Statev.Sivri

Supreme Court of ConnecticutAug 23, 1994
231 Conn. 115 (Conn. 1994)
231 Conn. 115646 A.2d 169

(14561)

Convicted of the crime of murder in connection with the disappearance of a woman who had gone to his residence to give him a massage, the defendant appealed to this court. Held: 1. The circumstantial evidence presented and the reasonable inferences that could be drawn from it were sufficient to establish beyond a reasonable doubt that the defendant had the specific intent to kill the victim notwithstanding the defendant's claim that the evidence could have supported a finding of recklessness or criminal negligence; the defendant's hypothesis of a different state of mind was only possible and not reasonable.

(Two justices dissenting)

2. The trial court improperly refused to instruct the jury regarding lesser included offenses; the evidence suggested at least the possibility that the defendant had acted with a lesser intent than that of the specific intent to kill and the proof on that issue was sufficiently in dispute to permit the jury to find the defendant not guilty of the greater offense but guilty of the lesser. 3. The affidavit supporting a warrant to search the defendant's residence, which resulted in the seizure of evidence he had sought, unavailingly, to have suppressed, established probable cause that a crime had been committed and that evidence of it would be found in his residence; the defendant admitted that the victim had been at his house, he was the last person to have seen her, and it was not unreasonable for the issuing magistrate to infer that evidence of the victim's whereabouts, or evidence that she had been physically assaulted, would be found there several days after her disappearance. 4. The trial court properly admitted evidence seized from the defendant's car on the ground that the car had been abandoned when it was searched; the defendant, having fled the country and having removed the car's license plates before leaving it on a public street miles from his home, could not claim a reasonable expectation of privacy in its contents. On the defendant's claim that the trial court improperly admitted into evidence expert testimony regarding population frequency calculations employed in the analysis of DNA typing evidence, this court directed the court on retrial to consider the conclusions and recommendations of the report of a committee assembled by the National Academy of Sciences to research and analyze the reliability of DNA testing to determine whether the probability calculations sought to be introduced into evidence conform to the committee's criteria or, if not, whether the evidence nevertheless passes appropriate scientific evidence standards.

Argued May 10, 1994

Decision released August 23, 1994

Substitute information charging the defendant with the crime of murder, brought to the Superior Court in the judicial district of Fairfield and tried to the jury before Gormley, J.; verdict and judgment of guilty, from which the defendant appealed to this court. Reversed; new trial.

John R. Gulash, Jr., with whom were Margaret M. Wynne and, on the brief, Charles W. Fleischmann, for the appellant (defendant).

Frederick W. Fawcett, assistant state's attorney, with whom, on the brief, were Donald A. Browne, state's attorney, and Jonathan C. Benedict, assistant state's attorney, for the appellee (state).


The defendant, Tevfik Sivri, appeals from the judgment of conviction, following a jury trial, of murder in violation of General Statutes § 53a-54a. The defendant claims that the trial court improperly: (1) determined that the evidence was sufficient to establish beyond a reasonable doubt that he intended to cause the death of the victim; (2) declined to instruct the jury on lesser included offenses; (3) denied his motion to suppress evidence seized from his home pursuant to a search warrant; (4) denied his motion to suppress evidence obtained as a result of a warrantless search of his automobile; and (5) admitted into evidence expert testimony regarding population frequency calculations employed in the analysis of DNA typing evidence. We agree with the defendant's claim regarding the necessity for instructions on the lesser included offenses. Accordingly, we reverse the judgment and remand the case for a new trial.

The defendant appealed directly to this court pursuant to General Statutes § 51-199 (b).

General Statutes § 53a-54a provides in relevant part: "MURDER. (a) A person is guilty of murder when, with intent to cause the death of another person, he causes the death of such person or of a third person. . . ."

The defendant also claims that the trial court improperly instructed the jury on the issue of intent and the definition of reasonable doubt. Because we reverse the defendant's conviction and order a new trial on the basis of the trial court's failure to instruct the jury regarding lesser included offenses, we decline to consider these claims, since they may not arise in the new trial.

The state produced the following evidence. At 7 a.m. on April 18, 1988, the victim, Carla Almeida, drove her boyfriend, Gerard Patano, to his job in her Volkswagen van. He expected her to pick him up by 5 p.m. that afternoon. The victim, who was twenty-one years old, had been living with Patano in Meriden for more than one year. That morning she also drove their son, who was less than one year old, to the home of Eunice McGuire, her aunt, for day care. McGuire expected the victim to pick up her child at approximately 5 p.m. that day. McGuire had been taking care of the child five days a week for approximately ten weeks, since the victim had started a new job. The victim picked up the child every day between approximately 5 and 5:30 p.m. The victim did not pick up Patano from work or their son from McGuire's home on April 18, 1988, however, and neither Patano nor McGuire has seen her again.

The defendant did not testify or produce any other evidence.

The victim worked five days a week for Andre's Massage (Andre's), which promoted "full body massage" services. Andre's did not maintain a massage parlor, but instead provided for house calls and retained a hotel room at the Hostways Motel in East Haven that was available for business twenty-four hours a day. An employee of Andre's, Paula Doak, received telephone orders for massage services at the home of one of the owners of Andre's, Gabriel Gladstone, in Hadlyme. According to Doak, the service offered was a choice of a one hour or a two hour full body massage. The fees for the employee's services were generally divided equally between Andre's and the employee. Doak denied any knowledge that any services other than massage services were being provided by employees of Andre's. She acknowledged, however, that Gladstone had previously been arrested for promoting prostitution.

The defendant lived with his mother in a house in northern Trumbull near the Monroe border at 37 Kitcher Court (Kitcher Court). At the time of the victim's disappearance, the defendant had been employed by a painting contractor, Patrick Burdo, for approximately eighteen months. On April 18, 1988, he went to work on an exterior job in New Haven in the morning, but because it began to rain at approximately 10 a.m., the defendant was sent home. At 2 p.m., he telephoned Andre's. Doak answered the telephone and wrote down the defendant's name, age, telephone number, and the fact that he was a painter who had been sent home from work. The defendant requested a one hour massage. Doak told him, as she routinely had told other customers, that he would be required to show the masseuse photographic identification, and that the masseuse would not take off her clothes or go to bed with him. The defendant did not request any particular masseuse.

After servicing her second customer of the day at the motel in East Haven, the victim telephoned Andre's and Doak told her that the defendant would be her next customer for a one hour appointment. Doak gave the victim the defendant's telephone number, and told her to telephone the defendant and obtain directions. Doak also believed that the victim was supposed to meet Gladstone at some point and give Gladstone his share of the receipts for the previous two days. The victim telephoned Doak and stated that she was going to the defendant's house. At 3:40 p.m., the victim telephoned Doak from Kitcher Court, and told Doak that she had checked the defendant's identification and verified his identity. A newspaper carrier noticed the victim's Volkswagen van in the driveway of the house at Kitcher Court between 3:30 and 4 p.m. while he was riding his bicycle past the house.

Andre's had a strict policy requiring each employee to call in at the beginning and conclusion of each appointment, and the victim had always adhered to this policy. When the victim had not called in by 4:40 p.m., Doak waited ten minutes, and then called the defendant's house. There was no answer. She called twice more without success, and then called the Trumbull police, told them of the situation and asked them to drive by Kitcher Court to see if the victim was still there. At 5 p.m., the newspaper carrier delivered a newspaper to the defendant's residence and noticed that the victim's van was no longer in the driveway.

At 6 p.m., Gladstone's wife, Rosanna Silva, arrived home and Doak expressed her concern about the victim to her. Silva called the defendant's residence repeatedly and finally got a busy signal, which she had the operator interrupt. A male voice then answered the telephone, and Silva asked if the victim was there, and if not, where was the victim. She repeated her questions, but received no response.

In response to Doak's telephone call, Lieutenant Henry DiJulio and Officer James Arlio of the Trumbull police department were dispatched to the defendant's residence. DiJulio arrived at Kitcher Court between 5:45 and 6 p.m. There was a young girl at the front door of the residence and, in the driveway, an automobile occupied by two high school age children. DiJulio asked the girl if anyone was home and she said no. The defendant's automobile, a 1987 Mitsubishi Tredia, was not in the driveway. DiJulio subsequently patrolled the area looking unsuccessfully for the victim's van.

Bridgeport police officer Paul Wargo, however, had noticed the victim's van parked on Waterview Avenue just north of the I-95 overpass while he was on routine patrol between 4:30 and 5 p.m. This location is approximately eight to ten miles from the defendant's home. There was nobody in the van, and its doors were unlocked. Wargo was unaware of the ongoing police investigation in Trumbull. When Wargo drove by approximately one-half hour later, the van was still there, and he had it towed to prevent it from being stolen.

Arlio patrolled the defendant's neighborhood in an attempt to discover any information regarding the victim's disappearance. At 6:15 p.m., Arlio met DiJulio at the defendant's residence, and Arlio observed that the defendant's automobile was still not in the driveway. At that time, DiJulio spoke with the defendant's mother and one of her sons, and explained the investigation, but did not enter the house.

Arlio continued to patrol the neighborhood. At approximately 7:30 p.m., while Arlio was across the street from the defendant's residence, he observed the defendant drive his automobile into his driveway. Arlio approached the defendant, ascertaining his identity by addressing him by name, and asked the defendant if he had been with a woman earlier in the day. The defendant answered yes. Arlio told the defendant that DiJulio wanted to speak with him. The defendant then entered the house and telephoned DiJulio, who asked him to come to the police station. The defendant exited the house, in the same clothing he had worn into the house, and told Arlio that he was going to the station to talk with DiJulio. At the station, the defendant told DiJulio that the victim had been at his house, that he had paid her $88, and that she had left at approximately 3:30 p.m.

On the day of the victim's disappearance, Refik Sivri, the defendant's brother, was working at a restaurant. He abruptly left work several hours early, at 7 p.m., after receiving a telephone call from a person whom his employer described as having a male voice.

The Trumbull police did not discover the victim's van until April 20, two days after her disappearance, when they became aware that it was in the custody of the Bridgeport police. At this point, according to Captain Frederick Nacovitch of the Trumbull police department, the police believed that there was probable cause to search the defendant's house, and they subsequently obtained a search warrant to do so.

The Trumbull police and the state police, with the assistance of Henry Lee, the director of the state forensic laboratory, began to search the defendant's residence on Thursday morning, April 21, continuing until approximately 3 a.m. the next morning. Lee and state police Detective James Craig testified regarding a number of blood stains and blood spatters that were found in the house, many of which were identified through chemical analysis as being human blood of the victim's blood type. There was a tissue with a blood-like stain in a garbage can in the garage. There were multiple blood droplets in the doorway that serves as an entrance to a hallway leading from the garage to the lower level family room of the house. On the bottom portion of the south wall of the hallway there were blood droplets, drop and contact blood smears, wiping patterns and medium velocity blood spatters. In Lee's opinion, human blood had been transferred to the wall, and an attempt to clean the blood had been made. Off the hallway there was a spare room. There were blood smears on the door to the room and on a small area of the wall adjacent to the light switch just inside the room.

