STATE of Arizona, Appellee,
v.
Bobby Joe GARRISON, Appellant.
No. 3977.
Supreme Court of Arizona, In Banc.
September 20, 1978.*256 Bruce E. Babbitt, Former Atty. Gen., John A. LaSota, Jr., Atty. Gen. by William J. Schafer, III, Carol Benyi, Asst. Attys. Gen., Phoenix, for appellee.
Robert L. Murray, Tucson, for appellant.
STRUCKMEYER, Vice Chief Justice.
Bobby Joe Garrison was convicted by jury of the crime of first degree murder and sentenced to life imprisonment. He appeals.
On the morning of October 24, 1976, the body of Verna Marie Martin was found on the desert near Tucson, Arizona. Her death was caused by strangulation and her body was mutilated by bite marks. On October 27, while the case was being investigated, Pima County Sheriff detectives interviewed appellant, who was seen with the deceased on the night of October 22. On December 22, 1976, after two months of investigation, an indictment was returned charging appellant with murder.
Appellant urges that the trial court erred in failing to suppress certain statements made at his interview with the detectives on October 27, for the reason that he had not at that time been advised of his Miranda rights. It is the State's position that the statements made by appellant on October 27 were made in the course of a police investigation and were not made under custodial circumstances.
Statements made while a person is being interviewed as part of the investigation of a crime are admissible without Miranda warnings. State v. Bainch, 109 Ariz. 77, 505 P.2d 248 (1973); State v. Mumbaugh, 107 Ariz. 589, 491 P.2d 443 (1971). In State v. Bainch, we said:
"The fact that an officer may be suspicious of an individual is not the test as to whether Miranda warnings must be given prior to questioning, nor is the mere presence of a police officer to be considered a restraint on the suspect's liberty. The vital point is whether, examining all the circumstances, the defendant was deprived of his freedom of action in any significant manner, and the defendant was aware of such restraint. In the latter instance the Miranda warnings are required to be given before the statements of the defendant may be received in evidence against him." 109 Ariz. 77, 79, 505 P.2d 248, 250 (1973).
We do not think that the questioning of appellant on October 27, 1976 rose to the level of a custodial interrogation. Appellant came to the sheriff's office voluntarily at the request of Detective Rick Lingel of the Pima County Sheriff's Department. He traveled to the sheriff's office in his own automobile. That the interrogation took place principally at the police station does not in itself require that Miranda warnings be given.
"Such a noncustodial situation is not converted to one in which Miranda applies simply because a reviewing court concludes that, even in the absence of any formal arrest or restraint on freedom of movement, the questioning took place in a `coercive environment.' Any interview of one suspected of a crime by a police officer will have coercive aspects to it, simply by virtue of the fact that the police officer is part of a law enforcement system which may ultimately cause the suspect to be charged with a crime. But police officers are not required to administer Miranda warnings to everyone whom they question. Nor is the requirement of warnings to be imposed simply because the questioning takes place in the station house, or because the questioned person is one whom the police suspect. Miranda warnings are required only where there has been such a restriction on a person's freedom as to render him `in custody.' It was that sort of coercive environment to which Miranda by its terms was made applicable, and to which it is limited." Oregon v. Mathiason, 429 U.S. 492, 495, 97 S. Ct. 711, 714, 50 L. Ed. 2d 714, 719 (1977).
The trial court did not err if it concluded that the circumstances of the questioning took place in a "`coercive environment'", but that Miranda warnings were *257 not required. The police had tried to contact appellant as early as October 25, 1976, a fact which appellant knew. They then waited two days before attempting to contact him again. This is indicative of the fact, and must have been so understood by appellant, that the police did not consider there was an urgent, pressing need to interview him. When appellant arrived at the station, he signed in as a visitor. At the start of the interrogation, he was told that the officers wanted to find out how long appellant had known the deceased and their activities on the evening that he had been seen with her. While appellant was interviewed over a period of some six hours, the record does not show that any pressure whatsoever was put on him during that time. We do not consider the length of time as critical for the reason that part of the time was consumed when appellant went out for breakfast, part while waiting to take a polygraph test which he agreed to take, and part used in a trip to and from appellant's home.
