Summary
finding product-rule estimate using ceiling principle to be admissible after finding unmodified product-rule calculations to be in admissible
Summary of this case from State v. HarveyOpinion
Nos. 91-380, 93-326
Decided December 9, 1994
1. Evidence — Scientific Tests — Admissibility "Ceiling principle," requiring testing of DNA samples from about 100 persons from each of 15-20 randomly selected ethnic populations, for creation of conservative estimate of population frequencies in order to account for possibility that certain genetic markers may occur more or less often within certain racial groups, has gained general acceptance in relevant scientific community.
2. Appeal and Error — Standards of Review — Generally
3. Evidence — Scientific Tests — Admissibility Frye test for admissibility of scientific evidence requires (1) general acceptance in relevant scientific community of the scientific theory or principle; and (2) general acceptance in relevant scientific community of the techniques, experiments, or procedures applying that theory or principle.
4. Evidence — Scientific Tests — Admissibility
5. Evidence — Scientific Tests — Admissibility
6. Evidence — Scientific Tests — Admissibility Frye test for admissibility of scientific evidence does not require scientific unanimity.
7. Evidence — Scientific Tests — Admissibility
8. Evidence — Scientific Tests — Admissibility
9. Appeal and Error — Harmless Error — Burden of Proof
10. Evidence — Weight and Sufficiency — Criminal Cases
11. Appeal and Error — Harmless Error — Criminal Cases
12. Criminal Law — Prosecutorial Conduct — Discretion of Prosecutor
13. Appeal and Error — Reversal — Prosecutorial Misconduct
14. Appeal and Error — Reversal — Prosecutorial Misconduct
15. Criminal Law — Confrontation of Witnesses — Particular Cases
16. Arrest — Probable Cause — Generally
17. Arrest — Probable Cause — Tests and Standards
18. Arrest — Probable Cause — Particular Cases
19. Criminal Law — Confrontation of Witnesses — Cross-Examination
20. Criminal Law — Judgment and Sentence — Statement by Defendant
Jeffrey R. Howard, attorney general ( Janice K. Rundles, assistant attorney general, on the brief and orally), for the State.
James E. Duggan, chief appellate defender, of Concord, and Albert E. Scherr, assistant appellate defender, of Concord ( Mr. Duggan and Mr. Scherr on the brief, and Mr. Scherr orally), for the defendant.
The defendant, Daniel Vandebogart, appeals from his conviction for first degree murder, RSA 630:1-a (1986), based on a jury verdict in the Superior Court ( Mohl, J.). After consideration of all the issues raised by the defendant on appeal, we affirm.
On the afternoon of September 12, 1989, Kimberly Goss, the victim, was taken from her home, bound, gagged, raped, and strangled, and her dead body was left partially buried in the woods. The defendant was arrested the following morning. Evidence was presented at the defendant's six-week jury trial as to the following facts.
At approximately 1:30 p.m. on September 12, the victim left work intending to meet a friend, Susan Thibodeau, to go shopping. The victim and her husband lived on Old Derry Road in Londonderry in a duplex with a large, secluded back yard, behind which was a large tract of woods. At approximately 2:30 p.m., neighbor Carl Currier heard a woman's scream coming from the area behind the victim's residence. Between five and fifteen minutes later, he heard a car start and thought the sound came from the area of a dirt driveway located near the victim's home. He then saw a black Pontiac Firebird with a white male driver going slowly past his house, headed toward Derry. Shortly thereafter, Currier saw the Firebird parked down the street and wrote down the vehicle's license plate number. Carl Currier's thirteen-year-old brother, Toby, also saw the car parked as described by Carl when he arrived home from school, and then saw it pass their residence again, headed toward Manchester at about 3:25 p.m.
Unable to reach the victim at the time of their planned shopping excursion, Thibodeau went to the victim's home to look for her. She arrived at about 4:00 p.m. and found the victim's car in the driveway, but not the victim. The victim's husband, David Goss, arrived home at 4:30 p.m. to find his wife missing and Thibodeau looking for her. The victim's purse and keys were on the table, and her cigarettes were on the kitchen counter; a cigarette butt was on the floor, and an ashtray was in the trash can; the dishwasher was standing open and half-unloaded; the family dog was shut in a closet. After searching for several hours and calling friends and relatives, David Goss called the Londonderry Police Department at about 7:30 p.m. Several officers responded, including Detective Scott Saunders and his supervisor, Lieutenant Joseph Ryan. Currier, seeing the commotion at the victim's residence, walked over and gave the officers the license number for the vehicle he had seen. A motor vehicle record check revealed that the defendant was a co-owner of the car.
At the victim's home, the Londonderry police located the victim's blue jacket buried under some leaves just beyond the cleared portion of the back yard. They also found a pair of sneakers with the laces removed and a kitchen knife belonging to a butcher block set in the victim's kitchen. Casts and photographs were taken of tire impressions and footprints found at the scene.
On September 15, the victim's body was discovered partially buried in the woods behind her home. Her arms were behind her back, bound at the wrists with shoe laces and pantyhose in a series of complicated knots, and a white athletic sock was knotted tightly around her neck. Her lower body was unclothed. Her denim skirt, underwear, a broken shoe lace, and a jar of Vaseline which had been taken from the victim's bathroom were found near the body. The medical examiner concluded that the victim had been strangled with the athletic sock, after it had been used as a gag. There were ligature marks around the victim's ankles, and it appeared that her ankles and wrists had been bound together behind her back. There was an abrasion on her chin and indications that she had struggled. A tear to her anus which had hemorrhaged, a bruise on her thigh, and the presence of Vaseline in her pubic hair suggested that the victim had been sexually assaulted while still alive. Semen, which proved not to be from David Goss, was found in the victim's vagina.