Craig testified that, with respect to crime scene investigations, blood "spatter" refers to any quantity of blood, as little as a single drop, that subdivides into droplets as a result of coming into contact with a surface.

In the family room, there was a small blood stain on a pillow that was on top of a couch against the west wall. Lee also detected medium velocity blood spatters on a wooden paneled wall on the east side of the room, under a radiator cover, on a framed photograph and videocassette recorder placed on top of a television set, and on the television console and screen. Further, Lee discovered a discoloration and dampness in an area of the carpet between the television set and the wall divider. The carpet was cut up to expose its underside. The bottom of the carpet, and the padding underneath, had an eleven by fifteen inch direct transfer stain containing fresh blood. The nature of this stain indicated to Lee that a large amount of blood had saturated the surface of the carpet, soaked through the backing of the carpet and was transferred to the padding underneath. The blood was also dispersed through the application of some other liquid, which had diluted the blood and contained soaps. The blood could not be identified by type but could only be chemically identified as constituting human blood. Lee determined that as a "very conservative estimate" it would take 500 to 1000 cubic centimeters of blood to produce the stain in the rug. One thousand cubic centimeters would constitute one fourth to one fifth of the blood in the body of a woman of medium build.

On the room divider in the family room, there were two hundred tiny blood spatters, and wiping patterns. The majority of the stains in that area were medium velocity blood spatters. One of the victim's earrings also was found on the rug near the divider.

In Lee's expert opinion, the majority of the blood spatters on the room divider, the paneling of the east wall and the baseboard heater, the television set, and the items on top of the television set, had been produced by a mechanical force projecting the blood in small, randomly distributed patterns. Lee performed experiments with a rug shampooing machine that produced similar patterns, and opined that the stains had been produced by the cleansing action of a mechanical rug shampooer on a bloody carpet. He offered his opinion on what had occurred in the family room: the evidence was "consistent with an individual receiv[ing] a serious injury, of that much volume deposited in front of the T.V. . . . that spot, caus[ed] quite a bit of blood transfer. A person lost some amount of blood. Exact amount, I don't know, some large quantity of blood soak[ed] through the carpet. Subsequently, some mechanical device was used with some liquid for cleanup. Shampoo the rug maybe, and caused those medium velocity spatters to deposit on the lower portion of the wall. Subsequently, some attempt to clean up those spatters, caused some wiping/swiping pattern in between those medium velocity spatters."

The state asked Lee to offer his opinion on what had caused the injury. Lee responded: "I have no idea. I cannot come here to tell you. I only can say lost sufficient amount of blood; as far as knife, bullet, I cannot tell." He testified that "unless you cut deep enough or cut into the vessel, you will not come up with quarts of blood." Lee did not testify to any evidence of a violent struggle.

The state also compared hair obtained from a hairbrush and necklace belonging to the victim to hairs found in the house. Expert testimony identified two telogen head hairs found on the carpet in the family room as similar to the victim's hair. Other hairs found were dissimilar.

Elaine Pagliaro of the state forensic laboratory testified that telogen hairs are hairs that would normally fall from the head because the root is no longer actively growing.

On April 25, 1988, one week after the victim's disappearance, the defendant's automobile was found by police parked on Noble Avenue in Bridgeport. There were no license plates on the car. The police identified the vehicle by its vehicle identification number. The police had the vehicle transported to the Trumbull police department garage and subsequently searched the vehicle.

In the automobile, stains containing human blood were found on the steering wheel, the floor of the trunk, and on a jack handle taken from the trunk. There were also dripping and smearing blood stains found on the front and back of a rear interior plastic panel in the trunk. Several stains appeared to be diluted. In addition, blood-like stains were observed on the car's rear bumper. Several of the blood stains were identified as being of the victim's type, and DNA profiling tests of a sample of blood found within the trunk by Cellmark Diagnostics indicated to a high degree of probability that the blood belonged to the victim. Inside the trunk, there was a bag containing an unopened package of two sponges, and a bottle of Pine Sol. There was also an air freshener hanging in the trunk. The Mitsubishi model owned by the defendant was sold with a floor matting or carpet in the trunk of the car. There was no carpet in the defendant's Mitsubishi.

Finally, there was an elongated three-quarter inch by five-eighth inch hole in the trunk, stained with blood of the victim's type. There were blue paint chips in the hole, similar to the blue paint on the car jack, which was missing paint. There also was red paint on the jack similar to the red paint of the trunk. In Lee's opinion, some unknown force initially created the hole, and then the hole was widened with the jack. Various chemical and other tests could not ascertain what initially caused the hole before it was widened.

In an attempt to locate the body of the victim, Trumbull police searched a ten mile radius around the defendant's residence, encompassing parts of the towns of Trumbull, Monroe, Easton, Fairfield and Newtown. Neither a body nor any evidence of a body was found.

The state also presented the following evidence regarding the defendant's consciousness of guilt. Soon after the victim's disappearance, the defendant telephoned Patrick Burdo, his employer, and told Burdo that he had experienced some personal problems, and would not be able to work for a couple of days. The defendant did not, however, return to work at any point, except to pick up a paycheck. Burdo did not see the defendant again until the trial.

Three days after the victim's disappearance, on Thursday, April 21, Refik Sivri borrowed his employer's car. The next day, the defendant extended his expired Turkish passport at the Turkish Embassy in Ottawa, Canada, and purchased, at the Ottawa International Airport, an airplane ticket to travel from Ottawa to Istanbul, Turkey. Refik did not return the car to his employer until early Saturday morning, April 23, and gave no explanation of his need for the car.

The defendant returned to New York on June 30. He remained in New York, working in a restaurant, until his arrest by the Federal Bureau of Investigation on February 26, 1989. The defendant had three false identification cards in his wallet, two exhibiting the name Steve Fontaine, and one exhibiting the name Abdullah Kuyaalb.

I

The defendant first claims that the evidence was insufficient to prove beyond a reasonable doubt that he had the specific intent to cause the death of the victim. We disagree.

"In reviewing the sufficiency [of the evidence] claim, we apply a two-part test. First, we construe the evidence in the light most favorable to sustaining the verdict. Second, we determine whether upon the facts so construed and the inferences reasonably drawn therefrom the jury reasonably could have concluded that the cumulative force of the evidence established guilt beyond a reasonable doubt." (Internal quotation marks omitted.) State v. Greenfield, 228 Conn. 62, 76, 634 A.2d 879 (1993). The specific intent to kill is an essential element of the crime of murder. "To act intentionally, the defendant must have had the conscious objective to cause the death of the victim. General Statutes § 53a-3 (11). . . ." (Internal quotation marks omitted.) State v. Raguseo, 225 Conn. 114, 120, 622 A.2d 519 (1993). Intent is generally proven by circumstantial evidence "because direct evidence of the accused's state of mind is rarely available." State v. Greenfield, supra, 77. Therefore, intent is often inferred from conduct; id., 76; and from the cumulative effect of the circumstantial evidence and the rational inferences drawn therefrom. State v. Raguseo, supra, 119. This does not "require that each subordinate conclusion established by or inferred from evidence, or even from other inferences, be proved beyond a reasonable doubt" because this court has held "that a jury's factual inferences that support a guilty verdict need only be reasonable." State v. Crafts, 226 Conn. 237, 244, 627 A.2d 877 (1993). Nevertheless, because intent to cause the death of a person is an element of the crime; State v. Raguseo, supra, 120; that intent must be proven beyond a reasonable doubt. Patterson v. New York, 432 U.S. 197, 204, 97 S.Ct. 2319, 53 L.Ed.2d 281 (1977); In re Winship, 397 U.S. 358, 364, 90 S.Ct. 1068, 25 L.Ed.2d 368 (1970). Furthermore, "[i]ntent to cause death may be inferred from the type of weapon used, the manner in which it was used, the type of wound inflicted and the events leading to and immediately following the death." (Internal quotation marks omitted.) State v. Raguseo, supra, 120.

With this background in mind, we begin our analysis by acknowledging the absence in this case of certain types of evidence that we have identified in other cases as supporting an inference of an intent to kill. Thus, there was in this case no body or evidence of body parts; compare State v. Crafts, supra, 226 Conn. 237; State v. Nicely, 39 Ohio St.3d 147, 529 N.E.2d 1236 (1988); State v. James, 819 P.2d 781 (Utah 1991); no evidence of the specific type of weapon used; compare State v. Raguseo, supra, 225 Conn. 120; no evidence of the specific type of wound inflicted on the victim; compare id.; State v. Chace, 199 Conn. 102, 106, 505 A.2d 712 (1986); and no evidence of prior planning, preparation or motive. Compare State v. Crafts, supra, 251; State v. Pinnock, 220 Conn. 765, 790, 601 A.2d 521 (1992); State v. Grant, 219 Conn. 596, 604, 594 A.2d 459 (1991). We conclude, nonetheless, that the circumstantial evidence and the reasonable inferences drawable therefrom are sufficient to support the jury's finding that the defendant had the intent to kill the victim.

First, according to Lee's testimony, the bloodstain in the carpet in the family room was large, measuring approximately eleven inches by fifteen inches. The stain, on the surface and underside of the carpet and on the underlying padding, was a vivid red, indicating a strong and fresh concentration of blood, as opposed to other liquid that may have entered the saturated area from the later shampooing of the rug. Lee testified that the bloodstain exhibited a direct contact and transfer pattern, and that the blood had soaked through "quite a few layers" of fabric, indicating that the source of the blood was on top of the carpet. By Lee's "very conservative" estimate, the amount of blood that had caused the stain was 1000 cubic centimeters. Lee equated that amount of blood to one liter, or a "little bit over" one quart. He testified that this represented approximately one-fourth of the total blood in the body of a woman of average build.

In this connection, Lee minimized the effect of the liquid from the shampooing process on his estimate of the amount of blood involved. He testified in this respect as follows: "If I use one c.c. of blood, [and] one c.c. of water, you can see this slight pinker color. In other words, fifty/fifty percent dilution. Don't forget, the blood already [has] a fifty/fifty percent dilution. Fifty percent are plasma, those are liquid. Only 45 percent are cells. So you already, if you dilute further, you are going to become one to four dilution. If you drop one drop of [blood] to four drop of water, you barely see a color, pinkish color. What I see is a vivid red color."

Second, there was sufficient evidence for the jury to infer that the fatal wound of the victim was caused by a weapon, rather than by human hands. This inference was permissibly based on the amount of blood found at the scene, and Lee's testimony to the effect that the amount of blood was caused by some instrument that was capable of cutting into a blood vessel or cutting deeply into the body.

Third, as noted above, Lee testified to the effect that, in order for the weapon to have caused a wound yielding that much blood, the weapon would have been required to "cut" very deeply into the body or to have cut a blood vessel, namely, a vein or artery. Thus, the jury could have inferred that the weapon that the defendant used in killing the victim had an edge or point sharp enough and had been used vigorously enough to cut or penetrate deeply enough into the body so as to cause the massive bleeding that Lee estimated.

