State v. Grant

Connecticut, like other jurisdictions, has long recognized that a defendant may introduce evidence which indicates that a third party, and not the defendant, committed the crime with which he is charged. State v.Perelli, 125 Conn. 321, 328, 5 A.2d 705 (1939). The freedom of defendants to offer such evidence is, however, limited by well established judicial doctrine. Under our law, the defendant "must show some evidence which directly connects a third party to the crime with which the defendant is charged." (Internal quotation marks omitted.) State v. Ortiz,252 Conn. 533, 564, 747 A.2d 487 (2000). The court is now called upon to apply this doctrine in a twenty-nine year old murder case that, over the course of its long and eventful life, has been graced with a multitude of suspects.

On July 16, 1973, the body of Concetta ("Penney") Serra was found in the stairway of the Temple Street parking garage in New Haven. The medical examiner determined that she died as the result of a stab wound to the heart. No one was apprehended at the scene. In 1984, a man named Anthony Golino was arrested for the crime, but the charges against him were dismissed in 1987 when a court-ordered blood test revealed that his blood type did not match that of the killer. Golino v. New Haven,950 F.2d 864, 866 (2d Cir. 1991), cert. denied, 505 U.S. 1221,112 S. Ct. 3032, 120 L. Ed. 2d 902 (1992). Much later, in 1999, Edward Grant, the defendant herein, was arrested, largely on the basis of the state's allegation that blood thought to be that of the perpetrator is consistent with a DNA profile of a sample of Grant's blood. Grant's case is now about to be tried, and he understandably wishes to point to the existence of suspects other than himself. No one, as far as the record indicates, has ever thought that the Serra murder was the work of more than one person. Evidence that Serra was killed by someone other than Grant would necessarily tend to exculpate Grant.

On October 30, 2001, the state filed the motion in limine now before the court. The motion seeks "to preclude the defendant from offering evidence that a third party other than the defendant may have committed the murder charged in this case."

On January 22, 2002, Grant filed a written offer of proof regarding evidence of third party culpability. Grant specifically seeks to submit evidence that will, in his opinion, connect two different suspects to the commission of this offense. The two suspects identified by Grant are Philip DeLieto and Selman Topciu. Grant does not seek to identify Golino as the author of the crime in question.

On January 23, 2002, the state informed the court that it did not claim its motion in limine with respect to DeLieto. The state does, however, claim its motion with respect to Topciu. The motion was argued on February 11, 2002.

The record establishes that Topciu has been the subject of three warrant applications in connection with the Serra murder case. On June 16, 1992, a Texas magistrate signed a warrant for the search and seizure of a sample of Topciu's blood. The search warrant was executed later that day. On July 5, 1994, the state submitted to the Connecticut Superior Court an application for an arrest warrant charging Topciu with the Serra murder (the first application). On July 13, 1994, Ronan, J., "Reviewed denied" the first application. On June 30, 1995, the state tried again, submitting another application (the second application) for Topciu's arrest to the Superior Court. On July 7, 1995, Hartmere, J., found "No Probable Cause." Both the first and second applications were signed by a prosecutor prior to submission. (The first application was signed by an assistant state's attorney; the second application was signed by a deputy chief state's attorney.) Grant now wishes to use the information contained in the second application in his defense. He does not seek to offer either the first or second applications themselves.

This is an unusual case. On two separate occasions, the state itself has formally submitted applications asserting that probable cause exists to arrest Topciu for the Serra murder. Each application was signed (and presumably reviewed and approved) by an experienced prosecutor. On each occasion, however, the court officially found that there was no probable cause. How should this history affect the court's decision with respect to third party culpability evidence?

The controlling standard is well established: "[A] defendant may introduce evidence which indicates that a third party, and not the defendant, committed the crime with which the defendant is charged. . . . The defendant, however, must show some evidence which directly connects a third party to the crime with which the defendant is charged. . . . It is not enough to show that another had the motive to commit the crime . . . nor is it enough to raise a bare suspicion that some other person may have committed the crime of which the defendant is accused. . . . The admissibility of evidence of third party culpability is governed by the rules relating to relevancy. . . . No precise and universal test of relevancy is furnished by the law, and the question must be determined in each case according to the teachings of reason and judicial experience. . . . The trial court has wide discretion in its rulings on evidence. . . ." (Citations omitted; internal quotation marks omitted.) State v.Ortiz, supra, 252 Conn. 564.

Our courts have explained the phrase "directly connects" largely by contrast with what it does not mean. Ortiz tells us that evidence showing "that another had the motive to commit the crime" and "a bare suspicion that some other person may have committed the crime" do not amount to a "direct connection." (Internal quotation marks omitted.) Id. The case law, however, does not tell us what a "direct connection" actually is.

