916 F2d 41 United States v. L Dove
916 F.2d 41
UNITED STATES of America, Appellee,
Michael L. DOVE, Defendant-Appellant.
No. 1291, Docket 89-1607.
United States Court of Appeals,
Argued June 5, 1990.
Decided Oct. 2, 1990.
Gary D. Weinberger, Asst. Federal Public Defender for the D. of Conn. (Thomas G. Dennis, Federal Public Defender for the Dist. of Conn., of counsel), Hartford, Conn., for defendant-appellant.
Joseph C. Hutchinson, Asst. U.S. Atty., D. of Conn. (Stanley A. Twardy, Jr., U.S. Atty., D. of Conn., of counsel), New Haven, Conn., for appellee.
Before FEINBERG, NEWMAN and CARDAMONE, Circuit Judges.
CARDAMONE, Circuit Judge:
This appeal from a three-day jury trial that resulted in a judgment of conviction for bank robbery presents a question of defendant's identity. That issue in this case is inextricably fused with defendant's contention that he is innocent. A number of the circumstances revealed in the record, to be detailed shortly, are quite out of the ordinary. Two warrant immediate mention: the first is that defendant's fingerprints that might be expected to have been found at the scene were not; the second is the failure of the government's key eyewitnesses to make an in-court identification of defendant.
Michael Dove appeals from a judgment of the United States District Court for the District of Connecticut (Daly, J.), convicting him of bank robbery in violation of 18 U.S.C. Sec. 2113(a). He was sentenced on October 3, 1989 to 45 months (three years, nine months) of incarceration, three years of supervised release, and a $50 assessment. On appeal, he challenges the district court's failure to give requested jury instructions, and contends that the court's instructions were unbalanced and failed adequately to inform the jury regarding how to assess the circumstantial evidence of innocence.
The unfolding of the unusual circumstances at trial required that the jury instructions point to the evidence indicating innocence as well as guilt, and that the jury be informed how to assess such evidence. Because in this case we think the defense theory was not adequately included, and that as a result a balanced charge was not given, we vacate the judgment of conviction and remand this case to the district court for a new trial.
A. The Robbery
At 4:50 p.m. on October 7, 1988 a lone robber displaying a handgun entered a branch of the Connecticut Bank & Trust Company in South Meriden, Connecticut. After ordering everyone in the room to lie down on the floor, he jumped over the teller counter, grabbed $1,000 off a countertop, tried unsuccessfully to open a cash drawer, and then fled back over the counter and out the door. Bank surveillance photographs show that the entire incident took no more than a minute, and all pictures of the perpetrator were from the back, except for one blurred shot from the side. Those witnesses who had an opportunity to observe the robber described him to the police and agents of the Federal Bureau of Investigation (FBI) as a white male, five foot eight inches in height, weighing about 150 pounds, with shoulder length hair, dark eyes, and several days growth of beard. All of the witnesses agreed that he was wearing blue jeans, tan work boots, and a ripped grey sweat shirt. Based on these observations the police prepared a composite drawing of the suspect.
B. The Eyewitness Identification Evidence
The only contested issue at trial was the bank robber's identity, the government asserting that the defendant was the party who committed the crime, and Dove asserting his innocence. The evidence linking defendant to the crime came from a curious course of events. After the robbery, the police searched the neighborhood around the bank but were unable to find a suspect.
An account of the crime was published the next day in a local newspaper describing the suspect generally as a white male in his early twenties with blonde hair. A 14-year-old girl named Teresa Botta read the newspaper account and called the police. She asked the desk sergeant if the robber had been caught yet, and when told that he had not, Botta asked for a more detailed description of him. At trial, she testified that the officer described the robber to her as a male in his early twenties, with light blue eyes and blonde hair, weighing between 150 and 170 pounds. This testimony is interesting because the police composite--from which the sergeant was presumably reading--listed the eye color of the suspect as "hazel," which is light brown, not blue. Based on this description alone, Botta testified, she told the officer the robber might be her neighbor, Michael Dove. Later, she identified Dove in a police photo used at trial by the sweatshirt he was wearing.
The desk sergeant informed Meriden Police Detective William Staton of the phone call. Detective Staton then obtained a photograph of Michael Dove along with the photographs of seven other males who, in his view, resembled Dove. From these he assembled a photo lineup, although this photo array was not based on the description given the police by the witnesses, nor on the police composite created from those descriptions. Several days after the robbery, Detective Staton showed the photographs to the two female bank employees who had the best opportunity to observe the perpetrator during the robbery. The detective told both witnesses beforehand that the police had a suspect whose picture was included in the array.