The defendant was arrested at approximately 5:15 a.m. on September 13. His clothing was seized and placed in eight separate bags. Combings of the defendant's pubic hair, which was coated in petroleum jelly, a photograph of what appeared to be a fresh scratch on his leg, and a sample of the defendant's blood were all taken for evidentiary purposes.
The samples of Vaseline and semen taken from the victim, the tire and footwear impressions found at the scene, and the clothing, petroleum jelly, and blood taken from the defendant were sent to various laboratories for testing. An expert in tire and footwear comparisons and in chemical analysis testified that the marks found at the scene were consistent with both the defendant's tires and the tennis shoes he was wearing at the time of his arrest. He additionally testified that the petroleum jelly found on both the victim and the defendant was chemically indistinguishable from that in the container of Vaseline found at the scene.
Analysis of the defendant's clothing revealed the presence of vegetation in his jeans, socks, and boxer shorts. This vegetation included hemlock needles, white pine bud scales, and seed-like structures. Identical needles and bud scales were found in the vicinity of the victim's body. Identical seed-like structures were found in the cleared area of the victim's back yard and were identified as barnyard grass. An expert testified that the combination of the vegetation was very unusual because barnyard grass is rare and does not exist in the same habitat as the hemlock and white pine.
Testimony regarding the testing of the defendant's blood for a match to the semen took a little more than a day of the six-week trial. The samples of semen and known blood taken from the victim's body as well as the sample of blood taken from the defendant were sent to the FBI forensic laboratory for comparison. The tests performed showed that the defendant's DNA profile matched that of the semen, and that he could therefore not be excluded as a source. The FBI calculated the odds of a person selected at random from the Caucasian population matching the semen sample as one in 50,000.
The defendant had previously been convicted of two very similar sexual assaults, one in Virginia and the other in Montana. In each case, the defendant had subdued the victim by threatening the victim with a knife. He had taken each victim to a secluded, wooded area, bound their wrists behind their backs with boot laces, gagged them with white athletic socks, and raped them. On one occasion, he told the victim that "[t]ying knots turn[ed] him on."
The defendant's mother had spoken with the defendant on the telephone after his arrest. After asking the defendant about another assault charge then pending against him, she said, "I have another question for you," to which the defendant replied, "The answer to that question is yes. . . . I can't tell you why. I don't know. I just went too far this time." She understood the defendant's statements as a confession of guilt for the victim's murder. Later that day, she visited him at the prison and asked him more questions about the murder. She asked him how he felt when he "dragged her body out," and he replied, "Scared." She asked him if he had buried her, and he replied, "[P]artially." He admitted that it was a sex crime. Other testimony established that at the time she spoke with her son, the news media had not yet reported that the victim had been raped or that she had been partially buried. The defendant also told his mother that he had attempted to commit suicide the day before their conversation.
In State v. Vandebogart (DNA), 136 N.H. 365, 616 A.2d 483 (1992), we bifurcated the defendant's appeal to expedite our consideration of issues relating solely to the admissibility of forensic deoxyribonucleic acid (DNA) profiling. After clarifying the two-part Frye standard for admitting novel scientific evidence, id. at 376, 616 A.2d at 490; see Frye v. United States, 293 F. 1013 (D.C. Cir. 1923), as requested by the parties, we concluded that the trial court had misapplied Frye by admitting population frequency estimates that had not found general acceptance in the field of population genetics. Vandebogart (DNA), 136 N.H. at 381, 616 A.2d at 494. We remanded to the trial court to conduct a hearing in order to determine whether a technique known as the "ceiling principle" has gained general acceptance in the relevant scientific community and whether the erroneous admission of the population frequency statistic in this case was harmless error. Id. at 383, 616 A.2d at 495.
[1] We now hold that the ceiling principle, as described infra, has gained general acceptance in the relevant scientific community. We also conclude that the overwhelming conventional evidence presented at trial against the defendant renders the erroneous admission of population frequency evidence in this case harmless beyond a reasonable doubt.
The defendant also asserts several alternative grounds upon which the trial court allegedly erred: (1) in failing to grant a mistrial or to give curative instructions following a portion of the prosecutor's closing argument that the defendant argues denied him a fair trial by improperly evoking sympathy for the victim; (2) in ruling that one of the State's witnesses, a prior victim of the defendant, need not disclose her current name and address while testifying; (3) in finding that there was probable cause to arrest the defendant; (4) in allowing the State to call a rebuttal witness to contradict the defendant's testimony on collateral issues; and (5) in denying the defendant's request to address the court before sentencing. We shall address these issues seriatim following our analysis of the DNA-related issues.
I. DNA-Related Issues
In Vandebogart (DNA), we held that the restriction fragment length polymorphism (RFLP) process employed by the FBI to determine a match between a sample taken from the suspect and genetic material found at the crime scene is a generally accepted technique in the scientific community. Id. at 379, 616 A.2d at 492. We concluded, however, that the FBI's method for determining the significance of any such declared match, as expressed by the probability that there is a coincidental match, was not a generally accepted technique because it did not account for population substructure, that is, the possibility that certain genetic markers may occur more or less often within certain racial groups. Id. at 381, 616 A.2d at 494. We noted that the National Research Council (NRC), in a pre-publication copy of its April 1992 report entitled "DNA Technology in Forensic Science," recommended a technique known as the "ceiling principle" to create a conservative estimate of population frequencies in order to account for the possible existence of population substructure. Vandebogart (DNA), 136 N.H. at 382, 616 A.2d at 494.