The phrase "directly connects" cannot sensibly refer to the distinction between direct and circumstantial evidence. "There is, of course, no legal distinction between direct and circumstantial evidence as far as probative force is concerned." (Internal quotation marks omitted.) Statev. Foster, 202 Conn. 520, 536, 522 A.2d 277 (1987). "Courts must necessarily rely on circumstantial evidence in many cases . . . ."Shaughnessy v. Morrison, 116 Conn. 661, 664, 165 A. 553 (1933). It has long been a legal commonplace that circumstantial evidence is sometimesmore probative than direct evidence. "A fact positively sworn to by a single eyewitness of blemished character, is not so satisfactorily proved, as is a fact which is the necessary consequence of a chain of other facts sworn to by many witnesses of undoubted credibility."Commonwealth v. Harman, 4 Pa. 269, 272 (1846). Just as the law allows a criminal defendant to be convicted by circumstantial evidence, it cannot prevent the same defendant from establishing a third party's "direct connection" to a crime by a similar use of evidence. Reason and experience do not support the proposition that "direct connection" evidence "occupies a special or exotic category of proof." People v.Primo, 96 N.Y.2d 351, 356, 753 N.E.2d 164, 728 N.Y.S.2d 735 (2001).

Ortiz provides a valuable hint of what a "direct connection" is by explaining that, "The admissibility of evidence of third party culpability is governed by the rules relating to relevancy." (Internal quotation marks omitted.) State v. Ortiz, supra, 252 Conn. 564. The New York Court of Appeals has recently explained that the phrase "directly connects" "merely reinforce[s] the notion that remote evidence of a third party's culpability — though relevant — will not be sufficiently probative to outweigh the risk of trial delay, undue prejudice or jury confusion."People v. Primo, supra, 96 N.Y.2d 356. Connecticut Code of Evidence § 4-3 provides that "Relevant evidence may be excluded if its probative value is outweighed by the danger of unfair prejudice or surprise, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time or needless presentation of cumulative evidence." The commentary to § 4-3 explains that "Section 4-3 establishes a balancing test under which the probative value of proffered evidence is weighed against the harm likely to result from its admission." The court's task here is to strike this balance.

In striking this balance, the court is aware that two judges have, on separate occasions, found that there is no probable cause to arrest Topciu for the Serra murder. The question before me, however, is not the question presented to those earlier judges. The question before Judges Ronan and Hartmere was whether there was a fair probability that Topciu committed the crime. State v. Velasco, 248 Conn. 183, 191, 728 A.2d 493 (1999). The purpose of third party culpability evidence, on the other hand, "is not to prove the guilt of the other person, but to generate a reasonable doubt of the guilt of the defendant." State v. Hawkins,260 N.W.2d 150, 158-59 (Minn. 1977); see also Smithart v. State,988 P.2d 583, 588 (Alaska 1999), and authorities cited therein. Connecticut juries are routinely instructed that a "reasonable doubt" is "a real doubt, an honest doubt, a doubt which has its foundation in the evidence or lack of evidence." (Internal quotation marks omitted.) Statev. Lemoine, 256 Conn. 193, 204, 770 A.2d 491 (2001). There is no legal requirement that third party culpability evidence must rise to the level of probable cause in order to generate reasonable doubt. Evidence implicating a third person in a crime can consequently fall short of establishing probable cause of the guilt of that person and, nonetheless, establish a reasonable doubt as to the guilt of the accused. The court's task here is to ascertain the probative value of the proffered evidence on the question of its ability to generate a reasonable doubt of the guilt of the accused and weigh that value against the harm likely to result from the admission of the evidence in question.

Grant has submitted an extensive offer of proof. His offer reflects many of the factual submissions made in the second application, although it differs from the second application in some details. The more forceful points of Grant's offer are as follows.

On July 16, 1973, after Serra's murder on that day, a 1972 Buick registered to her father, John Serra, was discovered on Level #8 of the Temple Street garage. Human blood, thought to be that of the killer, was found in the car. (Serra's blood was type A; the blood in the car was type 0.) Two stamped, sealed, unmailed envelopes were found over the driver's side visor. Both were addressed to "Mr. S. Topcier" of "10 Knoxs Street, West Haven, Conn." The return address of Eugene Horwitch, D.M.D., a New Haven dentist, was on both envelopes. John Serra had driven the car earlier that morning and had not seen any envelopes on the visor.

Prior to June 28, 1973, Serra had worked as a dental assistant for Dr. Horwitch. Topciu, who had used the name "Topcier" in the past, was a patient of Dr. Horwitch. He lived at 10 Knox Street. He worked as a cook at Murray's Restaurant, approximately one-half block from Dr. Horwitch's office. Serra went to Murray's every day to pick up Dr. Horwitch's lunch. Serra and Topciu almost certainly recognized each other from these connections.