We conclude, as we have before, that absent clear and manifest error, the trial court's determination of the admissibility of an accused's statements will not be upset by the appellate court. State v. Jordan, 114 Ariz. 452, 561 P.2d 1224 (1976); State v. Toney, 113 Ariz. 404, 555 P.2d 650 (1976).
Appellant complains of this portion of the prosecuting attorney's closing argument:
"Like I said before, if you are going to come back with a verdict of not guilty, ladies and gentlemen, you better have a reason, or you better have a reasonable explanation for all this evidence that's against the Defendant. You're turning this man loose and he will walk out the door, ladies and gentlemen, and as I said before, the evidence suggests in this case a total lack of regard for human life, such a lack of regard, ladies and gentlemen, if you don't expect such a crime could happen, or that a criminal necessarily, in committing such a crime as this, necessarily is going to stop at just this one death, you better think about that before you come in this courtroom and say not guilty. You think about that evidence I have presented to you, it's an important matter, and Mr. Murray tries to say no, it's not important anymore, her life is over, but it's very important for the rest of the lives in this community. Very important."
We do not think it was unreasonable for the prosecution to tell the jury as it did in the first sentence of the quoted argument that if it was going to come back with a verdict of not guilty, it had better have a reasonable explanation for all the evidence against the defendant. The appellant, however, directs the principal thrust of his claim of error to the remainder of the quoted argument, interpreting it to mean that the prosecution was saying there was a probability the appellant would take more lives if he were found not guilty and allowed to go free.
Appellant relies on State v. Makal, 104 Ariz. 476, 455 P.2d 450 (1969). There, the prosecution said of the defendant, "[S]ociety can't afford to have Mr. Makal take the life of any other innocent victims" and "Don't arrive at a verdict which will give Mr. Makal the opportunity to kill again." Id. at 478, 455 P.2d at 452. Those statements had to be considered in the light of Makal's plea of not guilty by reason of insanity and testimony in the case that persons who were found not guilty by reason of insanity were back on the streets soon afterward. The argument in Makal asked the jury to disregard the law by ignoring his plea of insanity.
The most that can be said about the prosecution's argument in the instant case is that the prosecution is telling the jury that the evidence shows a total lack of regard for human life and that a criminal who commits such a crime is not necessarily going to stop at one death. In the light of the total circumstances of how this homicide was committed, the argument that the criminal who committed it is not necessarily going to stop, while an emotional appeal, is one which is a permissible inference to be drawn from the nature of the evidence and did not unfairly prejudice appellant.
*258 Appellant complains of the testimony of an expert witness, Homer Richardson Campbell, Jr., a dentist, that there is an eight in one million probability that the teeth marks found on the deceased's breasts were not made by appellant. The witness, Campbell, is a board certified specialist in forensic dentistry and a member of the American Society of Orthodontology. His testimony is that the wounds in the deceased's breasts had ten points of similarity with appellant's teeth. He testified:
"My conclusion was that the bite marks on the deceased, and the bite marks produced by the model that I received, were consistent, the marks were consistent with those being made by the teeth that I received."
He further testified:
"* * * the probability factor of two sets of teeth being identical in a case similar to this is, approximately, eight in one million, or one in two hundred and fifty one in one hundred and twenty-five thousand people."
Appellant points to People v. Collins, 68 Cal. 2d 319, 66 Cal. Rptr. 497, 438 P.2d 33 (1968), which he asserts "is exactly on point with the situation in the present case." We, however, do not think so. In People v. Collins, the complaining witness was robbed by a blond woman wearing a ponytail who was driven from the scene of the robbery by a male Negro wearing a mustache and beard. At the trial, the prosecution called an instructor of mathematics at a state college, who testified that assuming the robbery was committed by a Caucasian woman with a blond ponytail whose companion was a Negro with a beard and mustache, there was an "overwhelming probability that the crime was committed by any couple answering such distinctive characteristics." The prosecution then, without presenting any statistical evidence whatsoever, argued that there was a probability of but one chance in twelve million that any couple would possess the distinctive characteristics of the defendants. On appeal, the court held that the "technique" employed by the prosecutor could only lead to wild conjectures without demonstrated relevance to the issues presented.