One of the two, Deborah Petroske, had observed the perpetrator twice on the day of the robbery. The first time was when he entered the bank earlier that day. Ms. Petroske stated that the same individual who later robbed the bank came into the bank for about ten seconds fidgeting with some papers, swore audibly, and hurried out. She said she remembered this individual because of his peculiar behavior, and further testified that she recognized him when he came in later and ordered everyone to lie down on the floor. Petroske was only able to observe the robber during a portion of the minute he was actually in the bank because she complied with the robber's request to lie down on the floor, which partially obstructed her view. When Ms. Petroske was shown the photo lineup, she observed that six of the eight individuals pictured were too young to be the robber, and identified Michael Dove--whose picture was one of two remaining--as the robber.
The second witness, Laurie Dinuzzo, was actually brushed by the robber when he approached her after vaulting the bank counter, she noted that he had a gun with black and silver on it. She described him to the police as a white male, 25 years of age with blonde hair, weighing 150 pounds, and wearing blue jeans, tan work boots, a plaid shirt, and a grey sweat shirt. She explained that he had a pock-marked face, and crooked, yellow teeth. These facial characteristics do not fit defendant. After being shown two sets of photos and told the suspect was among them, she tentatively identified Dove as the robber.
Despite the photographic identification, neither Petroske or Dinuzzo were able to identify Michael Dove at trial. When asked by the prosecutor whether the individual who robbed the bank was in the courtroom, both responded that he was not. Dinuzzo in fact pointed to a member of the jury as the individual who looked most like the robber. In a further strange twist, it was shown at trial that about three weeks following the robbery--and after her original photo identification of Dove--Petroske was working in the bank's drive-in teller window when an individual she thought was the robber tried to cash a check at her window. She stated to Dinuzzo--who also happened to be working that day--that the man she saw was either the robber or "his twin." Dinuzzo alerted her superior at the bank who in turn phoned the FBI. The man's name was Ronald Wujcik. He was not charged with the robbery, but was called by the defense as a witness at trial.
C. The Fingerprint and Footprint Evidence
Bank surveillance photos showed, and eyewitness testimony confirmed, that the robber had not worn gloves, and had placed his bare hands on a number of surfaces within the bank, including the counter surfaces and a pane of glass directly above the push-bar on the door that the perpetrator of the crime had used to exit the bank. Immediately after the robbery, police and the FBI lifted several sets of latent palm and fingerprints from these areas. At trial, the government fingerprint expert stated that none of the palm or fingerprints they had obtained matched Dove's palm or fingerprints. Further, the government was able to lift bootprints from areas where the robber had stepped, including the counter, and none of these prints matched the pair of tan work boots police recovered from Dove's residence.
D. The Confession
Special FBI Agent Michael Cizmar testified that he arrested Michael Dove on November 11, 1988, a little over a month after the robbery, in Waterbury, Connecticut and that he and several agents brought defendant to Hartford, Connecticut for interrogation. At first Dove denied involvement in the robbery. After Dove told them he drank heavily and occasionally suffered alcoholic blackouts, the agents questioned him regarding his drinking. He admitted to robbing a bank in Meriden, Connecticut, but stated he could not remember the robbery of the Meriden bank because he suffered from alcoholic blackouts. According to Cizmar, Dove admitted he purchased a chrome-plated revolver to commit the crime, and that he robbed the bank to get back at the South Meriden Police Department.
Based on Dove's account, Cizmar prepared a handwritten statement for Dove to sign. The statement did not include Dove's admissions that he purchased a revolver and robbed a bank to get back at the local police in South Meriden. It merely stated instead that Dove had viewed the surveillance photos, and that he (Dove) did "believe" the subject in those photos was him because he recognized a pair of "stone washed" jeans and a grey sweatshirt that he owned. Dove's admission that he recognized the stone washed jeans in the pictures as his own, oddly enough, was flatly contradicted by direct accounts of several witnesses in the bank at the time of the robbery who stated that the robber was wearing blue jeans.
Dove contends that the district court gave unbalanced instructions on the crucial issues of: (1) the proper manner to assess and weigh Petroske and Dinuzzo's inability to identify him in open court, and (2) the proper manner to assess and weigh the direct and circumstantial evidence in the case. In effect, the sole theory of the defense was that the police had the wrong man and that Dove was not the culprit.
The defendant first objected to the court's charge regarding the witnesses' failure to identify the defendant in the courtroom. The court had charged the jury
The government has the burden of proving his identity as the perpetrator beyond a reasonable doubt. In this connection, it is not essential that a witness be able to identify a defendant in open Court or be free from doubt as to the correctness of her identification of the defendant by other means. However, if you are not convinced beyond a reasonable doubt that the defendant was the person who committed the crime, you must find him not guilty.