Before we address the issue of whether the ceiling principle has gained general acceptance in the relevant scientific community, we first summarize the methodology underlying the principle. As with our first examination of DNA profiling, see id. at 368, 616 A.2d at 485, we derive this synopsis in part from the NRC's 1992 report; we also rely on evidence admitted at the trial court's hearing on remand [hereinafter "second Frye hearing"].
A. Allele Frequency Calculation
An allele is one of two or more alternate forms of a gene. Allele frequency is the proportion of a particular allele among the chromosomes carried by individuals in a population. For forensic purposes, there are two ways of calculating allele frequencies: the fixed bin method and the floating bin method. The presence of alleles is determined by measuring DNA fragments; these measurements determine the sample's bin, i.e., the window of measurements that are close enough to the sample's measurement to constitute a potential match. Under the fixed bin method, alleles are aggregated into a predetermined set of bins arranged according to molecular weight measurements. The frequency of a particular allele in the population is computed by determining the frequencies for the bins containing any alleles that fall within a statistical forensic matching window specified by the testing laboratory. The NRC recommends that when an allele falls within two or more bins, the frequencies of all of the bins containing that allele be added together; it is not sufficient to use only the allele frequency from the bin with the highest frequency of the particular allele.
Under the floating bin method, the testing laboratory counts the alleles that would be regarded as a match within the laboratory's statistical forensic matching window. Unlike the fixed bin method, however, the floating bin method does not aggregate alleles into predetermined bins. Instead, the allele frequency is derived from a bin centered on the allele of interest and surrounded on either side with a statistically appropriate sample size. In its report, the NRC recommends either the fixed bin or the floating bin method for implementing the ceiling principle; the NRC does not express a preference between the two methods. See NATIONAL RESEARCH COUNCIL, DNA TECHNOLOGY IN FORENSIC SCIENCE 85-86 (1992) [hereinafter NRC REPORT].
B. Ceiling Principle and Interim Ceiling Principle
The ceiling principle calls for a testing laboratory to analyze DNA samples from about 100 persons from each of 15-20 randomly selected ethnic populations and, using either the fixed bin or floating bin method, to calculate the allelic frequencies at a particular locus — that is, the specific physical location of a gene on a chromosome — for each population sampled. In other words, for a particular allele exhibited at a particular locus, the testing laboratory would calculate the percentage of individuals within each ethnic population exhibiting the same, or a statistically similar, allele at the same locus. Once these data are compiled, the ethnic population containing the largest percentage of individuals exhibiting the particular allele at the particular locus would be chosen to represent the potential "ceiling frequency" for that particular allele. If this potential "ceiling frequency" is less than a certain minimum (5% under the ideal ceiling principle and, as will be explained infra, 10% under the "interim" ceiling principle), the minimum would replace the computed frequency as the "ceiling frequency" to be used in subsequent calculations. The process of determining a ceiling frequency would then be repeated for each of the other alleles in the defendant's DNA profile. When all of the ceiling frequencies have been determined, they are multiplied in accordance with the "product rule," see Vandebogart (DNA), 136 N.H. at 372-73, 616 A.2d at 488, thereby providing a highly conservative (that is, generous) estimate of how common the defendant's DNA profile is within the relevant population.
Until the recommended population studies of 15-20 ethnic groups are completed, however, the NRC recommends that even more conservative adjustments be made to the ceiling principle and to the approach for presenting population frequency estimates. These adjustments, referred to collectively as the "interim" ceiling principle, do not require additional population studies, as the required frequencies can be derived from presently existing databases. Under the interim ceiling principle, the largest allelic frequency (expressed in terms of the upper 95% confidence limit) is gleaned from among all the ethnic databases. This frequency, or 10%, whichever is larger, is used to represent the ceiling frequency for that particular allele. As with the ideal ceiling principle, the product rule is then applied to the individual interim ceiling frequencies of each allele studied to derive an extremely conservative estimate of the probability of a random match. See NRC REPORT, supra at 84-85.
C. Second Frye Hearing
On remand, in February 1993, the trial court conducted its second Frye hearing in this case. The trial court heard testimony from five experts, four on behalf of the State and one on behalf of the defendant, on the question of whether the interim ceiling principle is generally accepted within the appropriate scientific community. The trial court found that there is "universal" consensus in the scientific community of geneticists and forensic DNA scientists that the interim ceiling principle properly accounts for the possibility of population substructure by providing a highly conservative estimate of the probability of a random match. In addition, the trial court found that although estimates provided by the interim ceiling principle may be so conservative as to be deemed not accurate, it is nonetheless generally accepted in the scientific community that any possible errors in such estimates favor the defendant.
[2] On appeal, we independently review the record and make our own determination of general acceptance without regard to the findings of the trial court. State v. Vandebogart (DNA), 136 N.H. at 376, 616 A.2d at 491.
The trial court found that the interim ceiling principle resolves any uncertainties in favor of the defendant by generating a conservative estimate of the probability of a random match, which compensates for the possibility of population substructure. Although the defendant's expert, Dr. Laurence Mueller, testified that he and several other scientists believed that the interim ceiling principle could result in errors that do not favor the defendant if underlying assumptions about the independence of populations prove to be incorrect, the other expert testimony and documentary evidence from the second Frye hearing resoundingly reflect the trial court's view. Based on a review of the record, we concur with the trial court's finding.
[3, 4] The Frye test for the admissibility of scientific evidence requires: "(1) general acceptance in the relevant scientific community of the scientific theory or principle; and (2) general acceptance in the relevant scientific community of the techniques, experiments, or procedures applying that theory or principle." Vandebogart (DNA), 136 N.H. at 376, 616 A.2d at 490. We held in Vandebogart (DNA) that the first prong of the Frye test has been satisfied; namely, that "the theory underlying DNA profiling is generally accepted in the relevant scientific community." Id. at 378, 616 A.2d at 492.