Lee testified that "unless you cut deep enough or cut into the [blood] vessel, you will not come up with quarts of blood." We do not think, however, that by Lee's use of the word, "cut," rather than "penetrate," for example, the jury was precluded from inferring that he also meant that the weapon could have been sharp enough to penetrate the body so as to cause the massive bleeding that Lee estimated to have occurred. Thus, from its own common experience, taken together with Lee's expert testimony, the jury could have inferred that the weapon penetrated into the body deeply enough to cause the bleeding.

Fourth, the evidence supports the inference that the defendant caused that wound by using a weapon in the family room. The jury, using its common experience, could have inferred that a family room, unlike perhaps a garage or basement, is not the kind of room in a house that would ordinarily contain a weapon capable of causing that kind of wound. From this, the jury could also infer that, before using the weapon, the defendant either had such a weapon in his possession while he was in that room or had obtained it from some other part of the house.

Contrary to the assertion in the dissent of Justice Berdon, the record does not support the inference that the defendant's family room was a natural place for a serious injury to occur. Photographic evidence reveals that the room does not contain a desk or other piece of furniture associated with scissors, and the fireplace instruments appear undisturbed. Further, there was no evidence that blood or cleansers were found on the fireplace instruments, or that any of the fireplace instruments was missing.

Fifth, there was evidence of the defendant's failure to summon medical assistance for the victim. In both State v. Greenfield, supra, 228 Conn. 78, and State v. Francis, 228 Conn. 118, 128-29, 635 A.2d 762 (1993), we noted the defendant's failure to summon medical assistance for his victim as part of the evidence from which the jury could have inferred an intent to kill. The reasoning underlying those cases is that it can be inferred that, if the defendant has caused a grievous wound that could cause the victim's death if not treated promptly, the failure to summon that treatment is consistent with an antecedent intent to cause death.

Sixth, there was very strong evidence of the defendant's consciousness of guilt. This evidence consisted of the defendant's: (1) hiding the victim's body and disposing of the victim's car within a very short period of time; (2) attempting to destroy the bloody evidence, in both his house and his car; (3) sudden flight to Turkey; and (4) use of aliases upon his return to the United States. We have in the past considered consciousness of guilt evidence as part of the evidence from which a jury may draw an inference of an intent to kill. See State v. Patterson, 229 Conn. 328, 333, 641 A.2d 123 (1994); State v. Francis, supra, 228 Conn. 131; State v. Reid, 193 Conn. 646, 656, 480 A.2d 463 (1984).

In sum, we conclude that there was sufficient evidence for the jury reasonably to have inferred that: (1) the defendant killed the victim; (2) by using a weapon with a sharp edge or point with sufficient force to cut a blood vessel, or to cut or penetrate her body deeply enough, so as to cause her to lose one fourth of the blood in her body; (3) prior to using the weapon, the defendant either had the weapon in his possession or had left the family room to obtain it; (4) the victim lay on the rug, bleeding massively from the wound; (5) after inflicting the wound, the defendant did not summon medical assistance for his victim; (6) shortly after killing the victim, the defendant took extraordinary measures to destroy the evidence, and extraordinarily successful measures to hide the body; and (7) he took extraordinary measures to avoid apprehension. Although the evidence certainly did not mandate an inference of an intent to kill, we conclude that, taken together, it reasonably supports the inference of an intent to kill that the jury drew.

We acknowledge that this reasoning requires us to posit a chain of inferences, each link of which may depend for its validity on the validity of the prior link in the chain. That is essentially what circumstantial evidence means, however, and it is what our case law generally permits. We also acknowledge that this is not an easy case. We conclude, nonetheless, that it is a case in which the jury's inference of an intent to kill was sufficiently supported by the circumstantial evidence.

The defendant's reliance on State v. Carpenter, 214 Conn. 77, 570 A.2d 203 (1993), on appeal after remand, 220 Conn. 169, 595 A.2d 881 (1991), cert. denied, 502 U.S. 1034, 112 S.Ct. 877, 116 L.Ed.2d 181 (1992), is misplaced. In Carpenter, we repeated the familiar language regarding proof beyond a reasonable doubt: "It is axiomatic that any conclusion, reasonably to be drawn from the evidence, which is consistent with the innocence of the accused must prevail. State v. Guilfoyle, 109 Conn. 124, 139, 145 A. 761. State v. Foord, 142 Conn. 285, 294, 113 A.2d 591 (1955); State v. Morrill, [ 193 Conn. 602, 610, 478 A.2d 694 (1984)]. The trier may not reach a conclusion of guilt where the facts, established by the evidence, including those reasonably and logically inferred from other proven facts, are rationally consistent with the innocence of an accused. A conclusion of guilt requires proof beyond a reasonable doubt, and proof to that extent is proof which precludes every reasonable hypothesis except that which it tends to support, and is consistent with the defendant's guilt and inconsistent with any other rational conclusion. State v. Smith, 138 Conn. 196, 200, 82 A.2d 816. State v. Foord, supra, 295; State v. Martin, [ 195 Conn. 166, 173, 487 A.2d 177 (1985)]; State v. Morrill, supra, 610-11. Moreover, inferences which do not have a basis in facts established by the evidence cannot be drawn or relied upon to sustain a verdict. State v. Jackson, 176 Conn. 257, 264, 407 A.2d 948 (1978). The jury may not resort to speculation and conjecture. State v. Saracino, 178 Conn. 416, 419, 423 A.2d 102 (1979). If the evidence is insufficient to sustain the burden of proof beyond a reasonable doubt, the verdict must be set aside. State v. Jackson, supra, 262." (Internal quotation marks omitted.) State v. Carpenter, supra, 84.

The defendant argues that, in light of these principles, his conviction of murder must be reversed because the evidence did not preclude the reasonable hypothesis that he killed the victim with a reckless or less culpable state of mind, as opposed to an intentional state of mind. Although the defendant's argument has plausibility based upon a literal reading of the language of Carpenter and, arguably, upon the way in which the court in Carpenter applied those principles to the facts of that case, we conclude that the argument is based upon a fundamental misconception of the relationship between the principles articulated in Carpenter and our well established and traditional scope of review of jury verdicts.

Just as the principles regarding the necessity of proof beyond a reasonable doubt articulated in Carpenter are "axiomatic"; id., 84; the principles regarding our scope of review of jury verdicts are also axiomatic. "We first review the evidence presented at trial, construing it in the light most favorable to sustaining the facts expressly found by the trial court or impliedly found by the jury. We then decide whether, upon the facts thus established and the inferences reasonably drawn therefrom, the trial court or the jury could reasonably have concluded that the cumulative effect of the evidence established the defendant's guilt beyond a reasonable doubt. . . . State v. Joyner, 225 Conn. 450, 455, 625 A.2d 791 (1993). [I]n viewing evidence which could yield contrary inferences, the jury is not barred from drawing those inferences consistent with guilt and is not required to draw only those inferences consistent with innocence. The rule is that the jury's function is to draw whatever inferences from the evidence or facts established by the evidence it deems to be reasonable and logical. . . . State v. Grant, [supra, 219 Conn. 604]." (Internal quotation marks omitted.) State v. Francis, supra, 228 Conn. 127; see also State v. Patterson, supra, 229 Conn. 332; State v. Greenfield, supra, 228 Conn. 76; State v. Rasmussen, 225 Conn. 55, 73-74, 621 A.2d 728 (1993).

If Carpenter means what the defendant suggests it means, its principles would be in irreconcilable conflict with our scope of review of jury verdicts, because those principles would require an appellate court reviewing the sufficiency of the evidence simultaneously to perform two inherently contradictory functions. The first appellate function, under Carpenter, would be to view the evidence adduced at trial in such a way as to require that "any conclusion, reasonably to be drawn from the evidence, which is consistent with the innocence of the accused must prevail." (Internal quotation marks omitted.) State v. Carpenter, supra, 214 Conn. 84. Moreover, the reviewing court must ensure that the evidence "precludes every reasonable hypothesis except that which it tends to support, and is consistent with the defendant's guilt and inconsistent with any other rational conclusion." (Internal quotation marks omitted.) Id. Read and applied literally, this language would mean that, in reviewing the sufficiency of the evidence to support a guilty verdict, an appellate court would be required to give deference, not to the view of the evidence taken by the jury that returned a guilty verdict, but to any reasonable hypothesis supplied by the defendant that, had it been accepted by the jury, would have resulted in an acquittal. Furthermore, it would mean that in any case in which the defendant has posed a reasonable hypothesis of innocence — a hypothesis, in other words, based on the evidence or purported lack thereof, that a reasonable jury could have accepted — the appellate court reviewing the defendant's sufficiency of evidence claim would be required to mandate a judgment of acquittal.

Indeed, under that view taken to its logical limit, it is difficult to imagine a case having gone to trial in which a guilty verdict would stand on appeal. For example, in any case in which the defendant presented plausible evidence of an alibi, or of a claim of self-defense, taking State v. Carpenter, supra, 214 Conn. 77, to mean what the defendant's argument suggests would require a judgment of acquittal on appeal.

This, of course, would be directly contrary to our traditional scope of review of jury verdicts, and to the way in which we traditionally employ it. Under that scope of review and its application, we give deference not to the hypothesis of innocence posed by the defendant, but to the evidence and the reasonable inferences drawable therefrom that support the jury's determination of guilt. On appeal, we do not ask whether there is a reasonable view of the evidence that would support a reasonable hypothesis of innocence. We ask, instead, whether there is a reasonable view of the evidence that supports the jury's verdict of guilty.

We believe, however, that there are two reasons that explain the apparent contradiction between the Carpenter principles and the principles regarding our traditional scope of review. The first reason is that the Carpenter principles, although not irrelevant to the appellate function of reviewing a claim of evidentiary insufficiency, are more properly viewed as principles that relate to the jury's function of applying the concept of proof beyond a reasonable doubt than to the function of an appellate court in reviewing the jury's verdict. That is, the jury, in sorting out the evidence and determining whether the state has proved its case beyond a reasonable doubt, must be instructed to, and must in fact, view the evidence in such a way that it cannot render a guilty verdict unless it is satisfied that the evidence meets the standards articulated in Carpenter. Once the jury has done that, however, and has determined that, in its best judgment, the hypothesis or hypotheses of innocence posed by the defendant are no more than "possible" as opposed to "reasonable"; see State v. Little, 194 Conn. 665, 672, 485 A.2d 913 (1984); that jury determination is entitled to deference on appeal. It would be inconsistent with the entire process of trial fact-finding for an appellate court to do otherwise. Thus, viewed through this prism, the Carpenter principles have their primary operation as rules of law for the guidance of the fact finder, rather than for the guidance of appellate courts in reviewing the sufficiency of evidence regarding the fact finder's verdict.