Grant additionally relies on Topciu's asserted resemblance to certain eyewitness descriptions of the killer. Two witnesses, Timothy Woodstock and Frederick Petzold III, saw a female being chased by a male in the Temple Street garage at about the time of the homicide and later saw the same man return to the Buick with a shiny object in his right hand. Woodstock described the killer as a white male, 5 feet 8 inches tall, 150 to 160 pounds, slim build, fair complexion, medium length black hair. Petzold described the man as a white male, dark hair, dark complexion, approximately 5 feet 10 inches tall. A third witness, Chris Fagan, was a parking attendant who took a bloody parking ticket from a man who was almost certainly the killer. (The bloody ticket is in evidence.) Fagan described the man as being "perhaps a Puerto Rican male or a white male [of] Italian [descent], thin face, color of his skin, that being the face was sort of an olive color, with a little whitish in same, black hair, medium cut, straight, and . . . approximately . . . 21 yrs. of age." Fagan also stated that the man did not speak "fluently, as american english." Fagan additionally observed that the man used only his right hand. This last observation is consistent with blood trail evidence in the garage, indicating that the killer bled profusely from the left side of his body.

Topciu was born in Albania in 1940. There is evidence that he had a foreign accent in 1973. A 1977 department of correction record describes him as a white male, 5 feet 11 inches tall, 150 pounds, brown hair. Photographs of his left arm in 1993 arguably show a scar in his wrist area, possibly produced by a cut that resulted in substantial bleeding. (It would, of course, be the rare cook who did not, on some occasion, suffer a serious cut).

Topciu's blood, a sample of which was taken as the result of the Texas search warrant, is type 0. He additionally shares the same DQ Alpha genetic marker as the killer. These combined findings — type 0 blood and the same DQ Alpha genetic marker — occur in approximately 5 percent of the Caucasian population of the United States.

In January 2002, long after the first and second applications (and, for that matter, after the filing of the motion in limine now before the court), the state did additional testing of Topciu's blood using the Short Tandem Repeat (STR) method of amplification, a form of DNA testing not yet developed at the time of the first and second applications. A state forensic report now in evidence eliminates Topciu as a contributor of blood found on a handkerchief at the scene. (The blood on the handkerchief is also thought to be that of the killer.)

This recent development gives the court considerable pause. The state's case against Grant is largely built on the same STR technique that has now, we are told, eliminated Topciu. If the jury ultimately finds that the STR technique is reliable and probative, Grant may be convicted regardless of the evidence he summons against Topciu. The very technique that points to Grant now eliminates Topciu as a suspect. It is, however, "the function of the trier of fact to determine the credibility of witnesses." State v. Gold, 180 Conn. 619, 647, 431 A.2d 501, cert. denied, 449 U.S. 920, 101 S. Ct. 320, 66 L. Ed. 2d 148 (1980). This is a function that the court cannot usurp.

The question under our law is whether Grant has provided evidence that "directly connects" Topciu with the Serra murder. If he has, that evidence is admissible regardless of the fact that other evidence, tending to exculpate Topciu, will then become admissible as well. Id. The state plainly recognizes this requirement. It (appropriately) does not oppose third party culpability evidence against DeLieto in spite of the fact that the state's testing of a sample of DeLieto's blood has eliminated him as a suspect as well. The court's limited function at this stage is to review Grant's offer of proof and determine whether the proffered evidence establishes the requisite "direct connection" if the trier of fact should choose to believe it. For this reason, other evidence tending to explain away the evidence summoned against Topciu, such as the fact that Serra sometimes brought patient bills from Dr. Horwitch home with her, cannot be considered by the court at this stage, although such explanatory evidence may ultimately play an important role in the jury's deliberations.

Grant's third party culpability evidence summoned against Topciu is relevant on the issue of reasonable doubt. Under § 4-1 of the Connecticut Code of Evidence, "relevant evidence" is any "evidence having any tendency to make the existence of any fact that is material to the determination of the proceeding more probable or less probable than it would be without the evidence." While the evidence in question fails to establish probable cause of Topciu's guilt, it is, nevertheless, probative on the issue of whether there is reasonable doubt of Grant's guilt. The real question is not whether the evidence is relevant, but whether it should be excluded under § 4-3 of the Connecticut Code of Evidence on the theory that "its probative value is outweighed by the danger of unfair prejudice or surprise, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time or needless presentation of cumulative evidence."

On the facts presented here, the court cannot exclude the evidence in question. Some factors enumerated in § 4-3 are plainly inapplicable here. There is no danger of "surprise," since the third party culpability issue is being determined well in advance of trial. The evidence is hardly "cumulative," since the state seeks to exclude all third party culpability evidence relating to Topciu. The remaining § 4-3 factors — "unfair prejudice," misleading the jury," "undue delay," and "waste of time" — merit consideration, but their force largely depends on the weight given to the DNA evidence that the state has recently developed against Grant as opposed to the earlier, largely circumstantial evidence that the state itself developed against Topciu. This is a matter for the jury, not the court, to decide.

The state plainly considered Topciu to be its prime suspect for a substantial period of time. It twice submitted applications seeking Topciu's arrest for the Serra murder. These applications were signed by experienced prosecutors. Although, as Judges Ronan and Hartmere duly held, there was no probable cause for Topciu's arrest, the evidence in question is nevertheless admissible to show reasonable doubt as to the guilt of Grant. Under these circumstances, the requisite "direct connection" has been established.

The motion in limine is denied.