Regarding the second prong of the Frye test, we have already held that the RFLP laboratory procedures used to determine a match between samples taken from the suspect and samples taken from the crime scene have gained general acceptance in the scientific community. Id. at 379, 616 A.2d at 492. Our remand order in Vandebogart (DNA) thus focused exclusively on the population frequency calculation aspect of the second prong of the Frye test; namely, whether the relevant scientific community generally accepts a method for calculating statistical probabilities of a random match that "yields a conservative estimate resolving all uncertainties [in DNA profiling] in favor of the defendant." Id. at 382-83, 616 A.2d at 494-95.
The defendant argues that the second prong of the Frye test requires that the technique selected for applying DNA methodology must be based on sound scientific principles. According to the defendant, a conservative result, even if guaranteed in every case, is not a replacement for a reliable scientific method. The defendant cites Daubert v. Merrell Dow Pharmaceuticals, Inc., 113 S. Ct. 2786, 2795 (1993), for the proposition that "[t]he adjective 'scientific' implies a grounding in the methods and procedures of science." Based on the testimony of several experts at the second Frye hearing, the defendant argues that the interim ceiling principle lacks such grounding, as it relies on flawed, or at best unproven, science. We disagree.
The Daubert Court grounded its decision on the Federal Rules of Evidence, holding that the Rules, particularly Rule 702, had superseded Frye in setting the standard for the admissibility of scientific evidence in federal courts. Id. at 2794. In light of the fact that Daubert is explicitly limited to the conduct of federal trials, and because the parties originally asked us to decide admissibility issues under Frye, see Vandebogart (DNA), 136 N.H. at 373, 616 A.2d at 488, we do not decide whether the Daubert standard should supersede the Frye standard in New Hampshire trial courts. In this case, however, the result under either standard would be the same.
[5] Daubert does not stand for the proposition that scientific knowledge must be absolute or irrefutable.
of course, it would be unreasonable to conclude that the subject of scientific testimony must be "known" to a certainty; arguably, there are no certainties in science. . . . But, in order to qualify as "scientific knowledge," an inference or assertion must be derived by the scientific method. Proposed testimony must be supported by appropriate validation — i.e., "good grounds," based on what is known.
Daubert, 113 S. Ct. at 2795.
We acknowledge that the experts who testified at the second Frye hearing generally agreed that the interim ceiling principle was not based on sound genetic principles. Indeed, the ten percent ceiling frequency represents such a conservative estimate of allele frequency that it may not exist in nature. However, ample testimony in the record supports the view that it is a valid scientific technique to build conservative features into statistical results to allow for experimental or sampling error. Thus, as the interim ceiling principle builds conservative safeguards into allele frequency estimates in order to account for the possibility of population substructure, it is a valid scientific technique.
[6] In addition, the Frye test does not require scientific unanimity. See United States v. Porter, 618 A.2d 629, 634 (D.C. 1992); Commonwealth v. Lanigan, 596 N.E.2d 311, 316 (Mass. 1992). As a result, the isolated criticisms cited by the defendant do not defeat the existence of a general consensus in the scientific community that the interim ceiling principle provides conservative estimates of DNA-derived population frequency calculations.
[7] Therefore, in examining the interim ceiling principle, the trial court did not err in its application of the Frye test to the evidence before it. Based on our review of the record, the available literature, and other courts' consideration of the matter, we conclude that the interim ceiling principle is generally accepted in the relevant scientific community as providing a highly conservative estimate of the possibility of a random match that resolves all uncertainties in favor of the defendant. Accord United States v. Bridgett, Crim. No. F-8544-90, slip op. at 30-31 (D.C. Super. Ct. May 29, 1992), reprinted in 120 Daily Wash. L. Rep. 1697, 1704-05 (1992). Although we have not had the opportunity to decide whether the Daubert standard should supersede the Frye standard in New Hampshire trial courts, the outcome today would not be different under a Daubert-type, rules-based approach. If we were to read New Hampshire Rule of Evidence 702 under the Daubert approach, which gives trial courts more discretion than the Frye test, the DNA evidence would have been admissible. Our holding today thus obviates the need for future relitigation on this issue; trial courts may henceforth take judicial notice that the interim ceiling principle has attained general acceptance.
We are not unmindful of the fact that the interim ceiling principle will inevitably become obsolete, as more accurate techniques for estimating the probability of a random match gain general acceptance, and as the presence or absence of population substructure is determined to a sufficient degree of certainty. We merely hold today that the interim ceiling principle provides a conservative vehicle upon which to carry forensic DNA evidence across the threshold of admissibility.
D. Harmless Error Analysis [8] We have already held that it was error to admit population frequency estimates that had not found general acceptance in the field of population genetics. Vandebogart (DNA), 136 N.H. at 381, 616 A.2d at 494. Furthermore, we have held that "evidence of a match will not be admissible if it is not accompanied by a population frequency estimate that has been produced from a generally accepted method." Id. at 382, 616 A.2d at 494. It necessarily follows, therefore, that the trial court's admission of forensic DNA evidence at all in this case was error. See Commonwealth v. Daggett, 622 N.E.2d 272, 274-75 (Mass. 1993) (plurality opinion).