Defense counsel objected that while this charge was correct insofar as it went, it was nevertheless "unbalanced." He asked the trial court to add a sentence stating that the jury was "free to consider and weigh the effects of a witness' failure to identify a defendant in Court in the course of weighing the evidence." The trial judge declined stating "I told them they can weigh the evidence of or lack of it. The charge will stand."
Next, defense counsel asked the district court to describe the instruction it intended to give regarding direct and circumstantial evidence, to which the court responded, "I give the one, depending on my mood at the time, whether Jack shot Mary or Mary shot Jack." The charge was as follows:
Now, to illustrate the difference between direct and circumstantial evidence, let us assume that the fact in issue in a case is whether Jack shot and killed Mary. If a witness testified that he personally saw Jack shoot Mary, then we would say we have direct evidence of that fact. On the other hand, if a witness testifies that an hour before Mary was shot he sold Jack the pistol which has been identified as the murder weapon, and it was found in Jack's possession shortly after the murder, we would say we have circumstantial evidence of the fact that Jack did shoot Mary. That, as I say, is a very simple illustration and has no direct bearing on this case at all, but is illustrative of what I mean by circumstantial evidence.
Again defense counsel objected asserting that it was "unbalanced" because the charge did not reflect that evidence can point toward innocence as well as guilt. Nonetheless, the trial judge gave the just-recited hypothetical at an early point in the charge.
In order to succeed when challenging jury instructions appellant has the burden of showing that the requested charge "accurately represented the law in every respect and that, viewing as a whole the charge actually given, he was prejudiced." United States v. Ouimette, 798 F.2d 47, 49 (2d Cir.1986), cert. denied, 488 U.S. 863, 109 S.Ct. 163, 102 L.Ed.2d 134 (1988). See also United States v. Lam Lek Chong, 544 F.2d 58, 68 (2d Cir.1976), cert. denied, 429 U.S. 1101, 97 S.Ct. 1124, 51 L.Ed.2d 550 (1977). Whether jury instructions were properly given is a question of law that this court reviews de novo and, when assessing prejudice to the defendant, we appraise the significance of an error in instructions to the jury by comparing the instructions actually given with those that should have been given. See Henderson v. Kibbe, 431 U.S. 145, 154, 97 S.Ct. 1730, 1736, 52 L.Ed.2d 203 (1977). On the close facts of this case, we believe that the defendant has met his burden in demonstrating prejudice.
The trial court's instructions on in-court identification were unbalanced. It was possible for the jury to conclude, based on the instructions given, that it need not attach any real significance to the fact that the government's two key eyewitnesses could not identify the defendant in open court. In this respect, the charge was unbalanced because it instructed the jury as to how the witnesses' inability to identify the defendant in the courtroom might bear on guilt without indicating how this rather significant evidence might bear on innocence. See United States v. Gleason, 616 F.2d 2, 15 (2d Cir.1979) ("Unquestionably, it is the court's duty, in instructing a jury on the subject of witnesses' credibility, to give balanced instructions."), cert. denied, 444 U.S. 1082, 100 S.Ct. 1037, 62 L.Ed.2d 767 (1980).
The imbalance could readily have been cured by the addition of defendant's one-sentence requested instruction advising the jury that they were, in fact, free to consider and weigh the effect of the witness' failure to identify the defendant in open court. Given the crucial nature of the eyewitnesses' identification, we are not convinced that simply stating that the jury must be convinced beyond a reasonable doubt that the defendant was the person who committed the crime cured whatever prejudice defendant may have suffered.
This issue takes added emphasis from Justice Harlan's insightful comments, in Simmons v. United States, 390 U.S. 377, 88 S.Ct. 967, 19 L.Ed.2d 1247 (1968), pertinent because both eyewitnesses were told the police suspect's picture was in the array.
The chance of misidentification is also heightened if the police indicate to the witness that they have other evidence that one of the persons pictured committed the crime. Regardless of how the initial misidentification comes about, the witness thereafter is apt to retain in his memory the image of the photograph rather than of the person actually seen, reducing the trustworthiness of subsequent lineup or courtroom identification.
Id. at 383-84, 88 S.Ct. at 971. Here, despite the suggestive presentation of the photo array to the two eyewitnesses, neither of them identified Dove in court as the robber. The dubious and suspect photo identification coupled with the failure of both eyewitnesses to make an in-court identification of the defendant should have alerted the trial court to the necessity for a carefully balanced charge on identification, one pointing to guilt and the other to innocence. Its failure in this regard cannot be deemed harmless.
Further, the district court's instructions on the difference between direct and circumstantial evidence were unbalanced as well. The government proposed and the defense joined in urging the court to use a neutral hypothetical to illustrate the difference between direct and circumstantial evidence. Counsels' proposed instruction illustrates how from observing a person entering a room with a dripping wet umbrella one can reasonably determine, based on this sort of circumstantial evidence, that it is raining outside. The trial court refused to give this neutral instruction using instead its hypothetical of "whether Jack shot Mary."