In the instant case, Dr. Campbell obtained the figure of eight in one million not from personal mathematical calculations, but from "articles written in the journals of the American Academy of Forensic Sciences" and two books, and "there are articles written throughout the literature that do mention the possibility or the numerical values of finding two [sets of teeth] of the same."
Arizona's Rules of Evidence, 17A A.R.S., effective September 1, 1977, provides by Rule 803(18) that statements contained in published treatises, periodicals, or pamphlets on the subject of medicine are not excluded by the hearsay rule. Although the Arizona Rules of Evidence were not in effect at the time of the trial of this case, we do not think the admission of the eight in a million statement is reversible error. Were we to reverse on this ground, it would only result in a retrial at which the same evidence would be admitted since Dr. Campbell's testimony, obtained from published treatises and periodicals, would be admissible under Rule 803(18). Courts should not engage in such futile practices.
We have stated before the principle on which Dr. Campbell's expert testimony was admitted for consideration by the jury:
"An expert is one whose opinions depend upon special knowledge with which he can assist the jury * * * He need not have the highest degree of skill or knowledge, but that lack of degree goes to the weight of his testimony before the trier of fact and not to admissibility. City of Phoenix v. Brown, 88 Ariz. 60, 352 P.2d 754 (1960); Udall, Arizona Law of Evidence, § 23." State v. Macumber, 112 Ariz. 569, 570, 544 P.2d 1084, 1085 (1976).
Finally, appellant complains that the trial court erred in failing to instruct the jury that if it found that witnesses for the State had caused the destruction of evidence, it could infer that the destroyed evidence was against the interest of the State. The instruction requested by appellant was one similar to that approved in State v. Willits, *259 96 Ariz. 184, 393 P.2d 274 (1964), and arises out of the evidence presented by Dr. Campbell.
Dr. Campbell performed a number of experiments to determine whether appellant's teeth matched the bite marks on the deceased's breasts. He made a model of appellant's teeth. The model was mounted on a dental articulator, a mechanical device which duplicates the movement of the human jaw. Then it was closed on wax, leaving an impression of appellant's teeth bite. A clear acetate sheet was placed over the wax impression and the bite marks traced onto the acetate. The acetate sheet was then laid over photographs of the deceased's breasts and the bite marks on the acetate sheet were compared with the bite marks found on the photographs. Dr. Campbell in the course of his experiments made about 50 wax impressions and about 30 acetate plates. Only three impressions and four acetate plates were produced at the trial. The remainder were destroyed. Dr. Campbell testified that he threw away probably 26 acetate plates and that they were thrown away "because essentially they have no value at all" and because "they didn't reproduce the marks as seen in the photographs." The reason some of the impressions and acetate sheets didn't reproduce the marks as seen in the photographs was because the bite on the wax was not made from the same angle or the same force as those on the deceased's breasts.
In State v. Willits, supra, if the evidence had not been destroyed, it might have tended to exonerate the defendant. But that is not true in the instant case. The wax impressions and acetate plates which were destroyed had no evidentiary value whatsoever. They were simply material from which no conclusions could be made, either as to innocence or guilt, because they did not duplicate the conditions under which the wounds on the deceased's breasts were inflicted. They would not have tended to exculpate appellant had they been preserved. Cf. State v. Hannah, 120 Ariz. 1, 583 P.2d 888 (1978). The court below did not err in refusing to give the requested instruction.
Judgment affirmed.
HAYS and HOLOHAN, JJ., concur.