"In determining the gravity of an error, this court asks whether it can be said beyond a reasonable doubt that the inadmissible evidence did not affect the verdict." State v. Dellorfano, 128 N.H. 628, 637, 517 A.2d 1163, 1169 (1986) (quotation omitted). "The evaluation of whether this standard has been achieved involves consideration of the alternative evidence presented at trial," id., and of the character of the inadmissible evidence itself, see State v. Lemieux, 136 N.H. 329, 331, 615 A.2d 635, 636 (1992). An error may be harmless beyond a reasonable doubt if the alternative evidence of the defendant's guilt is of an overwhelming "nature, " State v. Dumais, 126 N.H. 532, 535, 493 A.2d 501, 503 (1985), "quantity," State v. Sampson, 132 N.H. 343, 348, 565 A.2d 1040, 1043-44 (1989), or "weight," State v. Green, 133 N.H. 249, 260, 575 A.2d 1308, 1315 (1990), and if the inadmissible evidence is merely cumulative, Lemieux, 136 N.H. at 331, 615 A.2d at 636, or inconsequential in relation to the strength of the State's evidence of guilt, id. at 332, 615 A.2d at 636.
[9-11] "The burden is on the State to prove harmless error, and this burden is met only if we can conclude beyond a reasonable doubt that the evidence did not affect the verdict." State v. Silk, 138 N.H. 290, 291, 639 A.2d 243, 244 (1994) (citation omitted). The State presented alternative evidence, the nature, quantity, and weight of which can only be described as overwhelming. The defendant knew details of the crime and confessed to his mother that he had committed the murder. The defendant committed two previous rapes in a manner consistent with various aspects of the assault committed against the victim, including the location of the crime and the means used to subdue and restrain the victim. The defendant's car was spotted, by license plate number and specific description, in the vicinity of the crime within minutes after a scream was heard emanating from behind the victim's residence. The tire treads of the defendant's car were consistent with tire tracks found in the area where witnesses had spotted the car. The tread of the defendant's sneakers was consistent with sneaker prints at the victim's home. Three different types of vegetation found in the defendant's clothing matched vegetation from two distinct habitats that existed in the victim's yard and in the adjacent woods. Finally, the defendant and the victim had chemically identical Vaseline in their pubic hair, which was also identical to Vaseline removed from the victim's home and left in the woods beside the victim's body. The defendant's evidence, which consisted primarily of his own testimony, and that of his neighbor who testified that she saw him walking toward his apartment building the afternoon of the murder, did not seriously undermine the State's non-DNA evidence of guilt. To be sure, DNA evidence possesses a unique character for proving identity that cannot be overlooked. The State, however, presented equally compelling evidence of the defendant's identity as the perpetrator; namely, his confession to his mother. Moreover, the DNA evidence was subjected to vigorous challenge by the defense through cross-examination of the prosecution's expert witnesses; it was not presented to the jury with an "aura of infallibility." State v. Bible, 858 P.2d 1152, 1191 (Ariz. 1993), cert. denied, 114 S.Ct. 1578 (1994). Even taken in the light most favorable to the defendant, the non-DNA evidence of the defendant's guilt leads us to conclude beyond a reasonable doubt that the admission of DNA evidence did not affect the verdict. See State v. Bible, 858 P.2d at 1191-92; People v. Barney, 10 Cal.Rptr.2d 731, 747-48 (Ct.App. 1992); Daggett, 622 N.E.2d at 275-76 (holding error "not prejudicial").
Because we hold that the erroneous admission of DNA evidence in this case was harmless, we do not address the trial court's application of the interim ceiling principle and its calculation of the probability of a random match. For the same reason, we do not review the trial court's determination that the floating bin methodology of computing allele frequencies is superior to the fixed bin methodology, but we again note that the NRC approves of both methods, expressing no preference for one methodology over the other. See NRC REPORT, supra at 85-86. Finally, in light of our harmless error determination, we do not address the defendant's argument pertaining to the admission into evidence of the laboratory technician's reports.
II. Non-DNA Issues A. Closing Argument
The defendant argues that the prosecutor's closing argument denied the defendant a fair trial. He objected to the prosecutor's "evoking the memory of a victim in this case, talking about the victim talking to the jury in this case through the evidence or however," and moved for a mistrial or for curative instructions to the jury. Both forms of relief were denied by the trial court. The prosecutor commented toward the end of his closing argument that the evidence was impersonal and scientific in nature, but that through this impersonal evidence "Kimberly Goss is going to have the last word on [the defendant], because she's actually come back into the courtroom through all of that impersonal" evidence. After being interrupted by the defendant's objection, the prosecutor continued:
The evidence that has been produced through Kimberly Goss' body has in fact pointed a finger at that man, ladies and gentlemen, just as definitively, just as certainly as she would if she were in court today.
. . . .
I want you to imagine Kimberly Goss taking the stand, swearing to tell the truth and being asked whether or not she sees the man in the courtroom who did what was done to her on September 12th, 1989. Ladies and gentlemen, you know to a certainty that she would point her finger at him and she would turn to you and she would say, ladies and gentlemen of the jury, you have the right man. That is the right man.
The defendant argued below, and continues to argue on appeal, that the evocation of the victim's memory was an improper and deliberate attempt by the prosecutor "to influence the emotions and passions of the jurors by leaving them with the last lingering thought of what the trial would have been like if Kimberly Goss were still alive and had testified." He additionally suggests that the prosecutor was "obviously choked up and very emotional" during this portion of the argument, a charge that both the prosecutor and the trial court itself disputed. The trial court determined that there was no prosecutorial misconduct during the closing argument.
[12, 13] "A prosecutor may draw reasonable inferences from the facts proven, and has great latitude in closing argument to both summarize and discuss the evidence presented to the jury and to urge the jury to draw inferences of guilt from the evidence." State v. Sylvia, 136 N.H. 428, 431, 616 A.2d 507, 509 (1992) (citation and quotation omitted). To constitute prosecutorial misconduct, the government must, either intentionally or through gross negligence, have caused "aggravated circumstances to develop which seriously prejudiced a defendant, causing him reasonably to conclude that continuation of the tainted proceeding would result in his conviction." Id. We do not agree with the defendant that the statements in this case warrant reversal.
