the person exhales the carbon dioxide along with any impurities. Alcohol in the blood changes into a gas in the lungs and is exhaled as an impurity. Thus, the alcohol measured by an intoximeter comes from the blood but through the breath. The scientific principle of Henry's Law states that the concentration of alcohol in the blood is 2,100 times greater than the concentration of alcohol in the same sample of air from the lungs. The intoximeter applies this principle to measure the blood alcohol content from a breath sample. On cross-examination, Agent Wilder agreed that the actual concentration of alcohol in a person's blood varies between 1,100 to 3,000 times greater than the same amount of air but that an intoximeter applies the ratio of Henry's Law, which has been in use for over forty years. He disagreed with a particular study that states that a .03 error, as opposed to an error of .005 percent weight by volume, may commonly result from the standard of 2,100.

Agent Wilder testified that the Intoximeter EC-IR is one of the newest, state-of-the-art models. The EC stands for electrochemical and the IR stands for infrared. The Intoximeter EC-IR measures alcohol in two different ways. The infrared detector continuously monitors alcohol in the breath and when the alcohol gets to a certain peak and levels off, the electrochemical unit takes a sample of the breath. This ensures a sample of deep lung air, which is closer to the exchange of gases in the blood, rather than air from the mouth or throat. If the intoximeter detects alcohol before it detects carbon dioxide, then the reading is from the throat rather than the deep lung area, and the machine does not give a result. The Intoximeter EC-IR has safeguards built into the machine that prevent it from giving a reading if the machine is not working properly. If the machine indicates that it is regulating the temperature, it means that it is not ready for the sample.

Agent Wilder testified that an intoximeter measures the alcohol content in a person's blood at the time they breathe into the machine. He admitted that a person can continue to absorb alcohol for thirty to ninety minutes after their last drink. He said that given the .005 percent weight by volume margin of error, a result of .12 percent could be .115 percent or .129 percent, but either would be above a .08 percent.

TBI Agent Raymond Siler, Jr. testified as follows: He had been the head of the Chattanooga Crime Laboratory for eighteen years. He performed the ninety-day testing of the breath instruments in Hamilton County. In performing this test, he first blew a blank sample into the machine. He then tested the machine with a known amount of alcohol. He tested intoximeter number 1799 on May 7, 1997, and August 6, 1997. To his knowledge, intoximeter number 1799 had never given an inaccurate result on a test.

I. ADMISSIBILITY OF BREATH TEST RESULTS

The defendant contends that the trial court erred in admitting the results of the breath test because the officer administering the test failed to observe him for twenty minutes as required by State v. Sensing, 843 S.W.2d 412, 416 (Tenn.1992). The defendant argues that rather than observing him, the officer filled out paperwork from the time that they arrived at the jail until the officer administered the test. The state contends that the officer observed the defendant through his peripheral vision while he stood across from the defendant and completed paperwork and that the officer affirmed that the defendant did not eat, drink, smoke, vomit, or belch during this time.

The defendant moved to suppress the results of the breath test pretrial, arguing that the officer failed to observe the defendant for the required twenty-minute period. The trial court found that the officer was directly facing the defendant and would have detected the defendant's movements through his peripheral vision. It noted that it did not believe that the officer was required to stare unblinkingly at the defendant for twenty minutes. The trial court found that the officer's actions sufficiently complied with Sensing.

In Sensing, our supreme court relaxed the evidentiary foundation needed to admit breath test results obtained from an Intoximeter 3000. Id. at 416. Due to the reliability and general acceptance of the testing device, the court held that a certified operator no longer had to know the scientific technology behind the Intoximeter 3000's operation. Id. Instead, the court established six prerequisites to admissibility. The officer who administered the test must show:

(1) that the tests were performed in accordance with the standards and operating procedure promulgated by the forensic services division of the Tennessee Bureau of Investigation,

(2) that he was properly certified in accordance with those standards,

(3) that the evidentiary breath testing instrument used was certified by the forensic services division, was tested regularly for accuracy and was working properly when the breath test was performed,

(4) that the motorist was observed for the requisite 20 minutes prior to the test, and during this period, he did not have foreign matter in his mouth, did not consume any alcoholic beverage, smoke, or regurgitate,

(5) evidence that he followed the prescribed operational procedure,

(6) identify the print-out record offered in evidence as the result of the test given to the person tested.