The problem with the "whether Jack shot Mary" example is that Jack's guilt is assumed in the hypothetical's premise and the jury is merely instructed how to look for evidence of that guilt. As the trial court said itself about the hypothetical "it only points towards how guilt is proved." Although the example the trial court used was not analogous to the facts of this case, the use in a criminal case of a hypothetical that assumes guilt where defendant asserts his innocence is disfavored. Certainly the use of such an example was more prejudicial than helpful and would tend to skew the jury away from the truth rather than toward it. See United States v. Gaggi, 811 F.2d 47, 62 (2d Cir.), cert. denied, 482 U.S. 929, 107 S.Ct. 3214, 96 L.Ed.2d 701 (1987). The use of a neutral hypothetical is generally advisable, and was especially warranted in the unusual circumstances of this case. See, e.g., L. Sand, J. Siffert, W. Loughlin & S. Reiss, Modern Federal Jury Instructions, at p 5.01 (1987) (Instruction 5-2).
Again, the hypothetical that assumed Jack's guilt may have led the jury to discount the circumstantial evidence supporting the defense theory that Dove was innocent. Virtually all of the circumstantial evidence pointed towards the possibility of innocence. The palm and fingerprints were determined, for example, not to be defendant Dove's; similarly, none of the bootprints at the scene matched boots seized from his residence.
Quite plainly were Dove's fingerprints found at the scene or were his boots to have matched those prints lifted from the bank's counter, it would have been direct evidence that he was present in the bank. Where defendant's prints are not found, the converse obviously is not proved. That is, their absence does not conclusively show that defendant was not present in the bank, because there are many other explanations such as the presence of many other prints, or the prints were smeared, or imperfect police lifting procedures, or because Dove got rid of the incriminating boots.
Defendant argues that there is a probable inference that he was not present when a number of these explanations are purportedly negated. "An inference is probable ... if it is one of a class of arguments such that the conclusions are true with a certain relative frequency when the premises are true," see M.R. Cohen & E. Nagel, An Introduction to Logic and Scientific Method, ch. VIII, at 151, 154 (1934) (emphasis in original). To put it in other words, the absence of such prints in a large number of cases would mean that more times than not the person was not present. Because there are numerous possible explanations for the absence of his prints, it cannot be concluded that there is a probable inference that defendant Dove was not in the bank. Nonetheless, on the other hand, their absence is circumstantial evidence in defendant's favor, and, in a case like this where the photographic evidence indicated that the robber had placed his hands on surfaces likely to retain his fingerprints, required instruction by the trial judge in order that the jury properly assess that fact. See, e.g., Modern Federal Jury Instructions, supra, at p 6.01 (Instruction 6-1).
And, although Dove signed a statement identifying a pair of pants in the bank surveillance pictures as a pair of his own "stonewashed jeans," the evidence showed that he only owned two pairs of jeans--one grey, one black--while all the witnesses in the bank who described the perpetrator to the authorities stated that his jeans were definitely blue.
As a consequence, it was incumbent on the trial court to emphasize that the jury could consider the circumstantial evidence of innocence along with the evidence of guilt. Examining the district court's charge in its entirety, see Cupp v. Naughten, 414 U.S. 141, 146-47, 94 S.Ct. 396, 400, 38 L.Ed.2d 368 (1973), it was not sufficiently balanced in our view. See United States v. Assi, 748 F.2d 62, 67 (2d Cir.1984) ("instructions on the credibility of accomplices and criminal defendants, an issue on which the entire case turned, were not balanced").
Further, a criminal defendant is entitled to instructions relating to his theory of defense, for which there is some foundation in the proof, no matter how tenuous that defense may appear to the trial court. See United States v. Luis, 835 F.2d 37, 40 (2d Cir.1987), citing United States v. O'Connor, 237 F.2d 466, 474 n. 8 (2d Cir.1956). The theory of this defense was that the police had arrested the wrong person. This theory plainly emerged and was supported by the suspect nature of the photographic identification evidence, the failure of the government's two key eyewitnesses to identify the defendant at trial, the suggestion that the fact defendant suffered from alcoholic blackouts was used to secure his confession, the testimony regarding the fingerprints, the bootprints, and the defendant's inculpating statement regarding his clothing while other witnesses exculpated him. The trial court's jury instructions lacked sufficient balance to incorporate this defense theory. These errors of failing to balance the charge and compass the theory of the defense were not harmless, see Assi, 748 F.2d at 68, and necessitate therefore that defendant be given a new trial.
The judgment of conviction is accordingly reversed and vacated and the case is remanded to the district court for a new trial.