The second reason involves the appropriate, albeit more limited, role that the Carpenter principles do play on the appellate level. Rather than contradict the traditional scope of review, those principles supplement that scope of review in the following way. The appellate court's first task, in responding to a claim of evidentiary insufficiency, is to apply the traditional scope of review to the evidence. That requires that, as we have done in this case, we view all of the evidence, and the reasonable inferences drawable therefrom, in favor of the jury's verdict. We then determine the sum of that evidence and the inferences, and for purposes of analysis, we assume that the facts established by that sum are true.

If those facts, taken as true, are sufficient to have met the state's burden of proving all of the elements of the crime charged beyond a reasonable doubt, the verdict must stand. In that circumstance, the Carpenter principles do not come into play, because to do so would be to engage in the inherently contradictory process described above, and would in effect permit an appellate court to substitute a different view of the evidence from the reasonable view taken by the jury. Put another way, once those facts are taken as true for purposes of an appellate challenge to the sufficiency of the evidence, those facts necessarily preclude the hypothesis of innocence posed by the defendant, because if those facts are true then the defendant's hypothesis becomes only possible as opposed to reasonable.

There are cases, however, in which those facts, taken as true, do not establish all of the elements of the crime charged. In such a case, the Carpenter principles do come into play, and require a judgment of acquittal. Under this view, therefore, the application of the Carpenter principles, as applied to an evidentiary insufficiency claim on appeal, require the conclusion that the evidence did not preclude all reasonable hypotheses of innocence. In such a case, therefore, the Carpenter principles reinforce the appellate judgment that the evidence adduced was not sufficient to establish guilt beyond a reasonable doubt, without permitting an appellate court to substitute its judgment of the evidence from that reasonably reached by the jury. Put another way, an appellate conclusion that the evidence adduced falls short of establishing any element of the crime charged means, by hypothesis, that the evidence was consistent with a rational conclusion other than guilt. See State v. Carpenter, supra, 214 Conn. 84.

Applying this analysis to the defendant's reliance on Carpenter, we conclude that Carpenter is unavailing to the defendant. Having concluded that, from the circumstantial evidence and the reasonable inferences drawable therefrom, the jury could have reasonably inferred an intent to kill, we assume that fact — an intent to kill held by the defendant — to be true for the purpose of the defendant's challenge based upon evidentiary insufficiency. Therefore, for that purpose the defendant's hypothesis of a different state of mind that is inconsistent with a conviction of murder, namely, recklessness or criminal negligence, is only possible and not reasonable. Put another way, having so concluded, it would be inappropriate for us to view the evidence in the manner suggested by Carpenter.

II

The defendant also claims that the trial court improperly declined to instruct the jury regarding lesser included offenses of homicide. The defendant requested that the trial court charge the jury on the following lesser included offenses: manslaughter in the first degree in violation of General Statutes § 53a-55 (a)(1) and (3), manslaughter in the second degree in violation of General Statutes § 53a-56 (a)(1), and criminally negligent homicide in violation of General Statutes § 53a-58 (a). The trial court declined to instruct the jurors on any of these offenses, and the defendant took exception. We conclude that the defendant was entitled to have the jury consider lesser included offenses.

General Statutes § 53a-55 provides in relevant part: "MANSLAUGHTER IN THE FIRST DEGREE: CLASS B FELONY. (a) A person is guilty of manslaughter in the first degree when: (1) With intent to cause serious physical injury to another person, he causes the death of such person or of a third person . . . or (3) under circumstances evincing an extreme indifference to human life, he recklessly engages in conduct which creates a grave risk of death to another person, and thereby causes the death of another person." (Emphasis added.) General Statutes § 53a-3 (4) provides: "`Serious physical injury' means physical injury which creates a substantial risk of death, or which causes serious disfigurement, serious impairment of health or serious loss or impairment of the function of any bodily organ."

General Statutes § 53a-56 provides in relevant part: "MANSLAUGHTER IN THE SECOND DEGREE: CLASS C FELONY. (a) A person is guilty of manslaughter in the second degree when: (1) He recklessly causes the death of another person. . . ." General Statutes § 53a-3 (13) provides: "A person acts `recklessly' with respect to a result or to a circumstance described by a statute defining an offense when he is aware of and consciously disregards a substantial and unjustifiable risk that such result will occur or that such circumstance exists. The risk must be of such nature and degree that disregarding it constitutes a gross deviation from the standard of conduct that a reasonable person would observe in the situation."

General Statutes § 53a-58 provides in relevant part: "CRIMINALLY NEGLIGENT HOMICIDE: CLASS A MISDEMEANOR. (a) A person is guilty of criminally negligent homicide when, with criminal negligence, he causes the death of another person. . . ." General Statutes § 53a-3 (14) provides: "A person acts with `criminal negligence' with respect to a result or to a circumstance described by a statute defining an offense when he fails to perceive a substantial and unjustifiable risk that such result will occur or that such circumstance exists. The risk must be of such nature and degree that the failure to perceive it constitutes a gross deviation from the standard of care that a reasonable person would observe in the situation."

"A defendant is entitled to an instruction on a lesser offense if, and only if, the following conditions are met: (1) an appropriate instruction is requested by either the state or the defendant; (2) it is not possible to commit the greater offense, in the manner described in the information or bill of particulars, without having first committed the lesser; (3) there is some evidence, introduced by either the state or the defendant, or by a combination of their proofs, which justifies conviction of the lesser offense; and (4) the proof on the element or elements which differentiate the lesser offense from the offense charged is sufficiently in dispute to permit the jury consistently to find the defendant innocent of the greater offense but guilty of the lesser." State v. Whistnant, 179 Conn. 576, 588, 427 A.2d 414 (1980); see State v. Rodriguez, 180 Conn. 382, 407-408, 429 A.2d 919 (1980).

The defendant claims that, under Whistnant, he was entitled to a lesser included offense charge with respect to all of the lesser degrees of homicide. The state concedes that the first two Whistnant requirements are met. The states argues, however, that the third and fourth prongs of Whistnant are not satisfied by the facts of this case. We agree with the defendant.

The third prong of the Whistnant test requires the existence of some evidence justifying conviction of the lesser offense. The critical element distinguishing murder from its lesser included offenses is intent, "often the most significant and, at the same time, the most elusive element of the crime charged." State v. Rodriguez, supra, 180 Conn. 404. We must determine if the "evidence suggests at least a possibility" that the defendant acted with a lesser intent than that of the specific intent to kill. State v. Falby, 187 Conn. 6, 30, 444 A.2d 213 (1982). If "we cannot as a matter of law exclude this possibility"; id.; then the defendant was entitled to lesser included offense instructions.

Although we have concluded that the evidence was sufficient to support a finding of intent to kill, we agree with the defendant that the evidence was such as to permit the jury to infer that there was also a possibility that "there was a sudden or spontaneous incident on April 18, 1988, resulting in the unintended death of [the victim]." From the amount of blood, and the setting of the injury — a family room in a home, where one does not expect accidental conduct to produce an injury causing the loss of that much blood — a jury could reasonably infer that the defendant acted with an intent to cause the victim serious injury or with extreme indifference toward human life; see footnote 11; or with criminal recklessness, which is an awareness and conscious disregard of a substantial and unjustifiable risk; see footnote 12; or with criminal negligence, which is conduct manifesting a failure to perceive a substantial and unjustifiable risk. See footnote 13.

The defendant also claims that the jury should have been instructed on the theory that the defendant killed the victim under extreme emotional disturbance, pursuant to General Statutes § 53a-55 (a)(2). The defendant does not identify evidence that would support this affirmative defense in the present case. We do not consider this claim further, but merely note that the trial court, if a new trial is held, should instruct only on those lesser offenses and those defenses that are supported by the evidence. State v. Campbell, 225 Conn. 650, 659, 626 A.2d 287 (1993).

The fourth prong of Whistnant requires that we determine if the proof on intent that differentiates the lesser offense from the offense charged was sufficiently in dispute to permit the jury consistently to find the defendant innocent of the greater offense but guilty of the lesser. As the discussion in the previous section indicates, the evidence of intent to kill was not overwhelming. We also note that prior cases in which this court has held that the defendant was not entitled to any lesser included offense instructions are distinguishable from the present case. For example, in State v. Crafts, supra, 226 Conn. 251, there was evidence of prior planning and preparation — the defendant had arranged for the rental of a woodchipper to dispose of the victim's body prior to causing the victim's death, making the defendant's claim of a "sudden confrontation . . . too speculative to put the issue of intent sufficiently in dispute." We conclude that the trial court should have instructed the jury on those lesser included offenses that were supported by the evidence.

The defendant also argues that the trial court's failure to instruct the jury on lesser included offenses offended certain constitutional principles, as well as General Statutes § 53a-45 (c) ("[t]he court or jury before which any person . . . held to answer for murder . . . is tried may find such person guilty of homicide in a lesser degree than that charged"). Because we hold that the jury should have been instructed on lesser included offenses under State v. Whistnant, supra, 179 Conn. 588, we need not consider these claims.

III

The defendant next claims that evidence obtained pursuant to the search of his home by Trumbull and state police should have been suppressed by the trial court because the search warrant affidavit did not state probable cause and therefore was invalid under the fourth amendment to the United States constitution. We disagree.

The fourth amendment to the United States constitution provides: "The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no warrants shall issue, but upon probable cause, supported by oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized." The defendant also refers to article first, § 7, of the state constitution as a ground for suppression of the evidence obtained pursuant to the search of his house. The defendant, however, does not advance an analysis of this issue under the state constitution separate from his analysis under federal authorities, and accordingly we will review his claim only under the federal constitution. State v. Vincent, 229 Conn. 164, 168 n. 5, 640 A.2d 94 (1994).

On April 20, 1988, the Trumbull police obtained a search warrant for the defendant's Kitcher Court residence, his automobile, and his person. The warrant identified as the evidence sought to be seized the victim's person or body, her clothing, blood, body parts or fluids, hairs and other items that would circumstantially establish the victim's presence in the defendant's home. Pursuant to that warrant, the police searched Kitcher Court on April 21, 1988, as previously described in this opinion. The defendant filed a motion to suppress the evidence, which the trial court denied. We determine that the affidavit stated probable cause to search the defendant's residence.

See part IV of this opinion.

Under the fourth amendment, "[p]robable cause to search exists if: (1) there is probable cause to believe that the particular items sought to be seized are connected with criminal activity or will assist in a particular apprehension or conviction . . . and (2) there is probable cause to believe that the items sought to be seized will be found in the place to be searched." (Emphasis in original; internal quotation marks omitted.) State v. Vincent, 229 Conn. 164, 171, 640 A.2d 94 (1994).