Id. The state has the burden of proving that the breath test complied with the Sensing requirements. State v. McCaslin, 894 S.W.2d 310, 312 (Tenn.Crim.App.1994). On appeal, we presume that the trial court's determination regarding the Sensing requirements is correct unless the evidence preponderates otherwise. State v. Edison, 9 S.W.3d 75, 78 (Tenn.1999). The defendant disputes that Officer Hostetter was able to prove the fourth requirement.

Initially, we note that the parties tried this case with an understanding that the Sensing requirements apply to the Intoximeter EC-IR. Sensing dealt with the Intoximeter 3000. The Sensing court noted that the Intoximeter 3000 was thoroughly tested by the TBI and used by law enforcement throughout the state. Sensing, 843 S.W.2d at 415. The Intoximeter 3000 measures the alcohol in the breath sample with an error margin of the greater of five percent or .005 percent weight by volume. Id. The instrument must be tested and certified by the TBI before use, installed pursuant to strict requirements, and retested every ninety days. Id. Those operating the Intoximeter 3000 must be certified by the TBI. Id. In the present case, TBI Agent Wilder testified that the Intoximeter EC-IR is one of the newest, state-of-the-art breath testing machines. He described the testing and certification process performed by his laboratory on every machine. He stated that the machine must have a margin of error of no more than five percent or .005 percent weight by volume to be certified by the TBI. He said that the TBI retests the accuracy of these machines every ninety days. He also described safeguards built into the machine that prevent it from giving a result if the machine is not working properly or if the breath sample is not from the deep lung area. Officer Hostetter testified that he was trained and certified by the TBI to operate the Intoximeter EC-IR. In light of this testimony, we believe that the Intoximeter EC-IR is as reliable as the Intoximeter 3000. Thus, the foundational prerequisites established in Sensing for test results from an Intoximeter 3000 also apply to the Intoximeter EC-IR.

The defendant argues that the officer must keep his eyes on the DUI suspect for the entire twenty-minute observation period required by Sensing. The state counters that Sensing does not require the officer to stare unblinkingly at the defendant for twenty minutes. In State v. Deloit, 964 S.W.2d 909 (Tenn.Crim.App.1997), the officer observed the defendant for fifteen minutes while he was performing field sobriety tests and then watched the defendant in the rearview mirror for an additional ten to thirteen minutes while he filled out paperwork in the front seat of the police car. The officer admitted that he could not see the defendant while he was writing. This court held that the officer's actions did not meet the twenty-minute observation requirement noting that the officer must "``continuously observe[] the test subject, with his or her eyes, for the entire twenty-minute observational period.'" Id. at 916-17 (quoting State v. Harold E. Fields, No. 01C01-9412-CC-00438, Williamson County, slip op. at 5, 1996 WL 180706 (Tenn.Crim.App. Apr. 12, 1996)) (brackets in original); see McCaslin, 894 S.W.2d at 311-12 (holding that being in the presence of the officer while in the backseat of the patrol car did not meet the fourth Sensing prerequisite).

In Fields, the officer remained in the room with the defendant but filled out paperwork and entered data into the Intoximeter 3000 during the observation period. We noted that "in light of the relaxed standard for admitting breath alcohol test results, the threshold requirements for admissibility of such test results must be scrupulously followed." Fields, slip op. at 4. The officer must be able to show that during the twenty-minute period, nothing happened that would affect the validity of the test. Id., slip op. at 5. We agree that an unblinking gaze for twenty minutes is not required. However, the officer must be watching the defendant rather than performing other tasks. The state seeks to distinguish Fields by pointing to Officer Hostetter's testimony that the defendant did not eat, drink, smoke, vomit, or belch during the observation period. We note, though, that Officer Hostetter admitted that he was not looking at the defendant while he was writing. Despite Officer Hostetter's confidence that he would have heard or smelled any of the listed activities that he did not see, we have previously observed that while "often a belch or regurgitation will produce a noise capable of being heard by another person, this is not always the case." Id. Officer Hostetter's belief that he would have heard or smelled anything he did not see does not satisfy the prerequisite that the defendant must be observed for twenty minutes.