[14] The prosecutor did not ask the jury to base its decision upon impermissible grounds of sympathy or justice for the victim, or upon evidence not in the record. Rather, he argued that the evidence before the jury, including the semen found inside of the victim's body, the Vaseline found on her person, the vegetation found inside of the defendant's undergarments, the shoelaces used as bindings, and the sock used to gag and strangle the victim, was tantamount to an identification of the defendant by the victim. We find the argument to be a reasonable inference based upon the evidence. See State v. Batchelder, 126 N.H. 700, 701, 496 A.2d 346, 348 (1985). B. Witness Anonymity
The defendant next contends that the trial court's ruling that the defense could not elicit the current name and address of a witness put on by the prosecution denied the defendant's right to confront and cross-examine the witness. We disagree.
Generally, a witness is required to divulge both name and address on cross-examination in order to prevent prejudice to the defendant ensuing from a "`denial of the opportunity to place the witness in his proper setting and put the weight of his testimony and his credibility to a test, without which the jury cannot fairly appraise them.'" State v. Novosel, 120 N.H. 176, 184, 412 A.2d 739, 745 (1980) (quoting Alford v. United States, 282 U.S. 687, 692 (1931)); see Smith v. Illinois, 390 U.S. 129 (1968). The defendant argues that the Smith and Alford decisions require reversal based upon the refusal of the trial court to require the witness to divulge her current name and address, regardless of any showing that this refusal prejudiced the defendant. We disagree with the defendant's characterization of these cases. As the State points out, other courts considering this issue have determined that "the purpose of Alford/Smith was to safeguard the opportunity for a meaningful and open cross-examination, not to require that a witness always divulge his or her home address." United States v. Alston, 460 F.2d 48, 51 (5th Cir.), cert. denied, 409 U.S. 871 (1972). "Smith v. Illinois does not establish a rigid rule of disclosure [of the witness's name and address], but rather discusses disclosure against a background of factors weighing conversely, such as . . . harassment, prejudice, confusion of the issues or the witness's safety." Clark v. Ricketts, 958 F.2d 851, 855 (9th Cir. 1991) (quotation omitted), cert. denied, 113 S. Ct. 117 (1992); accord United States v. Cavallaro, 553 F.2d 300, 304-05 (2d Cir. 1977).
In Novosel, we similarly held that the defendant is not entitled in all cases to the disclosure of a witness's current name and address. Novosel, 120 N.H. at 184-85, 412 A.2d at 745. As we noted there, the trial court may consider the witness's circumstances in considering whether to require this disclosure. Id. Here, the witness was a prior victim of the defendant who testified regarding the sexual assault he perpetrated upon her in 1978, for which assault he was convicted and imprisoned in Virginia. The nature of her testimony makes "unnecessary any explicit disclosure of the threat of harassment or harm, because the threat is inherent in the situation." Id. at 186, 412 A.2d at 745. We find that the potential danger to the witness outweighs the defendant's right to the disclosure of her current name and address, and therefore find no abuse of discretion in the trial court's refusal to require this disclosure.
C. Probable Cause
The defendant next argues that the trial court erred in finding that there was probable cause to arrest him, contending that the facts did not support a finding of probable cause and that the arresting officer was not a credible witness at the suppression hearing. The defendant was arrested outside of his apartment building at approximately 5:15 a.m. on September 13, 1989, ten hours after the victim was reported missing to the Londonderry Police Department. Lieutenant Ryan, who made the arrest, was the chief of the detective unit of the Londonderry police force and had been investigating the disappearance of the victim since she was reported missing. At the time of the arrest, he had the following information linking the defendant to the victim. A neighbor of the victim, Carl Currier, had heard a woman scream in the vicinity of the victim's residence at around two in the afternoon on September 12. Shortly after the scream, Currier saw a black sports car pass his house slowly, driven by a white male. Shortly thereafter, he saw what he believed was the same car parked on the wrong side of the street, just down the street from the victim's residence. He recorded the vehicle's license plate number and gave this number to Ryan. Currier also told Ryan that the car he had seen had a stuffed, toy animal in the windshield area. The Londonderry police determined that the car seen by Currier was registered to the defendant. Ryan and another officer also investigated the area where the car had been parked, and noticed tire and footprint impressions in the dirt.
Additionally, after determining that the car seen near the victim's residence was registered to the defendant, Ryan discovered that the defendant had a criminal record which showed a conviction for "deviant sexual conduct," for which the defendant had been designated a "dangerous offender" and had received a twenty-year sentence. He additionally learned that the Manchester police had recently arrested the defendant for assault and criminal threatening. The police file on that case indicated that the victim's description of the car that the defendant drove was consistent with Carl Currier's description, and that the conduct committed against the victim in that case had occurred near that victim's residence.
Ryan and another officer arrived in the area of the defendant's apartment sometime after midnight but could not find the defendant's car. They returned to this area sometime after 1:00 a.m. to await the possible return of the car. At approximately 5:15 a.m. the defendant arrived. Ryan observed that the defendant's car matched the description given by Currier. He also noticed that the car appeared dirty and saw what appeared to be small fingerprints on the fender. He saw what he believed to be a yellow blouse with a stain on it inside the car. He noted that the defendant's arms were dirty from the elbows down.
Additionally, the defendant seemed evasive when Ryan spoke to him. He told Ryan that he had been working all night, which Ryan disbelieved because the defendant was a construction worker and Ryan knew of no all-night construction jobs. The defendant refused, when requested, to accompany Ryan to the Londonderry police station to answer some questions. Ryan then asked the defendant to show him the bottom of his sneaker, the pattern of which Ryan believed to be consistent with the shoe prints he had observed earlier. He placed the defendant under arrest.