"The standard of review for an issuing judge's determination that probable cause existed to issue a search warrant is to consider the information before the issuing judge at the time of the issuance of the warrant, together with the reasonable inferences drawn from such information, in the light most favorable to the issuing judge's determination of probable cause. . . . In determining whether probable cause exists to conduct a search, a totality of the circumstances test is used. Illinois v. Gates, 462 U.S. 213, 233, 103 S.Ct. 2317, 76 L.Ed.2d 527 (1983); State v. Johnson, [ 219 Conn. 557, 562, 594 A.2d 933 (1991)]; State v. Barton, 219 Conn. 529, 545, 594 A.2d 917 (1991). [P]robable cause is a fluid concept — turning on the assessment of probabilities in particular factual contexts — not readily, or even usefully, reduced to a neat set of legal rules. Illinois v. Gates, supra, 232. In determining the existence of probable cause to search, the magistrate should make a practical, commonsense decision whether, given all of the circumstances set forth in the affidavit . . . there is a fair probability that contraband or evidence of a crime will be found in a particular place. Id., 238; State v. Johnson, supra, 563. In making this determination [of probable cause], the magistrate is entitled to draw reasonable inferences from the facts presented. When a magistrate has determined that the warrant affidavit presents sufficient objective indicia of reliability to justify a search and has issued a warrant, a court reviewing that warrant at a subsequent suppression hearing should defer to the reasonable inferences drawn by the magistrate." (Citations omitted; internal quotation marks omitted.) State v. Zarick, 227 Conn. 207, 222-23, 630 A.2d 565 (1993).

The search warrant affidavit sets out the circumstances of a "missing person complaint." It alleges that the victim went to the defendant's house on April 18, 1988, to give the defendant a massage, and that as of the time of the warrant issuance, two days later, she had not been heard from since that appointment. According to established procedure, she telephoned Andre's when she arrived at Kitcher Court, and identified the defendant by his driver's license.

The affidavit provides: "That the affiant, Detective Robert Duva, is a regular member of the Trumbull, Connecticut police department and has been a regular member of said department for a period of 18 years and amongst his duties is the investigation of missing persons and related cases.
"That James Arlio is a regular member of the Trumbull Police Department and has been a regular member of said department for a period of 10 years and amongst his duties is the investigation of Missing Persons and related cases.
"That the undersigned was detailed to investigate a missing person complaint reported by `Andre's' Massage of Hadlyme, Connecticut on April 19, 1988.
"Subsequent investigation revealed that the identity of the person reported missing was said to be [the victim], DOB 04-22-65, a white female having blonde hair, approximately 135 pounds, who had been given a massage assignment at 37 Kitcher Court, Trumbull, CT. She arrived at this address and telephoned `Andre's' to check in at 3:40 p.m., April 18, 1988. This is a steadfast rule that is always followed. At this time, [the victim] related to `Andre's' operator that she had checked the client's Connecticut Photo Driver's license and said that everything was fine and that she would call in an hour. The procedure adhered to required her to telephone the `Andre's' operator when she was to depart.
"When [the victim] had not done so by 4:45 p.m., 04-18-88, `Andre's' operator Paula Doak called the telephone number given by the client who had identified himself to be Tevfik Sivri. Receiving no answer, the telephone number was tried again a few minutes later, also getting no response. At this time, she was instructed to notify the Trumbull Police Department and did so.
"The following investigation revealed that a red and white VW Van with New Hampshire plates had been observed in the driveway of 37 Kitcher Court sometime between 3:30 to 4 p.m. on April 18, 1988 by an area newspaper carrier. [The victim] had been driving a red and white 1976 VW Microbus bearing New Hampshire Registration 172656 when last seen by family members on 04-18-88.
"Officers detailed to 37 Kitcher Court, Trumbull, CT at 5:56 p.m., 04-18-88, found no one at home, continued to check the area, and, upon rechecking 37 Kitcher Court, found that Mrs. Sivri had returned home. After explaining the situation and asking permission to check the interior of the house, they were quickly denied by her son who interjected an emphatic denial.
"Tevfik Sivri, the person identified to be the client, returned home at 7:30 p.m., 04-18-88, while officers were at the 37 Kitcher Court residence. He was asked to come to Police Headquarters to be interviewed. Upon doing so at 8 p.m., 04-18-88, he admitted [the victim] had been at his home but said she had left. He would give no further information and asked to leave, which he did.
"Investigation further revealed that [the victim] was extremely punctual, that she never missed a check-in during her employ, that she never overstayed any appointment that would prevent her from picking up her infant son before 6 p.m., and that she also was timely in picking up her husband at work at approximately 5 p.m. She did none of these the night of 04-18-88, and has not contacted any family, friends, or co-employees since, nor has anyone heard from her since 3:40 p.m., 04-18-88, when she telephoned from the Sivri residence at 37 Kitcher Court, Trumbull, CT.
"The undersigned interviewed a coworker who had a close friend relationship with [the victim] and had been with her the morning and early afternoon of 04-18-88 prior to the 37 Kitcher Court, Trumbull appointment. She related that [the victim] was happy and content and had given no indication of going anywhere or doing anything out of her normal routine. She stated that in no way would [the victim] leave without her child, who is an infant child.
"On 04-20-88, the undersigned was notified by Bridgeport Police Department the red and white VW Microbus had been located in Bridgeport. Subsequent investigation revealed that the vehicle had been first observed on 04-18-88 approximately 8:00 p.m. abandoned on Waterview Avenue under the Route 95 Expressway just South of Father Panik Village.
"That based upon the facts set forth in the above affidavit, reasonable grounds exist to believe the aforementioned dwelling and vehicles contain evidence that a crime has been committed and a Search and Seizure Warrant should be issued and executed."

The warrant stated that the victim failed to report in by telephone after her appointment, which violated established procedure, although she had never before missed a check-in. She also failed to pick up her infant son and boyfriend, although she had always been timely in fulfilling these responsibilities. Her van was found abandoned on Waterview Avenue in Bridgeport at approximately 8 p.m. on the day of her disappearance.

The affidavit incorrectly refers to the victim's boyfriend as her "husband."

The warrant set out the following additional facts that weigh against the possibility that the victim voluntarily and permanently left the area, and support probable cause to believe that she disappeared as a result of criminal activity. According to her coworker, the victim had been happy and content and had given no indication of going away or doing anything out of her normal routine. The coworker also indicated that the victim would not have left without her infant son. That the victim's van was found abandoned hours after the victim was at Kitcher Court also weighed against an inference that the victim voluntarily and permanently left the area. We conclude that the affidavit set out sufficient facts to support probable cause that a crime had been committed.

The defendant claims, however, that the affidavit did not establish probable cause that evidence of a crime would be found at Kitcher Court. We disagree, and conclude that the warrant affidavit demonstrated probable cause to search the Kitcher Court premises.

The defendant admitted that the victim had been at his house on the date of her disappearance. He was the last person to have seen the victim. In view of the fact that there was probable cause to believe that a crime had been committed upon the victim, there was probable cause to search the last place the victim had been seen, particularly when that place was the home of the last person to have seen her.

The defendant also claims that the fact that almost three days had elapsed since the victim's disappearance rendered stale the information that was used to support the warrant. The defendant concedes that the passage of time "strengthened the inference that [the victim's] disappearance was due to foul play," but argues that any inference that evidence of the victim's disappearance might still remain at Kitcher Court would be unreasonable. We disagree.

"The determination of probable cause to conduct a search depends in part on the finding of facts so closely related to the time of the issuance of the warrant as to justify a belief in the continued existence of probable cause at that time. . . . Although it is reasonable to infer that probable cause dwindles as time passes, no single rule can be applied to determine when information has become too old to be reliable." (Citation omitted; internal quotation marks omitted.) State v. Vincent, supra, 229 Conn. 174. At one end of the spectrum is Vincent, in which we upheld an investigatory search warrant of Vincent's house that sought to locate evidence related to the disappearance of his daughter one year earlier because the items sought — children's photographs, childhood records and artifacts — are commonly kept by parents over extended periods of time. Id., 175. Although the items sought in the present case do not have similar lasting value, only three days had elapsed between the victim's disappearance and the issuance of the warrant. It was not unreasonable for the issuing magistrate to infer that evidence of the victim's whereabouts, or evidence that the victim had been physically assaulted, would be found at the defendant's residence several days after the victim's disappearance.

The defendant further argues that much of the information in the affidavit has its source in unidentified or ambiguously identified third parties, and therefore the issuing judge could not have ascertained the reliability of this information. Although the affidavit expressly identifies Paula Doak as the source of much of the crucial information relied on for probable cause, including the circumstances of the victim's visit to Kitcher Court, other facts set out in the affidavit have their origin in a newspaper carrier not identified by name, a coworker of the victim's also not identified by name, and the "Bridgeport police department."

When hearsay information from third parties is relied on by the affidavit, the "veracity" or "reliability" and "basis of knowledge" of those persons are "highly relevant" in determining the existence of probable cause. State v. Barton, supra, 219 Conn. 552. The veracity factor receives little emphasis, however, in a case such as this, in which the "informants" — a newspaper carrier, a coworker, and a police department — are not anonymous figures from the criminal underworld, but instead are citizens reporting their personal observations. See State v. Daley, 189 Conn. 717, 724, 458 A.2d 1147 (1983); see 1 W. LaFave, Search and Seizure (2d Ed. 1987) § 3.4. Furthermore, a sufficient basis of the knowledge of each of these citizen informants is indicated in the affidavit to support probable cause: personal observation by the newspaper carrier; discovery and possession of the victim's van by Bridgeport police; and the fact that the other informant is a coworker. Finally, the most significant fact alleged in the warrant supporting both that a crime had been committed, and that evidence of the crime could be found at Kitcher Court, namely, that the victim was last heard from at Kitcher Court, was provided by a named person, Doak, and confirmed by the defendant.

The defendant claims that because certain paragraphs of the affidavit contained ambiguous language, "the issuing judge was largely unable to determine from the affidavit which of the statements are based upon the affiant's personal knowledge, which are the result of hearsay and whether such hearsay was direct or not." Although the affidavit might have indicated more clearly the precise source of each piece of information relied on, the crucial information supporting probable cause in the affidavit was provided by named parties.

IV

The defendant next claims that the trial court improperly admitted certain evidence yielded by the search of the defendant's automobile. The trial court held that the search warrant affidavit discussed in the previous section of this opinion did not state probable cause to search the automobile because, although the warrant directed a search of the automobile, its affidavit did not mention any automobile and the automobile was not on the premises of the defendant's residence when it was searched. We assume, without deciding, that the trial court properly made this determination. Nevertheless, the trial court held a suppression hearing and heard testimony on the issue of abandonment, and ultimately admitted the evidence obtained pursuant to the search of the automobile on the ground that the automobile had been abandoned. The defendant argues that the trial court should have granted his motion to suppress. We disagree.

See footnote 18.

The paragraph of the search warrant listing places to be searched sets out the following: "Single family dwelling, #37 Kitcher Court Trumbull, Connecticut. 1987 Mitsubishi bearing Connecticut registration 226 EZP, vehicles normally parked at the premises, surrounding grounds." There is no language indicating that the Mitsubishi belonged to the defendant.