The state likens this case to that of State v. Gregory L. Parker, No. M1999-00209-R3-CD, 1999 WL 1296018 Davidson County (Tenn.Crim.App. Dec. 30, 1999), in which the officer administered field sobriety tests to the defendant for six minutes, then observed the defendant for an additional nineteen minutes. This court held that the evidence supported the trial court's admission of the breath test results because the officer was near the defendant for the six minutes preceding the nineteen-minute observation period. Id., slip op. at 5. In Parker, the officer was focusing on the defendant during the six minutes that he administered the field sobriety tests to him. Here, however, Officer Hostetter was looking at the paperwork. We believe the evidence preponderates against the trial court's finding that the officer observed the defendant for the required twenty-minute period.

The erroneous admission of the breath test results is not harmless error. Although Officer Hostetter testified that he saw the defendant weave across the center line, hit the curb, stop in the crosswalk, and pass the restaurant before turning back into the parking lot, the defendant's cross-examination of the officer emphasized that the officer did not attempt to stop him until he was in the restaurant parking lot. Officer Hostetter admitted that the defendant properly positioned his car in the parking space. Officer Hostetter testified that the defendant smelled of alcohol and admitted that he had been drinking. The defendant refused to perform the walk-and-turn and one-leg-stand field sobriety tests. In McCaslin, this court determined that the erroneous admission of the breath test results was not harmless error where those results constituted the only scientific evidence presented at trial and the videotaped field sobriety tests were inconclusive. 894 S.W.2d at 312. In Deloit, the officer testified that he observed the defendant do a "rolling stop" at a stop sign and that the defendant smelled of alcohol, admitted that he had been drinking, and performed poorly on two field sobriety tests. We concluded that in the absence of the erroneously admitted breath test results, this evidence fairly balanced the defendant's testimony that he had performed well on the field sobriety tests and the testimony of the defendant's friends that he appeared sober that evening. Id. at 917. We held that a jury should decide the sufficiency of the remaining proof. Id. In this case, we believe that the question of whether the remaining evidence proves that the defendant was driving while intoxicated should be left for a jury.

We note that before Sensing, Tennessee required the party seeking to admit breath test results to show that (1) the test instrument used was capable of measuring or calculating the data being used by the interpreting expert and was based upon scientific principles accepted by the scientific community, (2) the instrument was functioning properly and was used properly at the time of the test, (3) the test was conducted by a person qualified to administer it, and (4) the interpretation of the test results must have been by a qualified person, including the person having an understanding of both the scientific principles behind and the operative functions of the test. See Pruitt v. State, 216 Tenn. 686, 692-93, 393 S.W.2d 747, 750-51 (1965); Fortune v. State, 197 Tenn. 691, 696-98, 277 S.W.2d 381, 383-84 (1955); Tenn. R. Evid. 702; Neil P. Cohen, et al., Tennessee Law of Evidence, §§ 401.19-.21, at 124-27, § 401.24, at 130-32, §§ 702.2-.3, at 458-64 (3d ed.1995). The relaxed evidentiary foundation provided by Sensing does not foreclose the possibility of laying an evidentiary foundation through expert testimony on the reliability of the scientific test performed. Deloit, 964 S.W.2d at 913. Such testimony would still have to establish that the party performing the test used the breath testing device properly, which would include observing the test subject before administering the test if an observation period is required for the device. Although we hold that the erroneous admission of the breath test results requires that we reverse the judgment of conviction, we will briefly address the defendant's other issues to provide guidance if the state seeks to retry the case.

II. ADMISSIBILITY OF CERTIFICATION AND MAINTENANCE RECORDS

The defendant contends that the photocopies of the Intoximeter EC-IR's certification and maintenance records, introduced as exhibits four, five, and six, are inadmissible because the officer through whom the state introduced the copies is not the custodian of those records and they are not the best evidence. The state contends that copies of public records constitute the best evidence of those records if authenticated by a witness who has personally compared the copy with the original. See Tenn.R.Evid. 1005. It argues that Agent Lanny Wilder testified that exhibit four was a copy of the original certification of Intoximeter EC-IR, number 1799. It maintains that exhibits five and six, the copies of the ninety-day maintenance records on machine number 1799, were introduced by the certifying public officer, Agent Raymond Siler.