The defendant moved to have the evidence seized at the time of his arrest suppressed, arguing that the police did not have probable cause to arrest him. A hearing was held at which Ryan testified to the above information. The trial court found that there was probable cause to arrest. At trial, testimony from other witnesses showed that no yellow blouse was found in the car, and that there were fingerprints on the car, but in a different location from that testified to by Ryan at the suppression hearing. There was also testimony that it cannot be determined, by size, whether fingerprints belong to a male or female. Based upon these differences, the defendant moved during trial for the trial court to reconsider the suppression ruling, arguing that the testimony had shown Ryan's testimony about these items not to be credible. The trial court disagreed, finding Ryan's testimony credible and denying the motion to reconsider.
On appeal, the defendant argues that the trial court erred in denying both the motions to suppress and to reconsider, contending that there were insufficient facts to constitute probable cause and that the trial court erred in relying upon Ryan's testimony at the suppression hearing. We disagree.
[16] Probable cause to arrest exists when the arresting officer has sufficient, trustworthy information to warrant a reasonable person to believe that the arrestee has committed a crime. State v. Stevens, 121 N.H. 287, 290, 428 A.2d 1241, 1243 (1981). Courts are not bound by mathematical calculations in making this determination, but rather must "approach the issue with a concern for the factual and practical considerations of everyday life on which reasonable and prudent men, not legal technicians, act." Id. (quotation omitted). We will not overturn a trial court's determination of probable cause unless, when the evidence is viewed in the light most favorable to the State, the decision is contrary to the manifest weight of the evidence. State v. Reid, 135 N.H. 376, 379-80, 605 A.2d 1050, 1052 (1992).
The defendant would have us view each piece of evidence individually to determine that probable cause was absent at the time of the arrest. To this end, he cites cases holding that the presence of a similar vehicle in the vicinity of the crime, or the defendant's prior record, or a seemingly similar shoe print would not separately constitute probable cause. Compare Beck v. Ohio, 379 U.S. 89, 96-97 (1964) (defendant's criminal record and physical description, standing alone, cannot support probable cause); State v. Reynolds, 122 N.H. 1161, 1164, 453 A.2d 1319, 1321 (1982) (same brand of tennis shoes alone not enough to support probable cause) with State v. Lemire, 121 N.H. 1, 5, 424 A.2d 1135, 1138 (1981) (similar vehicle, combined with occupant's possession of gold jewelry and officer's knowledge of nearby jewelry robbery, sufficient for probable cause). The defendant argues, additionally, that Ryan's testimony regarding what he believed to be a stained yellow blouse and the fingerprints which he believed to be those of a female did not add anything to the determination of probable cause because there was no evidence that the victim wore a yellow blouse or had been near the car.
[17, 18] Contrary to the defendant's suggestion, we do not view each element of the evidence separately, but rather as a whole, and from the arresting officer's point of view at the time the arrest was made. Ryan knew that the victim had been missing for more than ten hours, and he suspected foul play. The victim's jacket had been found buried under leaves in the woods behind her home, and a knife and tennis shoes absent their laces had been found beneath a stump. He feared that the victim may have been buried somewhere. Given this fear, the defendant's dirty arms and car would reasonably seem significant. Additionally, Ryan did not know that the victim was not wearing a yellow blouse, and he might reasonably have found any stained, apparently female garment suspicious. There was also an accurate description of the defendant's car and license plate number tying him to the area of the Gosses' residence at the time the woman's scream had been heard, the defendant's prior conviction for a sex-related crime, and the similar tennis shoe pattern; in sum, the evidence combined to give the police reasonable cause to believe that the defendant was involved in the victim's disappearance. We hold that the trial court did not err in finding that, based upon all of the evidence in Ryan's possession at the time of the arrest, he had probable cause to arrest the defendant in connection with the disappearance of the victim.
Furthermore, while the defendant has the right to request the trial court to reassess a ruling on a motion to suppress based upon evidence produced at the trial, see Rouse v. United States, 359 F.2d 1014 (D.C. Cir. 1966); cf. United States v. Bradley, 455 F.2d 1181, 1185-86 (1st Cir. 1972), aff'd, 410 U.S. 605 (1973), we hold that the trial court did not err in denying the motion to reconsider. We find little to support the defendant's claim of inconsistencies between Ryan's testimony at the suppression hearing and the evidence at trial. Ryan testified at the suppression hearing that what he had believed to be a stained yellow blouse turned out instead to be a stained sleeveless yellow t-shirt. Thus, no inconsistency is shown between Ryan's testimony on this issue at the suppression hearing and the testimony at trial. The difference in the positioning and type of the fingerprints Ryan thought he saw on the defendant's vehicle and those actually found on the vehicle after it was impounded is not sufficient to cause us to hold that the trial court erred in finding Ryan's testimony credible.