On April 25, 1988, police officers located the defendant's car on Noble Avenue in Bridgeport. Its license plates had been removed. Police officer Joseph Velky identified the car by the vehicle identification number. A citizen, Edward Gomes, spoke with Trumbull police officer Shawn Sember and told Sember that the vehicle had been parked there for approximately two or three days, during which period Gomes had not seen anyone enter or exit the vehicle. The police towed the vehicle to the Trumbull police department garage. The defendant correctly points out that the search of his automobile, even if predicated on probable cause, was not lawful under the "automobile exception" to the warrant requirement. Under the state constitution, the police must obtain a search warrant to search an automobile, even if there is probable cause to search the automobile after it has been towed to the police department garage. State v. Miller, 227 Conn. 363, 377, 630 A.2d 1315 (1993). The state argues, however, that Miller is not relevant because in the present case the defendant's conduct evinced an abandonment of his reasonable expectation of privacy in the contents of the automobile. We agree with the state.

The court took judicial notice of Bridgeport ordinances indicating that the defendant's car had been legally parked on Noble Avenue.

The state makes no claim that its search of the defendant's vehicle was an inventory search. See State v. Miller, supra, 227 Conn. 377-78 n. 14.

"Whether property has been `abandoned' . . . does not depend on where legal title rests, or whether one asserting a Fourth Amendment right has a legally enforceable possessory interest in the property; the question, rather, is whether the person claiming the protection of the Fourth Amendment has a legitimate expectation of privacy in the invaded place. . . . In essence, what is abandoned is not necessarily the defendant's property, but his reasonable expectation of privacy therein." (Citations omitted; internal quotation marks omitted.) State v. Mooney, 218 Conn. 85, 107, 588 A.2d 145, cert. denied, 502 U.S. 919, 112 S.Ct. 330, 116 L.Ed.2d 270 (1991). Therefore, the fact that the defendant retained legal title to his automobile does not compel a finding that he retained a reasonable expectation of privacy in its contents. United States v. Thomas, 864 F.2d 843, 845 (D.C. Cir. 1989); 1 W. LaFave, supra, § 2.6(b), pp. 464-65.

Although the defendant argues that the dissent in State v. DeFusco, 224 Conn. 627, 620 A.2d 746 (1993), and the majority opinion in State v. Miller, supra, 227 Conn. 363, are authority for a heightened right of privacy in Connecticut under the state constitution, the defendant does not adequately articulate reasons why the abandonment doctrine under the state constitution provides greater substantive protection than the doctrine provides under the federal constitution. Accordingly, we review the defendant's claim solely under the federal constitution.

"[I]t is relevant although not necessary to the fourth amendment abandonment inquiry whether the defendant manifested by his conduct an intent to shed, albeit temporarily, his expectation of privacy in the item or container involved." State v. Mooney, supra, 218 Conn. 107. "[I]ntent may be inferred from words, acts and other objective facts." (Internal quotation marks omitted.) United States v. Kendall, 655 F.2d 199, 201 (9th Cir. 1981), cert. denied, 455 U.S. 941, 102 S.Ct. 1434, 71 L.Ed.2d 652 (1982).

In the present case, the defendant's automobile was found parked on a street in Bridgeport one week after the victim's disappearance. Of course, "cars are regularly parked on the street for brief periods of time without an expectation that they will thereby be subject to entry." 1 W. LaFave, supra, § 2.5(b), pp. 446-47.

Nevertheless, the facts and circumstances revealed at the suppression hearing, and known to the police at the time of their search of the defendant's automobile, support the trial court's determination that the defendant abandoned his expectation of privacy in his vehicle. First, on April 18, the day of the victim's disappearance, the police spoke with the defendant, who admitted that he had been with the victim at his house. On the morning of April 21, 1988, three days after the victim's disappearance, the defendant's house was searched. The defendant's brother, Refik, was on the premises at the time of the search, and was served with the warrant, which identified the automobile by make and marker number as a target of the search. There were blood stains visible in the house. The defendant left the country on the same day as the search of his home. The police learned that the defendant had flown from Ottawa, Canada, to Istanbul, Turkey. When the police found the defendant's car on April 25, they obtained information that it had been left on that street for days and had not been moved, and they observed that its license plates had been removed. The police had all of this information at the time that they searched the defendant's vehicle.

From this information, it is a reasonable inference that Refik informed the defendant that the police were searching his home for evidence of a killing, and also intended to search his car and his person. It is also a reasonable inference that, upon obtaining this knowledge, the defendant abandoned his automobile on a public street miles from his home, where it sat for several days prior to its discovery by the police, and had Refik drive him to Canada so that he could flee the country. Having fled the country, and having removed the automobile's license plates so that it could not be driven legally and was no longer readily identifiable as the property of the defendant, the defendant cannot claim a reasonable expectation of privacy in its contents.

This case is similar to United States v. Gulledge, 469 F.2d 713 (5th Cir. 1972). In Gulledge, a service station attendant gave permission to two men to leave a U-Haul trailer in the parking lot for two to three days. After the trailer had remained for ten days and the men had not returned, the attendant became suspicious and called the police. The police searched the trailer and found stolen goods. Under these circumstances, the Fifth Circuit Court of Appeals upheld the trial court's finding of abandonment. Id., 715.

V

Finally, the defendant claims that the trial court improperly admitted expert testimony regarding the population frequency calculation used by Cellmark Diagnostics (Cellmark) to assess the significance of deoxyribonucleic acid (DNA) profiling evidence. The defendant filed a motion in limine, seeking to suppress evidence of the results of comparisons of DNA extracts from the blood of the victim's parents and blood found in the automobile's trunk. The trial court held a pretrial evidentiary hearing (DNA hearing) in December, 1991, and January, 1992, at which four expert witnesses testified regarding the scientific validity of the DNA analysis conducted by Cellmark. The trial court denied the motion to suppress and subsequently allowed the admission of this evidence at trial.

As the trial court noted, at the time of trial DNA typing analysis was performed in only three laboratories in the United States: Cellmark; Lifecodes, of New York; and a laboratory operated by the Federal Bureau of Investigation.

"DNA, the fundamental genetic material of all animals and plants, is composed of two parallel chain-like structures, each segment of which is a pair of chemical bases. The sequence of the base pairs determines each individual's genetic traits. Most DNA does not vary from person to person, but certain parts of the DNA, called `polymorphic loci,' vary distinctly from one person to another." State v. Hammond, 221 Conn. 264, 281, 604 A.2d 793 (1992). This court has examined DNA typing evidence in only two appeals. State v. Skipper, 228 Conn. 610, 613-24, 637 A.2d 1101 (1994); State v. Hammond, supra, 264. These cases dealt with different aspects of DNA evidence however, and the issues raised by the defendant in the present case are of first impression in this state.

In State v. Hammond, supra, 221 Conn. 288-89, this court ordered the trial court to reconsider the defendant's motion to set aside a verdict of guilty of sexual assault in the first degree because of an examination of blood typing and DNA typing evidence that had excluded the defendant as a possible donor of sperm found on the victim's underwear. In State v. Skipper, supra, 228 Conn. 624, we reversed the defendant's conviction because testimony as to a "probability of paternity" statistic was based on assumptions that violated the presumption of innocence. In both cases, this court noted that the broader issues as to the admissibility of DNA typing testimony in the courts of this state have yet to be decided. Id., 614 n. 5; State v. Hammond, supra, 280 n. 8.

At the DNA hearing, the state relied on the expert testimony of Robin Cotton, Ph.D., deputy director at Cellmark, Lisa Forman, Ph.D., a population geneticist employed by Cellmark, and Kenneth Kidd, Ph.D., a professor at Yale University. The state's experts described a three step process used to compare different samples of DNA.

Henry Lee also testified about DNA evidence generally, but did not discuss the population frequency calculation at issue in this appeal.

First, the experts described the process, called restriction fragment length polymorphism (RFLP), employed to extract discrete polymorphic DNA segments, called alleles, from blood samples for the purpose of comparison. Second, the expert testimony indicated that computer imaging equipment is utilized to produce a visual pattern of paired bands of different lengths representing the alleles. These can be visually compared in order to determine whether a "match" has occurred. Cotton testified that on three of the four DNA probes used to examine the blood samples in this case, there were paired bands that matched bands from the victim's mother and father.

The court in State v. Hammond, supra, 221 Conn. 281-82, summarized the RFLP procedure as follows: "`Restriction enzymes' are able to recognize a specific sequence of base pairs along a single strand of DNA, and to cut the DNA at that location. This enables the researcher to conduct restriction fragment length polymorphism (RFLP) analysis, which involves a number of steps: (1) extraction of the DNA from the sample; (2) fragmentation of the DNA by restriction enzymes; (3) gel electrophoresis, in which the fragmented sample is placed upon an electrically charged agarose gel, where it forms a distinctive pattern while moving from the negative to the positive charge; (4) Southern blotting, in which the DNA pattern is transferred to a more durable nylon membrane; (5) hybridization, in which DNA patterns specific to the individual are revealed by the use of radioactive probes designed to seek out and bond with polymorphic regions of the DNA; and (6) autoradiography, in which a piece of film is developed on top of the nylon membrane, revealing the location of the radioactive probes by bands on the film."

Of the two bands resulting from the fourth probe, one had a band that matched a band of the father's but the other did not match that of either the mother or the father. Cotton testified that this mismatch did not demonstrate a misidentification, but instead most likely represented a mutant gene inherited from the mother, which happens about 5 percent of the time.

The trial court properly noted that the RFLP procedure has only been in use forensically in this country for several years, "and the system and lab techniques are being changed and refined continuously." Nevertheless, the basic procedures employed, both in theory and in laboratory practice, repeatedly have been recognized as generally accepted in the scientific community, and are not the subject of challenge by the defendant in this appeal. Indeed, the trial court noted that the experts who testified were in agreement on the general acceptance of these procedures. See, e.g., United States v. Jakobetz, 955 F.2d 786, 799 (2d Cir.), cert. denied, 506 U.S. 834, 113 S.Ct. 104, 121 L.Ed.2d 63 (1992); State v. Bible, 175 Ariz. 549, 582, 858 P.2d 1152 (1993), cert. denied, ___ U.S. ___ 114 S.Ct. 1578, 128 L.Ed.2d 221 (1994) (holding that Cellmark's "method of declaring a match . . . [is] generally accepted in the relevant scientific community"); State v. Cauthron, 120 Wash.2d 879, 896, 846 P.2d 502 (1993) ("no court has rejected RFLP testing on the basis that it was not generally accepted by the scientific community"; collecting cases).

In this opinion, we refer to the Frye standard of "general acceptance" for determining the admissibility of scientific evidence; Frye v. United States, 932 F. 1013, 1014 (D.C. Cir. 1923); because we have applied that standard in the past to novel scientific evidence that is beyond the ken of the average juror; see State v. Hasan, 205 Conn. 485, 489-90, 534 A.2d 877 (1987); and because neither the state nor the defendant claimed in the trial court or on appeal that any test for the admission of scientific evidence should apply to DNA evidence other than the Frye test. We note, however, that, interpreting the Federal Rules of Evidence, the United States Supreme Court recently abandoned the Frye test. Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993); see State v. Borrelli, 227 Conn. 153, 163 n. 10, 629 A.2d 1105 (1993).

