982 F2d 1483 Grubbs v. D Hannigan

982 F.2d 1483

61 USLW 2467

Clarence E. GRUBBS, Petitioner-Appellant,
v.
Robert D. HANNIGAN; Attorney General of Kansas,
Respondents-Appellees.

No. 91-3303.

United States Court of Appeals,
Tenth Circuit.

Jan. 6, 1993.

Tom Jacquinot (Kathleen Levy, on the brief), Staff Counsel, Kansas Defender Project, University of Kansas School of Law, Lawrence, KS, for petitioner/appellant.

Kyle G. Smith, Asst. Atty. Gen., Kansas Bureau of Investigation, Topeka, KS, for respondents/appellees.

Before BRORBY, Circuit Judge, SETH, Senior Circuit Judge, and EBEL, Circuit Judge.

EBEL, Circuit Judge.

1

In this appeal, the appellant is seeking habeas corpus relief pursuant to 28 U.S.C. § 2254. The appellant was convicted by a jury in the district court of Shawnee County, Kansas of raping and robbing two women, T.J. and S.S. The appellant collaterally attacks these convictions on two grounds: 1) that the evidence supporting his conviction for the rape and robbery of T.J. was insufficient, and 2) that the admission of S.S.'s pre-trial identification testimony violated his right to due process. We find these arguments to be without merit and therefore affirm.1

FACTS

2

On October 22, 1985, S.S. was working as a receptionist in an optometrist's office in Topeka, Kansas. At 4:35 p.m., just after the office closed, a black man walked into the lobby and approached S.S.'s desk. The man told S.S. that he had missed an appointment and wanted to reschedule it. As S.S. looked down at her appointment book, the man walked behind S.S. and put his hand over her mouth. In his other hand, he brandished a knife. He instructed S.S. to lie down on the floor in front of her desk and proceeded to rape her. After raping S.S., the man took $62.00 from S.S. and left the office.

3

The night of the attack, S.S. gave a statement to the police in which she described the man who raped and robbed her as a black male in his early twenties, 5'8" tall, 160 pounds, with a smooth, clean face and a short afro hair style. She viewed a six-picture photo-lineup but determined that her assailant was not among these pictures. The next week she was shown another six-picture photo-lineup which included the picture of the appellant and one picture from the previous lineup. Upon viewing the second photo-lineup, S.S. immediately identified the appellant as her assailant.

4

On October 28, 1985, T.J., a 17-year-old high school student, was working at a dry cleaning store in Topeka, Kansas. At approximately 6:00 p.m., a black man entered the store and asked about cleaning prices. T.J. answered his inquiry and returned to her work at the back of the store. Instead of leaving the store, the man came around the counter and headed toward T.J. When T.J. asked him what he was doing, he pulled a knife and instructed T.J. to lock the front door. He proceeded to take $100.00 from the cash register and a money bag sitting on the counter. After taking the money, the man pushed T.J. to the back of the store where he raped her.

5

Following the attack, T.J. called the police and was taken to the police station. She described the man who raped and robbed her as a black male, 5'7"' tall, 137 pounds, with straight hair and a smooth complexion. She also described his attire as a black jacket with a gold stripe down the left side of the chest and a pair of three color fatigue pants. The police later discovered a jacket and a pair of pants matching this description in the appellant's bedroom closet. Fibers found on the jacket were tested by Phillip Aviles, a fiber expert from the Kansas Bureau of Investigation, and were found to match other fibers found on T.J.'s clothing and on the floor where she was raped, though these fibers were determined to be available from a number of sources.

6

While at the police station, T.J. looked through approximately 50 photographs but determined that her assailant was not among them. The following day, she viewed a six-picture photo-lineup containing the picture of the appellant and immediately identified him as her assailant.

7

The night of the attack T.J. was also taken to a local hospital where seminal fluid samples were taken from her vagina. These samples were transported to the Kansas Bureau of Investigation, along with blood samples taken from T.J. and the appellant. Forensic examiner Susan Scholl subjected each of these samples to blood group substance testing, which identifies the subtype of a particular enzyme found in human blood known as PGM. The seminal fluid sample taken from T.J. showed the presence of a PGM subtype of 2+1+. The blood sample taken from T.J. contained a PGM subtype of 1+. The blood sample taken from the appellant contained a PGM subtype of 1+1-. Scholl retested blood samples from both T.J. and the appellant and found the same results. Based on the test results, Scholl concluded that the seminal fluid taken from T.J. could not have come from the appellant.