D. Rebuttal Testimony
The defendant next contends that the trial court erred in allowing the State to call a rebuttal witness to contradict the defendant's testimony on collateral issues. The defendant complains about three statements in particular, all made by one witness. This witness, on rebuttal, contradicted the defendant's testimony in several respects, three of which the defendant complains were collateral in nature and thus not subject to impeachment by third-party testimony: (1) she testified that she got to know the defendant when they were playing "cat and mouse" on the highway, which the defendant had denied, stating that he had known the witness since grade school; (2) she testified that the defendant told her that he was going to break up with his fiancee, Laura Rumphrey, and complained about his sex life with her; and (3) she testified that she had seen the defendant one or two more times after August 13, the last time being September 8, 1989, when she told him not to "come around anymore," which contradicted the defendant's testimony that he last saw the witness on August 13. The defendant objected that this testimony was inadmissible as collateral, and argues now that the trial court's decision to admit this testimony violated New Hampshire Rule of Evidence 608 (b), which provides:
Specific instances of the conduct of a witness, for the purpose of attacking or supporting the witness' credibility, other than conviction of crime as provided in Rule § 609, may not be proved by extrinsic evidence. They may, however, in the discretion of the court, if probative of truthfulness or untruthfulness, be inquired into on cross examination of the witness (1) concerning the witness' character for truthfulness or untruthfulness, or (2) concerning the character for truthfulness or untruthfulness of another witness as to which character the witness being cross-examined has testified. The giving of testimony, whether by an accused or by any other witness, does not operate as a waiver of the privilege against self-incrimination when examined with respect to matters which relate only to credibility.
The defendant argues that this case is governed by State v. Brooks, 126 N.H. 618, 495 A.2d 1258 (1985), and does not fall within the exception to Rule 608 discussed in the more recent State v. Mello, 137 N.H. 597, 631 A.2d 146 (1993). In Brooks, an arson case, an eyewitness testified that he "recognized the defendant as the person he saw fleeing the scene of the fire." Brooks, 126 N.H. at 622, 495 A.2d at 1261. He admitted on cross-examination, however, that he had stated to a police officer that he could only remember "that the subject looked to be between 5' 6" and 6' tall, had a dungaree jacket on and possibly a beard." Id. The trial court refused to allow the defendant to then bring in the officer to testify to the statement given her by the witness. Id. This court upheld the trial court's refusal to allow rebuttal on this ground, ruling that where the witness had admitted to the prior inconsistent statement on cross-examination, "[t]he additional testimony concerning the prior statements . . . was merely cumulative, and the trial court's refusal to admit additional extrinsic evidence of those statements for impeachment was a proper exercise of judicial discretion." Id. at 623, 495 A.2d at 1262. Brooks, with its widely different fact pattern, is not relevant to a determination of the case before us.
In Mello, the defendant testified on direct that he had never been in trouble with the police before. On cross-examination, the defendant stated that he had been arrested, but only for drinking, and never for kidnapping or an offense involving sexual assault. The prosecutor brought in a police officer in rebuttal, who testified that the defendant had indeed been arrested in 1983 for rape and kidnapping. Mello, 137 N.H. at 598-99, 631 A.2d at 147. We rejected the defendant's claim that extrinsic, rebuttal evidence of a prior arrest could not be admitted because the evidence "was used not to impeach him generally but to prove that, when he testified that he had never been arrested and then that he had only been taken in for drunkenness," he gave knowingly false testimony under oath. Id. at 600, 631 A.2d at 148. In so doing, we clarified our interpretation of Rule 608, holding that "Rule 608, like its counterpart, Rule 609 relative to convictions, is directed at the use of [extrinsic] evidence to impeach a witness's general credibility." Mello, 137 N.H. at 600, 631 A.2d at 148.
[19] The defendant would have us limit Mello's applicability to those situations where the defendant has testified falsely on direct, arguing that to apply the rule in Mello here, where the defendant's testimony on cross-examination was at issue, would completely abrogate the intent of Rule 608. We disagree. "The object of a trial is not solely to surround an accused with legal safeguards but also to discover the truth." State v. Duke, 100 N.H. 292, 293, 123 A.2d 745, 746 (1956). Because the defendant elected to testify, his credibility could be impeached in the same manner as any other witness. See, e.g, State v. Lavallee, 119 N.H. 207, 211, 400 A.2d 480, 483 (1979). The State is permitted proper and fair cross-examination in an effort to uncover the defendant's false statements and elicit the truth. See, e.g, United States v. Havens, 446 U.S. 620, 626-27 (1980). In this case, we see no reason to distinguish the defendant's statements on direct examination from his responses to questions asked on cross-examination. Cf. id. The rebuttal testimony went to show that the defendant had testified falsely on cross-examination. This rebuttal did not attack the defendant's general credibility, but rather the truthfulness of his testimony. The trial court did not abuse its discretion in allowing this rebuttal testimony.
E. Right of Allocution
The defendant lastly contends that the trial court denied him his "constitutional right to be heard" at sentencing by refusing to allow him to address the court. After the verdict, defense counsel informed the trial court that the defendant wished to be heard briefly before sentence was imposed. The trial court denied this request. The defendant argues that the denial of this right warrants resentencing. We disagree.
The defendant makes passing reference to the State Constitution in his brief; he made no mention of the State Constitution, however, in raising his objection to the trial court's ruling below. He has thus waived any argument on appeal based upon the State Constitution, and we will review the trial court's decision only for a Federal constitutional violation. See State v. Turgeon, 137 N.H. 544, 546, 630 A.2d 276, 277 (1993).
[20] "[T]he right of allocution is not constitutional." United States v. Fleming, 849 F.2d 568, 569 (11th Cir. 1988). In an analogous situation, the United States Supreme Court held:
The failure of a trial court to ask a defendant represented by an attorney whether he has anything to say before sentence is imposed . . . is an error which is neither jurisdictional nor constitutional. It is not a fundamental defect which inherently results in a complete miscarriage of justice, nor an omission inconsistent with the rudimentary demands of fair procedure.Hill v. United States, 368 U.S. 424, 428 (1962). Additionally, because the sentence imposed upon the defendant was a mandatory life sentence without the possibility of parole, see RSA 630:1-a, III, we find that the trial court did not abuse its discretion in refusing to allow the defendant to speak to the court before it imposed the only possible sentence.
Accordingly, we affirm the defendant's conviction.
Affirmed.
All concurred.