GORDON, Justice (concurring in part, dissenting in part):
Although I concur with the majority on the Miranda and jury instruction issues, I dissent from their position on the closing argument and Dr. Campbell's testimony.
At the close of his argument to the jury, the prosecutor stated:
"You're turning this man loose and he will walk out the door, ladies and gentlemen, and as I said before, the evidence suggests in this case a total lack of regard for human life, such a lack of regard, ladies and gentlemen, if you don't expect such a crime could happen, or that a criminal necessarily, in committing such a crime as this, necessarily is going to stop at just this one death, you better think about that before you come in this courtroom and say not guilty. You think about that evidence I have presented to you it's an important matter, * * * it's very important for the rest of the lives in this community. Very important."
Clearly this type of argument preys on the fears of jurors, inviting them to prevent hypothetical future crimes by this defendant even though the evidence as to the crime charged may not be beyond a reasonable doubt. Although we have not previously considered this precise argument, an analogous situation was presented in State v. Makal, 104 Ariz. 476, 455 P.2d 450 (1969). There the prosecutor asked the jury to disregard insanity as a defense because
"He is essentially dangerous to other people; he is very dangerous to himself. We can't afford society can't afford to have Mr. Makal take the life of any other innocent victims. * * * Don't arrive at a verdict which will give Mr. Makal the opportunity to kill again." Id. at 478, 455 P.2d at 452.
*260 In Makal, Justice Struckmeyer reversed, quoting Farris v. Commonwealth, 209 Va. 305, 163 S.E.2d 575, 577 (1968):
"When evidence introduced is not of a subsequent act but of a possible future act, it does not shed any material light on an accused's mental state at the time of the offense charged. It can only have relation to the possibility or even probability that an accused will in the future commit a criminal act or will be a danger to society, and such evidence tends to destroy by fear the recognized defense of not guilty by reason of insanity." State v. Makal, 104 Ariz. at 477-478, 455 P.2d at 451-452.
"Evidence of what defendant's future conduct might be could only serve to divert the attention of the jurors from the real issue in the case * * *." Farris v. Commonwealth, supra at 577.
Obviously the prejudicial effect is the same regardless of whether the prosecutor urges the jury to disregard the state's burden of proof or the defendant's insanity defense. See, e.g, Grant v. State, 194 So. 2d 612 (Fla. 1967); Lime v. State, 479 P.2d 608 (Okl.Cr. 1971); Potter v. State, 511 P.2d 1120 (Okl.Cr. 1973). Furthermore, I cannot sanction an argument which is unsupported by the evidence properly before the jury. See DR 7-106(C)(1), Rule 29(a), 17A A.R.S.; State v. Filipov, 118 Ariz. 319, 576 P.2d 507 (1977).
I agree with the majority that "[a]n expert is one whose opinions depend upon special knowledge with which he can assist the jury." In fact, it is precisely because the value of an expert's opinion lies in his mastery of a field that will aid the jury that I am unable to agree with the majority's conclusion on the admissibility of the evidence probability presented by Dr. Campbell.
A reading of Dr. Campbell's testimony reveals that, not only did he not perform any of his own mathematical calculations in reaching the eight in one million figure, but also that he was unaware of the formula utilized to arrive at that figure other than that it was "computerized", and that he was ignorant of the statistical weight assigned to each variable used in the equation. (See discussion below). In short, it is obvious that while Dr. Campbell may have a great deal of expertise in the actual comparison techniques of bite-mark identification, he is totally out of his field when the discussion turns to probability theory.
Since Dr. Campbell was able to testify that the figure was obtained from articles and books in the field, the majority presses into service the newly adopted hearsay exception for "learned treatises" to justify admission of the testimony, in spite of Dr. Campbell's lack of expertise. The majority's theory is apparently that so long as probability studies have been performed and published in the past by experts in the practical application of probability theory, it is not necessary that the expert on the witness stand be capable of independently generating the results reached by the published study. I cannot agree.