Public records and reports are specifically excluded from the prohibition against hearsay by Rule 803(8), Tenn. R. Evid.:

Unless the source of information or the method or circumstances of preparation indicate lack of trustworthiness, records, reports, statements, or data compilations in any form of public offices or agencies setting forth the activities of the office or agency or matters observed pursuant to a duty imposed by law as to which matters there was a duty to report, excluding, however, matters observed by police officers and other law enforcement personnel.

Our supreme court has recognized that the initial TBI certification and the subsequent maintenance records of breath testing instruments are public records under Rule 803(8), Tenn.R.Evid. Sensing, 843 S.W.2d at 416. Unlike the business records exception, the public records exception does not require that the record be introduced through the testimony of a "custodian or other qualified witness." See Tenn.R.Evid. 803(6), (8); State v. Baker, 842 S.W.2d 261, 264-65 (Tenn.Crim.App.1992). In Baker, the defendant challenged the admission of his driving record as a business record through the testimony of a state trooper, arguing that the trooper was not the custodian of those records. Although agreeing with the defendant that the trooper did not adequately testify to the identity of the record, the mode of preparation, and whether the record was made in the ordinary course of business near the time of the event recorded as required by Rule 803(6), this court held that the trial court properly admitted the driving record under the public records exception in 803(8). Id. Furthermore, Sensing contemplates that an officer certified to perform the breath test will testify that the TBI certified the testing instrument and regularly tested it for accuracy. See 843 S.W.2d at 416. Thus, the state could introduce the certification and maintenance records through the testimony of Officer Hostetter.

The defendant also contests the introduction of the certification and maintenance records because they are copies of the actual documents and, therefore, not the best evidence of those records. See Tenn.R.Evid. 1002 (requiring that a party introduce the original of a writing to prove its contents unless a rule or statute provides otherwise). However, Rule 1005, Tenn.R.Evid., permits the use of a copy to prove the content of a public record:

The contents of an official record, or of a document authorized to be recorded or filed and actually recorded or filed, including data compilations in any form, if otherwise admissible may be proved by copy certified as correct in accordance with Rule 902 or testified to be correct by a witness who has compared it with the original. If a copy which complies with the foregoing cannot be obtained by the exercise of reasonable diligence, then other evidence of the contents may be given.

The Advisory Commission's comments to this rule note that copies "of public records are the best evidence if certified or otherwise authenticated under the rule."

We note that the copies of the TBI certification and maintenance records are not self-authenticating under Rule 902(4), Tenn.R.Evid., dealing with certified copies of public records, because the copies themselves contain no certification of their correctness. Furthermore, Officer Hostetter did not testify that he had compared the copies of the TBI certification and the maintenance records of Intoximeter EC-IR, number 1799 to the originals. In fact, he stated that he did not recall seeing the original of the TBI certification hanging near the intoximeter at the jail. Nevertheless, Agent Wilder testified that he recognized exhibit four as a copy of the original TBI certification of Intoximeter EC-IR, number 1799. Agent Siler testified that he performed the ninety-day maintenance tests on the breath testing machines in Hamilton County; that he had tested machine number 1799 on May 7, 1997, and on August 6, 1997; and that he recognized exhibits five and six as the certifications of those tests. The trial court did not err in admitting copies of the TBI certification and maintenance records on Intoximeter EC-IR, number 1799 rather than requiring the state to produce the originals.

III. CROSS-EXAMINATION ON HGN TEST

The defendant contends that the trial court erred in refusing to allow him to cross-examine Officer Hostetter about his scoring the defendant on the HGN test despite his saying that the defendant refused to take the field sobriety tests. He claims that he sought to cross-examine the officer about this in order to impeach his credibility. He maintains that impeaching the officer on this matter was part of his argument that the officer was exaggerating the majority of his findings relating to the defendant's intoxication. He also summarily argues that this evidence constitutes a prior inconsistent statement under Rule 613, Tenn.R.Evid. The state contends that the defendant waived cross-examination on this issue by objecting to any testimony concerning the HGN test, which resulted in a ruling that the officer's testimony on the field sobriety tests would be limited to the walk-and-turn and the one-leg-stand tests. It also argues that the officer's testimony would have prejudiced the defendant, and its absence is harmless in light of the overwhelming proof of the defendant's intoxication.