Of course, the RFLP process, like any other scientific process, is subject to human error. In addition, various problems arising in the process can call into question the results in a particular case, including contamination of samples and other technical problems. See State v. Cauthron, supra, 120 Wash.2d 898. Although some courts have excluded DNA typing evidence because of these potential technical problems; People v. Castro, 144 Misc.2d 956, 545 N.Y.S.2d 985 (1989); others have reasoned that in view of the overwhelming scientific and judicial acceptance of the process itself, potential problems in its administration in a particular case "go to the weight rather than the admissibility of the testimony." State v. Cauthron, supra, 889; see also Fishback v. People, 851 P.2d 884, 893 (Colo. 1993); State v. Vandebogart, 136 N.H. 365, 376, 616 A.2d 483 (1992).

The RFLP process and subsequent visual comparisons, however, are only the first and second steps in the DNA analysis used to identify criminal suspects or, as in the present case, to establish parentage. The third step — the population frequency calculation — requires the use of statistical analysis to determine the significance of the match. This third step was never reached in State v. Hammond, supra, 221 Conn. 264, because the DNA comparisons used in that case excluded the defendant as a possible source of the bodily fluids analyzed.

The Hammond court expressly noted that "[o]ther difficulties, such as . . . the use of population statistics to estimate the likelihood of a coincidental match" were not implicated in that case "because no match was found and the mismatch was sufficiently obvious that even a layperson could verify it." State v. Hammond, supra, 221 Conn. 280-81 n. 8.
The distinction between a case like Hammond, where visual analysis of the DNA patterns excludes the defendant, and a case like this one, where a match is observed, is crucial. "A declared match means that the samples could have come from the same individual. Conversely, if samples do not match, they must have come from different individuals." (Emphasis in original.) State v. Bible, supra, 175 Ariz. 581.

In order to understand the significance of the population frequency calculation used by Cellmark and other DNA laboratories, it is important to understand the significance of a "match." The DNA matches identified by Cellmark in the present case do not indicate that the blood in the defendant's trunk necessarily came from a child of the parents. Instead, the match indicates that the particular strands of DNA observed are consistent with that of a child of the parents. "Even if there is a perfect match, there is a possibility that the two samples came from different people whose DNA patterns at the targeted loci are indistinguishable. Thus the [third] procedure, population frequency calculation, generates a ratio which accompanies a match in order to express the statistical likelihood that an unrelated individual chosen at random from a particular population could have the same DNA profile as the suspect." State v. Vandebogart, 136 N.H. 365, 370, 616 A.2d 483 (1992); see State v. Bible, supra, 175 Ariz. 582. Accordingly, the trial court correctly noted that because "a match between two DNA bands means little without data on probabilities, the calculation of statistical probabilities is an integral part of the process, and the method used must meet the Frye criteria."

The state presented the following testimony with regard to the population frequency calculation. Cellmark uses separate databases containing DNA extracts from Caucasians, African-Americans and Hispanics. The Caucasian database used by Cellmark in this case consists of blood samples containing DNA from over 300 persons, taken from a Red Cross blood bank in Wilmington, Delaware. This database provides a means of assessing the significance of the fact that the three probes of the DNA samples from the automobile's trunk and the DNA samples of the parents matched. For each of the three alleles tested, the database is consulted to determine the probability that a match would have occurred coincidentally. The result is six separate probabilities — three representing the match with the mother's DNA, and three representing the match with the father's DNA.

Next, the three probabilities referenced above are multiplied together for each parent, using the "product rule" familiar to statisticians. The result of the product rule is an overall probability that a person selected randomly would coincidentally have these same three matches. This probability with respect to the victim's mother is that approximately one in 1400 unrelated persons would have the same alleles. The relevant probability with respect to the victim's father is approximately one in 26,000. Forman explained that these calculations are "not meant to be . . . precise . . . but simply [are] estimate[s] of how common or rare a particular event is likely to be in the population as a whole." Cotton testified that the population frequency calculation methodology is generally accepted in the scientific community.

The defendant presented the expert testimony of Laurence Mueller, Ph.D., of the University of California at Irvine. Mueller testified that the application of the product rule necessarily depends on an assumption that there is statistical independence between the DNA alleles examined, an assumption that requires the presence of a population genetics principle called linkage equilibrium. Linkage equilibrium in turn requires that the population referred to be genetically homogeneous, rather than consisting of population substructures, or genetic subgroups which would "most likely occur along racial and ethnic lines and present significant pockets of genetic variation within the larger population group." Commonwealth v. Lanigan, 413 Mass. 154, 160-61, 596 N.E.2d 311 (1992). Mueller testified further that he was not aware of any study that has demonstrated the existence of linkage equilibrium in the Caucasian population, and believed that the available evidence was to the contrary. Mueller also criticized the composition of the Cellmark database. These criticisms have been echoed in recent court decisions. See, e.g., State v. Bible, supra, 175 Ariz. 585-86; State v. Vandebogart, supra, 136 N.H. 380-81. Mueller expressed the opinion that these criticisms were shared by other scientists to the extent that there was "substantial disagreement" as to the validity of the principles underlying the population frequency calculation. Although the trial court correctly noted that the population frequency calculation is "calculated differently" in this case than in an identification case because this is a parentage case, Mueller's criticisms are nonetheless relevant to the calculation employed in the present case.

The court in State v. Bible, supra, 175 Ariz. 585-86, stated that "these probability calculations are flawed in three ways: (1) they are impermissibly based on the disputed assumption of linkage equilibrium; (2) the database relied on is of disputed statistical validity; and (3) the database relied on is not in Hardy-Weinberg equilibrium."

Courts have questioned the validity of a population genetics assumption called Hardy-Weinberg equilibrium that is implicated in the use of DNA evidence to identify suspects. See, e.g., State v. Bible, supra, 175 Ariz. 585-86. Mueller, however, conceded that because the DNA typing evidence was used in the present case to establish parentage rather than to identify a suspect, whether Cellmark's database is in Hardy-Weinberg equilibrium is irrelevant to this case.

At the time of the DNA hearing in this case, there was, as Kidd acknowledged, substantial debate among population geneticists regarding the scientific validity of the population frequency calculation admitted by the trial court. See State v. Bible, supra, 175 Ariz. 584-85; People v. Barney, 8 Cal.App.4th 798, 814-16, 10 Cal.Rptr.2d 731 (1992); State v. Vandebogart, supra, 136 N.H. 381. The controversy centered around two opposing articles published in the December, 1991 edition of Science, "a respected scientific journal with articles subject to peer review . . . [that] stat[ed] radically conflicting views of [the scientific merit of] statistical probability calculations." State v. Bible, supra, 584. These articles were entered into evidence before the trial court.

In the article that sparked the controversy, two highly respected scientists wrote that the theoretical assumptions underlying the population frequency calculation "are unjustified and generally unreliable." R. Lewontin D. Hartl, "Population Genetics in Forensic DNA Typing," 254 Science 1745, 1750 (Dec. 20, 1991). The opposing article was coauthored by one of the experts who testified in the present case, Kenneth Kidd. R. Chakraborty K. Kidd, "The Utility of DNA Typing in Forensic Work," 254 Science 1735 (Dec. 20, 1991).

Another crucial development in the scientific field, however, was not before the trial court. The National Academy of Sciences, which Mueller described as "the most prestigious collection of scientists in the country," assembled a committee of eminent scientists and jurists, including Henry Lee, to research and analyze the current viability of DNA typing evidence. The expert witnesses in the present case discussed this committee at the DNA hearing, but its report and findings had not yet been published and were not available to the experts. When the report was finally published in 1992, after the trial in the present case, it significantly changed the scientific landscape. See Committee on DNA Technology in Forensic Science, National Research Council, "DNA Technology in Forensic Science," (1992) (Committee report); see also State v. Bible, supra, 175 Ariz. 584-85 and 585 n. 9; People v. Barney, supra, 8 Cal.App.4th 816.

The Barney court expressly distinguished a prior decision of another district of the California Court of Appeals that had upheld DNA probability evidence; People v. Axell, 235 Cal.App.3d 836, 1 Cal.Rptr.2d 411 (1991); as follows: "Whatever the merits of the prior decisions on the statistical calculation process — including Axell — the debate that erupted in Science in December 1991 changes the scientific landscape considerably, and demonstrates indisputably that there is no general acceptance of the current process." People v. Barney, supra, 8 Cal.App.4th 820. Likewise, the Supreme Court of Arizona distinguished earlier cases that had upheld statistical probability evidence by stating that "recent developments — scientific and judicial — drastically alter the relevant analysis." State v. Bible, supra, 175 Ariz. 584.

First, the Committee report fully endorsed the DNA typing technology itself, even going so far as to recommend that courts take judicial notice of the scientific acceptability of the procedures used to extract and compare DNA alleles. Committee report, p. 133, cited in State v. Cauthron, supra, 120 Wash.2d 895-96.

Second, the Committee report acknowledged that "`[s]ubstantial controversy has arisen concerning the methods for estimating the population frequencies of specific DNA typing patterns. Questions have been raised about the adequacy of the population databases on which frequency estimates are based and about the role of racial and ethnic origin in frequency estimation.'" State v. Cauthron, supra, 120 Wash.2d 902 -903, quoting Committee report, pp. 74-75. The report, however, chose not to take a position in the debate, but instead "`assume[s] for the sake of discussion that population substructure may exist. . . .'" United States v. Porter, 618 A.2d 629, 637-38 (D.C.App. 1992), quoting Committee report, p. 80.

Third, the Committee report recommended that courts admit into evidence population frequency calculations, but it set out various recommended criteria for the admission of this evidence, including the reliance on conservative population frequency estimates, and the use of a ceiling principle, which is a method of estimating probabilities that attempts to account for population substructures. State v. Cauthron, supra, 120 Wash.2d 908, citing Committee report pp. 80, 82-83; see State v. Alt, 504 N.W.2d 38, 51 and n. 21 (Minn.App. 1993) (holding that "statistical frequencies of individual loci should be admitted in this case if calculated according to the [Committee] modified ceiling principle"; noting, however, that even the [Committee] report recommendations have become the object of scientific controversy).

Since the publication of the articles in Science and the release of the Committee report, some courts have rejected population frequency testimony as not generally accepted in the scientific community. See, e.g., State v. Bible, supra, 175 Ariz. 585-86; People v. Barney, supra, 8 Cal.App.4th 820; Commonwealth v. Lanigan, supra, 413 Mass. 162-63; see also People v. Wallace, 14 Cal.App.4th 651, 660, 17 Cal.Rptr.2d 721 (1993).

On the other hand, other courts have remanded appeals involving population frequency calculations for further consideration of the conclusions and recommendations of the Committee report. United States v. Porter, supra, 618 A.2d 642; State v. Vandebogart, supra, 136 N.H. 383 (remanding so trial court can "conduct a hearing in order to determine whether the [Committee's] recommended ceiling principle is a generally accepted technique"); see State v. Cauthron, supra, 120 Wash.2d 885, 909 (after oral argument, court requested additional briefing on Committee report; subsequently court remanded case to trial court to determine "if the empirical evidence utilized by Cellmark is valid under the criteria set forth by the Committee prior to allowing an expert to testify about the [DNA typing] results in Cauthron's case"). Still other courts allow testimony regarding population frequency calculations without qualification. See, e.g., State v. Montalbo, 73 Haw. 130, 140-41, 828 P.2d 1274 (1992); People v. Axell, 235 Cal.App.2d 836, 867-68, 1 Cal.Rptr. 411 (1991).