Preliminarily, I have serious reservations as to the applicability of Rule 803(18) 17A A.R.S. After discussing the circumstances in which statements contained in learned treatises are admissible, the last sentence of Rule 803(18) reads as follows, "If admitted, the statements may be read into evidence but may not be received as exhibits." It seems to me that this sentence qualifies the exception with a requirement that the articles and books within which the relevant statements are contained be physically present in the courtroom and that the witness read the statement directly from the book. Of course, this construction of the sentence does nothing more than affirm the well established requirement that the contents of a writing be proven by production of the original writing. Arizona Rules of Evidence, Rule 1002 17A A.R.S. See also Rule 1001(3) 17A A.R.S. for the definition of "original".
Dr. Campbell's testimony provides a classic example of why a witness' memory of the contents of a writing is not a preferred method of proving those contents. As indicated in the majority opinion, Dr. Campbell was unsure as to precisely where he obtained *261 the figure "eight in one million". My independent research reveals that of the two treatises which he could name as containing statistical information, only Warren and Harvey, Dental Identification and Forensic Odontology (1976) lists any figures on the uniqueness of a bite-mark. Rather than the eight in one million figure vouched for by Dr. Campbell, though, that treatise, at page 139, contains the figure eight in one hundred thousand.[1]
Assuming that the jury took seriously Dr. Campbell's "eight in one million" apparently erroneous probability figure, and further assuming that the members of the jury tried it out on the estimated population of Pima County at the time of trial (approximately 465,000)[2] it would come to the conclusion that only 3.72 people in Pima County could make teeth marks similar to those found on the victim's body. If, however, the figure eight in one hundred thousand is the correct figure, there would be thirty-seven people in Pima County whose teeth could make similar marks. This example graphically illustrates just one of the reasons I am so concerned about allowing such unfounded probability figures into evidence a little error can paint a drastically distorted picture. The other reason is that the stunning impact of dramatic figures makes the picture indelible in the minds of the jurors, especially when the posture of Dr. Campbell's testimony left counsel unable to dispute the figure.
I would hold that when, as in this case, an expert offers a statement that is represented as being a direct quote from a book, the testimony is admissible only if the source material is present in the court room and available for inspection by opposing counsel.
Even if the statistical testimony qualifies for treatment as a hearsay exception, I nevertheless feel that there was an inadequate foundation laid for its admission in this case. Because of the overwhelming impact on a jury of statistical evidence that indicates an almost insurmountably conclusive probability of an accurate identification, and because of the ever present danger that the presumptions on which such calculations are based might be flawed, evidence of mathematical probabilities should be admitted only if there is a thorough foundation articulated by the expert supporting the legitimacy of the conclusion reached. See People v. Collins, 68 Cal. 2d 319, 66 Cal. Rptr. 497, 438 P.2d 33 (1968). McCormick on Evidence § 204 (2d Ed. 1972). To permit an expert to offer a probability figure without a complete explanation of how the number was calculated would not only intensify the mystery surrounding pronouncements of such huge probability figures but also, "foreclose(s) the possibility of an effective defense by an attorney apparently unschooled in mathematical refinements." People v. Collins, 66 Cal. Rptr. at 502, 438 P.2d at 38. In this case, the defense attorney's difficulty in attempting to defuse the impact of the large numbers was compounded by the fact that Dr. Campbell had not performed his own calculations and could not cite the source of the general formula. Because of his ignorance of not only the method by which the figure was calculated, but also the treatise where it could be verified or disputed, his testimony was essentially immune from the test of cross examination.
The science of comparing bite-marks begins with the assumption that any individual's bite-mark is unique. In attempting to identify a bite-mark by comparing it with the known mark of an accused, the forensic orthodontologist selects several distinct features or characteristics of a bite-mark such as the individual shape of the teeth (as *262 evidenced by the mark), the spacing between the teeth, and the size of the arch. The expert then conducts a methodical comparison of each aspect of the two bite-marks, determining whether there is similarity with respect to any or all of the features. As in fingerprint analysis, as the incidence of the finding of similarity among the "points of comparison" increases, so does the assurance that any overall identification is "positive". See Warren and Harvey Dental Identification and Forensic Odontology (1976).