Before Officer Hostetter testified, the state sought to clarify whether he would be allowed to testify regarding the HGN test given to the defendant. The defendant objected to the officer testifying about the HGN test, stating that he was making the same objection that he did in the first trial, which ended in a hung jury. The state offered to elicit only the testimony that the officer asked the defendant to perform field sobriety tests and the defendant refused. The defendant argued that he did not refuse the HGN test but that, instead, the officer felt that he did not perform the test properly. The following colloquy ensued:

[Prosecutor:] I mean, I'm willing to go either way. I'll leave it up to the defense. Either I say "refused all tests," or "attempted to give one, didn't comply and refused all others." I don't care which way it goes, but I don't want to make it an appellate issue, so I'll go with whatever they want to do.

....

[Defense Counsel:] But what Officer Hostetter testified to is when he asked Mr. Korsakov to do the HGN test, that instead of following his instructions and complying with the test, that Mr. Korsakov just stared straight ahead and basically was unable to do the test. Again, since the HGN is not admissible—

The Court: All right. So he's not going to go into the HGN, ... why can't [the prosecutor] bring out that your client refused the other tests?

[Defense Counsel:] Oh, he can. That I'm not objecting to, I do not object to that, but I don't think it's a proper characterization to say he refused the HGN. He didn't refuse it; he just, in Officer Hostetter's opinion—

The Court: I think the best thing is [to say] that, He refused to do any field sobriety tests. .... Or you can say he refused to do the walk-and-turn or the one-leg stand.

....

[Prosecutor:] So we won't ... even mention the HGN?

The Court: So don't say "all" or anything. Just use those two specific [tests].

Despite securing this ruling, defense counsel mentioned in his opening statement that the officer's notations on a form relating to a test given to the defendant were inconsistent with the officer's testimony. During his cross-examination of Officer Hostetter, the defendant attempted to question him about the HGN test. In a bench conference, the state protested that the parties had "a specific ruling and a specific agreement not to talk about the HGN." Defense counsel argued that he thought that the court had ruled that the officer could testify that the defendant failed the walk-and-turn test and the one-leg-stand test but not that he had failed the HGN test as well. The court responded that it had limited the officer's testimony that the defendant refused the test. The state pointed out that it had not questioned the officer on the HGN test. The trial court refused to allow the defendant to cross-examine the officer on the HGN test, finding the test inadmissible.

In an offer of proof, Officer Hostetter testified that he attempted to give the defendant the HGN test. He explained that the test involves holding his finger twelve to fifteen inches away from the subject's face and asking the subject to follow the finger with his or her eyes. He said that he watches for the presence of three clues in each eye: (1) the subject's inability to follow his finger smoothly, (2) the onset of involuntary jerking before forty-five degrees, and (3) involuntary jerking when his finger is as far out as the eye can follow. He testified that with regard to the subject's ability to follow his finger smoothly, sometimes the subject is too drunk to see his finger. He stated that he gives the subject one point for the presence of each of the three clues in each eye. He said that a failing score would be a four, five, or six. He said that the defendant did not follow his finger but, instead, stared straight ahead. He said that in scoring the defendant's performance, he indicated the presence of all three clues for each eye. He said by marking the sheet in this way he intended to show that the defendant failed the test. He admitted that he should not have filled in the boxes indicating involuntary jerking. He said that when he realized that the defendant did not pass the test, he left the total blank and wrote a narrative. The narrative states "when asked if he saw my finger he constantly looked at my light or at me as though my finger was not there."

Denial of the defendant's right to effective cross-examination is "``constitutional error of the first magnitude'" and may violate the defendant's right to a fair trial. State v. Hill, 598 S.W.2d 815, 819 (Tenn.Crim.App.1980) (quoting Davis v. Alaska, 415 U.S. 308, 318, 94 S. Ct. 1105, 1111, 39 L. Ed. 2d 347 (1974)). "The propriety, scope, manner and control of the cross-examination of witnesses, however, rests within the sound discretion of the trial court." State v. Dishman, 915 S.W.2d 458, 463 (Tenn.Crim.App.1995); Coffee v. State, 188 Tenn. 1, 4, 216 S.W.2d 702, 703 (1948). This court will not disturb the limits that a trial court has placed upon cross-examination unless the court has unreasonably restricted the right. Dishman, 915 S.W.2d at 463; State v. Fowler, 213 Tenn. 239, 253, 373 S.W.2d 460, 466 (1963).