The District of Columbia Court of Appeals noted, however, that "[i]n a majority of cases in which . . . match probabilities have been admitted, the defense failed to present evidence of the controversy among scientists which has since been recognized in the [Committee report] . . . ." United States v. Porter, supra, 618 A.2d 639.

We determine that the most appropriate course to take, in accordance with the decision of the New Hampshire Supreme Court in State v. Vandebogart, supra, 136 N.H. 383, and other recent decisions, is to remand this issue for further consideration by the trial court. On the retrial, if this issue again becomes relevant, the trial court should consider the conclusions and recommendations of the Committee report and any other relevant evidence, including expert testimony, and determine whether the probability calculations sought to be introduced conform to the criteria set out in the Committee report, or, if not, whether the evidence nevertheless passes appropriate scientific evidence standards under the circumstances of this case. See State v. Cauthron, supra, 120 Wash.2d 908-909.


I agree with every part of the majority opinion except for its conclusion, in part I, that the state produced sufficient evidence to sustain the defendant's conviction of murder in the first degree in violation of General Statutes § 53a-54a. On that issue, I respectfully dissent.

I cannot and do not disagree with the basic principles that govern this appeal. "There is no distinction between circumstantial and direct evidence so far as probative force is concerned. State v. Smith, 212 Conn. 593, 599, 563 A.2d 671 (1989); State v. Uretek, Inc., 207 Conn. 706, 715, 543 A.2d 709 (1988); State v. Walker, 206 Conn. 300, 315-16, 537 A.2d 1021 (1988); State v. Magnano, 204 Conn. 259, 287, 528 A.2d 760 (1987); State v. Rodgers, 198 Conn. 53, 58, 502 A.2d 360 (1985); State v. D'Antuono, 186 Conn. 414, 421, 441 A.2d 846 (1982). Further, the evidence presented at trial must be construed in a fashion most favorable to sustaining the jury's verdict. State v. Mandrell, 199 Conn. 146, 154, 506 A.2d 100 (1986); State v. Haddad, 189 Conn. 383, 387, 456 A.2d 316 (1983); State v. Brice, 186 Conn. 449, 459, 442 A.2d 906 (1982); State v. Jackson, 176 Conn. 257, 262, 407 A.2d 948 (1978); State v. Brunori, 22 Conn. App. 431, 435, 578 A.2d 139, cert. denied, 216 Conn. 814, 580 A.2d 61 (1990). . . . [The trier's] findings of fact are entitled to great weight and . . . a conviction based on the facts found by the trier will be affirmed if the trier of fact `could reasonably have inferred [from the evidence] that the defendant was guilty beyond a reasonable doubt.' State v. Cobbs, 203 Conn. 4, 7, 522 A.2d 1229 (1987); State v. Mandrell, supra, 153-54; State v. Scielzo, 190 Conn. 191, 196, 460 A.2d 951 (1983); State v. D'Antuono, supra, 421; State v. Perez, 182 Conn. 603, 607, 438 A.2d 1149 (1981)." State v. Osman, 218 Conn. 432, 436, 589 A.2d 1227 (1991).

Our cases also teach us, however, that the trier's findings of fact are not conclusive "where the state's evidence is improbable and unconvincing and where all the facts found are insufficient to prove the guilt of the defendant beyond a reasonable doubt. State v. Cobbs, supra, [ 203 Conn. 4] 11, 13; State v. Mandrell, supra, [ 199 Conn. 146] 154; State v. Payne, 186 Conn. 179, 184, 440 A.2d 280 (1982); State v. Mayell, 163 Conn. 419, 427-28, 311 A.2d 60 (1972); State v. Kelsey, 160 Conn. 551, 553-54, 274 A.2d 151 (1970)." State v. Osman, supra, 218 Conn. 437. Because of the great deference that we afford to the factual findings of the trier, such cases are and should be rare. I am nonetheless persuaded that this case, like State v. Carpenter, 214 Conn. 77, 570 A.2d 203 (1990), on appeal after remand, 220 Conn. 169, 595 A.2d 881 (1991), cert. denied, 502 U.S. 1034, 112 S.Ct. 877, 116 L.Ed.2d 181 (1992), is one of those rare instances in which the reasonable inferences to be drawn from the record do not support a finding of guilt.

The issue of sufficiency of the evidence relates to whether the state established, beyond a reasonable doubt, that the defendant had the intent to kill the victim. We are all in agreement that the record discloses no evidence of motive, premeditation or preplanning on the defendant's part. The record is entirely barren of any facts that would indicate the nature of the precipitating events that led the defendant to kill the victim. The majority opinion hypothesizes certain scenarios consistent with the jury's finding of the defendant's intent to kill, all of which include the defendant's possession of a weapon at the crucial time in the family room.

The absence of any factual basis for the circumstances of the killing makes other scenarios, inconsistent with the defendant's intent to kill, equally as credible as those hypothesized by the majority opinion. It is quite possible that the defendant, not believing that the services that the victim was prepared to offer were limited to a massage, became enraged at her refusal to have intercourse, and that their verbal sparring escalated to a physical confrontation. In the course of that escalation, if the defendant displayed a weapon to frighten the victim into submitting to intercourse, the victim may have, to the defendant's surprise, forcibly resisted, causing the defendant to panic and lash out, recklessly or negligently, without consciously intending to cause the victim's death. Alternatively, the victim herself, having previously encountered clients who misunderstood her proffered services, may have carried a knife with which she threatened the defendant when he sought to force himself on her. The defendant's efforts to disarm the victim may have led to her injury and to the same ultimate result. The evidence produced at trial does not enable us to determine which, if any, of these scenarios accurately depicts what transpired between the defendant and the victim.

Events that transpired after the killing do not furnish a factual predicate from which inferences may properly be drawn about the defendant's intent at the time of the killing. The majority opinion attaches significance to the defendant's failure to summon medical assistance for the victim. That failure would be probative if there were evidence that the victim's injury might have responded to prompt medical intervention. See State v. Francis, 228 Conn. 118, 129, 635 A.2d 762 (1993); State v. Greenfield, 228 Conn. 62, 78, 634 A.2d 879 (1993) (victim found alive and died at the hospital); see also State v. Carpenter, supra, 214 Conn. 83-84. In this case, however, there was no evidence that the injuries inflicted did not cause immediate death. Similarly, we have never held that evidence of consciousness of guilt independently supports a reasonable inference about the defendant's mental state when he caused the death of the victim, and other courts have rejected such inferences as untenable. See State v. James, 819 P.2d 781 (Utah 1991); Stafford v. People, 154 Colo. 113, 388 P.2d 774 (1964). To the contrary, we have declared it impermissible for a jury to infer intent to commit murder from the fact of the death of the victim, even from the fact of the victim's death at the hands of the defendant. State v. Crafts, 226 Conn. 237, 248, 627 A.2d 877 (1993).

In sum, the defendant's conviction of murder in the first degree can only be sustained by a series of speculations about the manner and circumstances in which the defendant brought about the victim's death. Neither singularly nor cumulatively can speculative evidence suffice to support an inference or series of inferences from which an element of a crime can be found to have been proven beyond a reasonable doubt.

Concededly, no clear line of demarcation exists between a permissible inference and an impermissible speculation. Nonetheless, on this record, I am persuaded that this case, in which the state had to prove not only that the defendant had killed the victim but that he had done so with intent to kill, falls on the side of impermissible speculation. When there are no ascertainable facts from which the appropriate inferences may reasonably be drawn, the presumption of innocence requires a conviction to be set aside. See State v. Skipper, 228 Conn. 610, 621-22, 637 A.2d 1101 (1994); State v. Hammond, 221 Conn. 264, 287-88, 604 A.2d 793 (1992).

It is not difficult to reconstruct what probably happened in the jury room in this case. The jury's decision to convict the defendant of this crime was very likely the result of the unenviable choice that the jury faced because it was not given a proper charge on lesser included offenses. We are all in agreement that the state produced substantial evidence that the defendant had been materially involved, in some fashion, in the victim's death. The jury could either speculate that the defendant had entertained the intent to kill the victim, or it could permit the defendant to walk away free. In these circumstances, it is entirely understandable that the jury chose to return a verdict of guilty of first degree murder.

I recognize that, if the defendant is acquitted of the only crime with which he was charged, a retrial on lesser included offenses may encounter constitutional difficulties. I would not, however, anticipate at this juncture whether the principles of double jeopardy would bar such a retrial.

Accordingly, I respectfully dissent.


I agree with the dissenting opinion of the Chief Justice that the evidence was insufficient to establish, beyond a reasonable doubt, that the defendant intended to kill the victim. I write separately to point out that the majority's analysis relies on pure speculation.

In addition to the conjecture in the majority opinion pointed out in the dissent of the Chief Justice, the majority relies on the claim that because the injury occurred in a family room, the jury could reasonably have inferred that the defendant had prior possession of a weapon or had to go to another part of the house to get the weapon. There is no basis for this speculation. As the majority indicates, there was no evidence of what type of instrument or weapon was used to cause the victim's death except that the amount of blood found would support the permissible inference that the instrument used cut very deeply into the body or cut into a blood vessel. Many household objects that can be found in family rooms, such as scissors, could cause the injury described in the majority opinion. In fact, the state introduced into evidence a photograph of the family room, which indicates that there was a fireplace in that room, several feet away from the blood stain on the rug, and that there are various metal fireplace tools on a rack next to the fireplace. These sharp pointed instruments, which can be found in many family rooms, could cause penetrating injuries similar to those described in the majority opinion.

If the majority's conjecture — the location of a weapon, failure to summon medical assistance and consciousness of guilt — is stripped from the circumstantial equation, there is simply insufficient evidence to support the jury's conclusion. The remaining evidence relied on by the majority consists of the large quantity of blood found and the inferences that could be drawn regarding the manner in which this bleeding was caused. This evidence can permissibly support the inference that the defendant caused the victim's death, but is insufficient to prove that he had the conscious objective to cause her death. We have held that " intent cannot be inferred directly from results. The fact that a victim was struck by a bullet would not, in itself, support the inference that the perpetrator intended to kill the victim, because the perpetrator might have acted with a variety of mental states. . . . In this case, similarly, the jury could not properly have inferred an intent to commit murder from the mere fact of the death of the victim, even from her death at the hands of the defendant." (Emphasis added.) State v. Crafts, 226 Conn. 237, 248, 627 A.2d 877 (1993). Speculative evidence, no matter how cumulative, cannot support an inference or series of inferences upon which an element of a crime can be proven beyond a reasonable doubt.

I respectfully dissent.