While Dr. Campbell's testimony on the matter is of little assistance, it appears that the probability figure, "eight in one million" is reached by application of the so called "product rule" to the comparison technique described above. In simple form, the rule is that if two or more events are mutually independent, the probability of their co-occurrence is equal to the product of their individual probabilities. See McCormick on Evidence, § 204 (2nd Ed. 1972). In bite-mark analysis, the underlying events used as the basis for calculating a product are the individual probabilities that there will be similarity in two bite-marks with respect to any one of the isolated characteristics described above.
The California Supreme Court's primary objection to the testimony offered in People v. Collins, supra, was that there was no empirical evidence offered to justify the probability figure assigned to each "event" used to calculate the figure "one in twelve million". The difficulty with the testimony of this case is just as serious. There was no evidence offered to justify the statistical weight assigned to each identification factor. Moreover, the absence of an explanatory foundation disguised the fact that the eight in one million figure taken from the literature was completely misleading when applied to the comparison study actually performed in this case.
Assume, for ease of calculation, that the probability of finding similarity in two bite-marks with respect to each discrete aspect is one in four. Assuming the independence of each variable, the probability of finding similarity with respect to two features is one in sixteen; with respect to three features, one in sixty-four. It should be apparent that as the number of findings of similarity increases, the denominator of the probability fraction rises exponentially. Far more significant jumps occur, for instance, when the findings of similarity increase from seven to ten.
The problem with Dr. Campbell's having utilized the figure "eight in one million" in this case is that while he was able to testify that that figure represents the product of the probabilities that there will be similarity in ten major points of comparison, he also testified that because of "inconsistency," he did not employ all of the ten major factors in making his identification. It is clear that the discrepancy between the comparative assumptions underlying the figure announced by Dr. Campbell and the actual comparison study performed by him in this case render the eight in one million figure totally irrelevant. However, because of his unfamiliarity with the statistical weight assigned to each comparison factor, Dr. Campbell would have been completely unable to explain to the jury the degree to which his figure was exaggerated when applied to this pair of bite-marks.
As an illustration, assume that Dr. Campbell failed to consider three of the ten major factors in making his comparison. Assume further that the variables not considered are independent and that each has a statistical weight of one in four. It would seem then that considering only the remaining seven points of comparison, the probability of the bite-marks being identical would be reduced to below one in two thousand.[3]
When this failure of Dr. Campbell to perform his own calculations is compounded by the fact that the figure in the literature for a study based on all ten factors is eight in one hundred thousand, it is obvious that a proper application of the product rule *263 would have yielded a substantially less imposing figure than eight in one million.
In conclusion this case highlights the numerous pitfalls that can be encountered by permitting probability evidence without a demonstration of the validity of the figure announced by the expert. I would hold that such testimony is admissible only if the witness is thoroughly familiar with the relevant probability formula and can justify the statistical weight assigned to each variable in that formula. I would further require that the expert be able to fully explain to the court the calculations he performed in arriving at his ultimate conclusion. CAMERON, Chief Justice (concurring in dissent).
I concur in the dissent of Justice Gordon concerning the prosecutor's closing arguments to the jury.
NOTES
[1] Moreover, the applicability of even an eight in one hundred thousand figure to the defendant is dubious. Warren and Harvey are referring to the probability of a specific combination of distinctive tooth characteristics occurring in Scotland where "not more than sixty percent of the adult population over sixteen have some natural teeth". Dr. Campbell never suggests that any of these same specific characteristics were present in the bite-marks that he studied.
[2] This figure has been rounded off, to facilitate calculations, from the actual 1977 population of Pima County which was 468,100. Valley National Bank of Arizona, Arizona Statistical Review, (33rd ed., 1977) at 8.
[3] This figure is reached by calculating the product of the hypothetical probabilities of the excluded factors, and dividing the figure eight in one million by that product.