8

The appellant was indicted for raping and robbing both T.J. and S.S., and these charges were tried together before a jury in Shawnee County, Kansas. At trial, S.S. testified as to her pre-trial identification of the appellant and stated that she was certain about her identification. S.S. testified that the assailant had been "within a foot" of her face at the time of the rape, and that she had concentrated on his face in order to "remember what he looked like." S.S. also stated that she recognized the appellant from a previous visit to the office.

9

On cross-examination, S.S. admitted that three of the men in the second photo-lineup had hairstyles significantly different from that of the appellant and that a fourth man had a swollen eye. She also admitted that the police had informed her prior to the second photo-lineup that they had a suspect and that she therefore assumed that the photo-lineup contained the suspect's picture. S.S. stated that she was unaware, however, that any of the pictures in the second photo-lineup had been reused from the first photo-lineup.

10

T.J. also testified about her pre-trial identification of the appellant and stated that she was positive of her identification. T.J. testified that the appellant's face was "within inches" of her own at the time of the rape and, like S.S., she "intentionally look[ed] at his face" in order to identify him. T.J. concluded her testimony by identifying the clothes found in the appellant's closet as the clothes worn by the man who raped and robbed her. Following T.J.'s testimony, the parties stipulated that T.J. was a virgin prior to the rape.

11

In addition to the testimony of the victims, the state presented the testimony of Aviles and Scholl.2 Aviles testified as to the matching fibers he found on the appellant's jacket, T.J.'s clothing, and the floor of the dry-cleaning store. He stated on cross-examination, however, that the fibers in question were quite common and could have come from "a number of different sources."

12

Scholl testified that the blood group substance tests she performed eliminated the appellant as a possible donor of the seminal fluid sample taken from T.J. She also testified that blood group substance testing was a well-tested and accurate scientific procedure and that it would yield consistent results provided "the procedure was done correctly." Scholl testified that the procedure was performed properly in this case.

13

The appellant testified on his own behalf. He stated that between 3:20 p.m. and 4:50 p.m. on October 22, 1985, he was visiting his foster mother, Helen Kingcannon. Kingcannon confirmed that the appellant had been to visit her, but could not confirm the time or date. The appellant also testified that on October 28, 1985, he was at home with his wife from 5:00 p.m. to 8:00 p.m., at which time he began moving from one apartment to another with the help of his uncle, Reverend Ronald Lassiter. The appellant's wife corroborated his testimony. However, Detective Halford testified that when he initially interviewed the appellant and his wife they told him that Reverend Lassiter had begun helping them move between 5:30 and 6:00 p.m.

14

The jury found the appellant guilty on all counts. The Kansas Supreme Court affirmed the appellant's conviction and the United States District Court for the District of Kansas rejected the appellant's petition for habeas corpus 771 F.Supp. 1159.

DISCUSSION

15

I. Was there Sufficient Evidence to Sustain the Conviction for Raping T.J.?

16

The appellant's first claim is that the evidence supporting his conviction for the rape and robbery of T.J. was insufficient.3 The evidence supporting a conviction is insufficient if no rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt. Jackson v. Virginia, 443 U.S. 307, 319, 99 S.Ct. 2781, 2789, 61 L.Ed.2d 560 (1979); United States v. Thurston, 771 F.2d 449, 452 (10th Cir.1985). "If substantial evidence supports the verdict, it cannot be set aside." United States v. Leach, 749 F.2d 592, 600 (10th Cir.1984).

17

In this case, there was strong evidence indicating that the appellant committed the rape and robbery in question. T.J. immediately identified the appellant upon viewing his picture in a photo-lineup and testified at trial that she had a good opportunity to view her assailant at the time of the rape. Additionally, the police discovered a jacket and a pair of pants in the appellant's bedroom that matched T.J.'s initial description of her assailant's attire, and fibers discovered on the jacket matched fibers found at the scene of the crime and on T.J.'s clothing. Finally, the appellant's alibi testimony was inconsistent with the testimony of other witnesses.