In this case, the trial court did not abuse its discretion by refusing to allow the defendant to cross-examine Officer Hostetter about scoring the defendant on the HGN test. Exclusion of evidence of the HGN test came at the defendant's insistence and was carefully crafted to avoid the implication that the defendant had refused the HGN test along with the other field sobriety tests. The state abided by this ruling in its direct examination of the officer. The defendant argues that he wanted to show that Officer Hostetter entered points on the HGN form in order to impeach the officer's testimony that the defendant refused to perform any field sobriety tests. On direct examination, Officer Hostetter did not testify that the defendant refused to perform any tests; instead, he testified that the defendant refused to perform the walk-and-turn and one-leg-stand tests. The trial court did not unreasonably limit the defendant's right to cross-examine the officer.

IV. ADMISSIBILITY OF THE COGNAC BOTTLE

The defendant contends that the trial court should not have permitted the state to introduce an unopened, miniature bottle of cognac found in his car. He argues that the bottle was irrelevant and could only show that the defendant had a propensity to drink. The state contends that the officer's discovery of the bottle in the defendant's car is relevant to the defendant's intoxication even though the bottle was unopened. It also argues that the defendant only objected to the bottle's relevance at trial and cannot now lodge the additional objection that the bottle constitutes propensity evidence under Rule 404(b), Tenn.R.Evid. Finally, the state maintains that even if the bottle is inadmissible, it did not prejudice the defendant in light of the overwhelming evidence of intoxication.

Evidence is relevant if it has "any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence." Tenn.R.Evid. 401. An unopened, sealed bottle of cognac has no bearing upon whether the defendant was intoxicated while driving on the morning in question. However, we believe that although irrelevant, the admission of the cognac bottle was harmless. Officer Hostetter testified that the defendant admitted that he had been drinking. The risk that the presence of a sealed miniature bottle affected the jury's view of the defendant's condition could only be minuscule.

We agree with the state that the defendant may not base his objection to the admission of evidence on relevance at trial and then base his argument on another ground, such as the violation of Rule 404(b), Tenn.R.Evid., on appeal. See State v. Miller, 668 S.W.2d 281, 285 (Tenn.1984). The defendant contended in his motion for a new trial that he was prejudiced by the admission of the irrelevant cognac bottle because it improperly suggested propensity. However, because the defendant did not raise 404(b) as a basis for his objection at trial, the trial court did not determine whether the evidence was admissible for a purpose other than to show propensity and did not weigh the probative value of the evidence when used for this purpose against the danger of unfair prejudice. See Tenn.R.Evid. 404(b). Thus, the defendant has waived his argument on this basis. See T.R.A.P. 36(a).

V. IMPROPER COMMENT UPON THE EVIDENCE

The defendant contends that the trial court invaded the province of the jury and improperly commented upon the evidence in violation of article IV, § 9 of the Tennessee Constitution when it instructed the jury to disregard the address on a tag attached to the cognac bottle. The state contends that the defendant acquiesced in the court's response to the jury. It also argues that the trial court's response was not an improper comment upon the evidence.

After retiring to deliberate, the jury submitted a question to the court noting that the tag attached to the cognac bottle lists the location at which it was found as 1501 East Main Street. The jury asked what this location represents. Before responding to the jury, the trial court discussed the matter with the parties, who agreed that the bottle was found in the defendant's car. The court stated that it would respond: "bottle found in car, address of no significance." After conferring with the defendant, defense counsel suggested that the proper response would be to tell the jury to rely on their own recollections about where the bottle was found. The trial court stated, "I won't put ``bottle found in defendant's car.' I'll just put ``address is of no significance.'" To which the defense responded, "That's fine." We do not believe that the trial court's comments to which the defense agreed regarding an admittedly irrelevant matter gives rise to any constitutional concern.

VI. CUMULATIVE ERROR

The defendant contends that the cumulative effect of the errors at trial denied his fundamental right to a fair trial, especially in light of the fact that his first trial ended in a hung jury. However, we see nothing deserving reversal other than the trial court's error in admitting the breath test results.

Based upon the foregoing and the record as a whole, we reverse the judgement of conviction and remand the case for further proceedings consistent with this opinion.