18

The appellant does not dispute that this evidence, standing alone, would be sufficient to support his conviction. However, the appellant argues that this evidence is insufficient in light of the blood test evidence presented by Susan Scholl. According to the appellant, this evidence indicated that he was not the donor of the seminal sample taken from T.J. on the night of the rape, and since T.J. was a virgin prior to the rape, the seminal sample had to belong to the rapist. Thus, the appellant concludes, no rational trier of fact could have found him guilty beyond a reasonable doubt.

19

However, in reviewing a sufficiency claim, "we are not permitted to weigh conflicting evidence or consider witness credibility." United States v. Davis, 965 F.2d 804, 811 (10th Cir.1992), cert. filed, 61 U.S.L.W. 3371 (Nov. 3, 1992). Rather, we must view the evidence in the light most favorable to the prosecution. Jackson, 443 U.S. at 319, 99 S.Ct. at 2789; Thurston, 771 F.2d at 452. This standard requires us to accept the jury's resolution of the evidence as long as it is within the bounds of reason. United States v. Edmondson, 962 F.2d 1535, 1548 (10th Cir.1992) ("A jury is free to choose among reasonable constructions of the evidence.") (quoting United States v. Perkins, 748 F.2d 1519, 1526 (11th Cir.1984)); see United States v. Hager, 969 F.2d 883, 887-88 (10th Cir.) (concluding that the evidence was sufficient to support conviction for possession despite testimony that drugs belonged to defendant's friend, since jury's decision to disbelieve testimony was not unreasonable), cert. denied, --- U.S. ----, 113 S.Ct. 437, 121 L.Ed.2d 357 (1992). The jury acts within these bounds when its decision to disregard evidence has a basis in the record. See United States v. Hines, 696 F.2d 722, 730 (10th Cir.1982). Here, the record indicates that the blood test evidence was sufficiently impeached by the prosecution to permit the jury to disregard this evidence.

20

Generally, a prosecutor may impeach expert testimony introduced by a defendant by "(1) introduc[ing] its own expert testimony in rebuttal; or (2) discredit[ing] the defendant's expert testimony on cross-examination; or (3) rely[ing] upon evidence from which the jury may infer that the defendant's expert testimony depends on an incorrect view of the facts." United States v. Bodey, 607 F.2d 265, 269 (9th Cir.1979) (quoting United States v. McGraw, 515 F.2d 758, 760 (9th Cir.1975)); see also United States v. Dube, 520 F.2d 250, 252 (1st Cir.1975) ("[E]xpert testimony is not conclusive even where uncontradicted ... and it may be rebutted in various ways apart from the introduction of countervailing expert opinion."). Although scientific expert testimony often involves less subjective judgment than other types of expert testimony, the same principles of impeachment apply. Thus, a prosecutor may impeach exculpatory scientific evidence either directly, by means of rebuttal evidence or cross-examination, or indirectly, by presenting evidence from which the jury can logically infer that the scientific evidence must be in error. To accomplish this indirect method of impeachment, the prosecutor need not pinpoint the source of error. Such a requirement would demand a greater degree of scientific acumen than the scientific process often permits. Rather, the prosecutor need only introduce evidence that is inconsistent with the conclusions drawn by the defendant's scientific expert, such that a rational individual could not believe both sets of evidence simultaneously.

21

The fact that scientific evidence introduced by a defendant may be impeached indirectly does not mean that a jury may disregard such evidence whenever the prosecution points to inculpatory evidence yielding an inference that the scientific evidence is erroneous. The jury may not disregard scientific evidence that is so conclusive that to disregard it would amount to mere conjecture or speculation. "When science acquires knowledge whereby the ascertainment of the truth ... is certain, courts [may not] ignore such knowledge." Commonwealth v. D'Avella, 339 Mass. 642, 162 N.E.2d 19, 21 (1959). However, when the jury is presented with scientific evidence that falls short of this high degree of incontrovertibility, the jury may choose to disregard it in the face of strong evidence of guilt. This rule is consistent with the reverse principle that a jury may formulate its opinion on the basis of any evidence in the record that is not "inherently unbelievable," United States v. Narcisco, 446 F.Supp. 252, 282 (E.D.Mich.1977) (citation omitted), and does not "def[y] physical laws," United States v. Gardea Carrasco, 830 F.2d 41, 44 (5th Cir.1987) (citations omitted); see also Born v. Osendorf, 329 F.2d 669, 672 (8th Cir.1964) ("No jury can be allowed to return a verdict based upon oral testimony which is flatly opposed to physical facts, the existence of which is incontrovertibly established.... However, so frequently do unlooked for results attend the meeting of interacting forces that courts should not indulge in arbitrary deductions from physical law and fact except where they appear to be so clear and irrefutable that no room is left for the entertainment, by reasonable minds, of any other deductions.").

22

Here, the prosecutor relied on the indirect method of impeachment to attack the blood test evidence. He argued that although he could not establish the precise error underlying the PGM test results, the jury could infer that the results were erroneous in light of T.J.'s eyewitness testimony, the matching fiber evidence, and the inconsistencies in the defendant's testimony. The jury's verdict indicates that it chose to accept this inference. The question then is whether the blood test evidence was sufficiently conclusive that the jury's decision must be considered irrational. We do not believe that the blood test evidence was so conclusive.

23

There is no testimony in the record on the general accuracy of PGM testing. Nor is there any testimony in the record on when the seminal fluid sample was taken from T.J., even though Susan Scholl testified that PGM in semen begins to break down within a few hours after intercourse. Finally, the record indicates that a key portion of the blood test results were uncorroborated. Although Scholl was able to retest blood samples taken from T.J. and the appellant, she was unable to retest the seminal sample. In light of these omissions, we do not believe the accuracy of the blood test evidence was established to a sufficient degree of incontrovertibility that the failure to treat the blood test evidence as conclusive would be tantamount to suspending the laws of nature. Accordingly, although we might not have reached the same result, we cannot say that the jury's decision to disregard the blood test evidence in favor of the countervailing inculpatory evidence violated the appellant's constitutional rights under Jackson v. Virginia, 443 U.S. 307, 99 S.Ct. 2781, 61 L.Ed.2d 560 (1979).

24

The Florida Court of Appeals reached a similar conclusion in Robinson v. State, 462 So.2d 471 (Fla. 1 DCA 1984), a case with virtually identical facts. In Robinson, as in the instant case, there was expert scientific testimony that the semen sample taken from the rape victim contained a different PGM subtype than that contained in the defendant's blood. Additionally, there was expert scientific testimony that the sperm count in the semen sample was significantly lower than the sperm count in the defendant's semen. Other physical evidence, however, supported the victim's testimony that the defendant was the individual who raped her. The Florida Court of Appeals rejected the defendant's claim that the evidence was insufficient, holding that the exculpatory scientific evidence did not "vitiate the legal sufficiency" of the evidence establishing the defendant's guilt. Id. at 475.4

25

To support his conclusion that the jury acted irrationally in disregarding the blood test evidence, the appellant relies on the field of paternity law where a majority of courts have treated exculpatory blood test evidence as conclusive of paternity. See, e.g., G.M.H. v. J.L.H., 700 S.W.2d 506, 517 (Mo.App.1985); Anonymous v. Anonymous, 10 Ariz.App. 496, 460 P.2d 32, 35 (1969); Commonwealth v. D'Avella, 339 Mass. 642, 162 N.E.2d 19, 21-22 (1959); Jordan v. Mace, 144 Me. 351, 69 A.2d 670, 672 (1949); see also John P. Ludington, Annotation: Admissibility and Weight of Blood-Grouping Tests in Disputed Paternity Cases, 43 ALR 4th 579, § 8(a) (1986). We find this reliance to be misplaced. The unique characteristics of paternity proceedings imbue the blood test evidence in these cases with a degree of reliability not present in the instant case.

26

In paternity cases, all of the blood samples can be drawn right in the laboratory, so there is little risk of pre-test contamination or chain-of-custody problems. Furthermore, each of the unknown components--the blood of the mother, the father, and the baby--is available for repeated testing. If any doubt exists about whether the tests were properly conducted, therefore, the results of the test can be easily verified. See V.L.P. v. J.S.S., 407 A.2d 244, 249 (Del.Fam.Ct.1978) (court required defendant to undergo a second paternity test where plaintiff alleged that there had been some mix-up in blood samples). Consequently, paternity cases constitute "one of the few classes of cases" in which scientific certainty can be achieved to such a degree that the courts are justified in according conclusive weight to the blood test evidence in these cases. D'Avella, 162 N.E.2d at 21.

27

By comparison, the blood test evidence in the instant case is not so incontrovertible as to rise to the level that all reasonable jurors would be compelled to accept it even in the face of strong inculpatory evidence. As pointed out previously, there was no testimony on the degree of reliability of PGM tests nor was there any detailed testimony on how the semen sample was obtained and preserved. Furthermore, the test results were based upon but a single test of the seminal sample since the seminal sample was not available for retesting. Consequently, the certainty of the blood test evidence in the instant case does not compare to that of the blood test evidence accorded conclusive weight in paternity cases.

28

It is not our intention here to decide exactly how extensive the guarantees of reliability must be before scientific evidence may be said to be incontrovertible. We will leave a more precise drawing of the line of scientific certitude for future cases. Wherever the line exists, however, we are convinced that the blood test evidence in this case falls short. Consequently, we believe the jury could rationally disregard this evidence in light of the countervailing inculpatory evidence and find the appellant guilty beyond a reasonable doubt. Accordingly, we find that the evidence was sufficient to support the appellant's conviction for the rape and robbery of T.J.

29

II. Did S.S.'s Photo-Identification Violate Due Process?

30

The appellant's second claim is that admission of S.S.'s pre-trial identification testimony violated his right to due process.5 In examining the constitutionality of pre-trial identification procedures, we must engage in a two-tier analysis. First, we must determine whether the procedure was unnecessarily suggestive. Archuleta v. Kerby, 864 F.2d 709, 711 (10th Cir.), cert. denied, 490 U.S. 1084, 109 S.Ct. 2108, 104 L.Ed.2d 669 (1989). If the procedure is found to have been unnecessarily suggestive, we must then weigh the corrupting influence of the suggestive procedure against the reliability of the identification itself. Manson v. Brathwaite, 432 U.S. 98, 114, 97 S.Ct. 2243, 2253, 53 L.Ed.2d 140 (1977). "[T]he 'central question is whether under the totality of circumstances, the identification was reliable.' " United States v. Thurston, 771 F.2d 449, 453 (10th Cir.1985) (quoting Neil v. Biggers, 409 U.S. 188, 198, 93 S.Ct. 375, 381, 34 L.Ed.2d 401 (1972)). Only when a pre-trial identification procedure is so unnecessarily suggestive that it is "conducive to irreparable mistaken identification" does the procedure violate due process. Kirby v. Illinois, 406 U.S. 682, 691, 92 S.Ct. 1877, 1883, 32 L.Ed.2d 411 (1972). "[R]eliability is the linchpin in determining the admissibility of identification testimony...." Brathwaite, 432 U.S. at 114, 97 S.Ct. at 2253.

31

In Neil v. Biggers, 409 U.S. 188, 93 S.Ct. 375, 34 L.Ed.2d 401 (1972), the Supreme Court set forth five factors to be considered in determining the reliability of a pre-trial identification. These factors are:

32

the opportunity of the witness to view the criminal at the time of the crime, the witness' degree of attention, the accuracy of his prior description of the criminal, the level of certainty demonstrated by the witness at the confrontation, and the length of time between the crime and the confrontation.

33

Biggers, 409 U.S. at 199-200, 93 S.Ct. at 382. It is these factors which must be weighed against the corruptive effect of a suggestive pre-trial identification procedure to determine whether the identification testimony should have been suppressed. Archuleta, 864 F.2d at 711; Thurston, 771 F.2d at 453.

34

In this case, we find that the pre-trial identification procedures followed by the police were unnecessarily suggestive. The second six-photo array viewed by S.S. contained four individuals who had facial characteristics noticeably dissimilar from those of the appellant and a fifth individual who had been seen by S.S. in a prior lineup. Although a photo-lineup is not necessarily suggestive merely because the individuals in the lineup differ in facial characteristics, see United States v. Merryman, 630 F.2d 780, 786-87 (10th Cir.1980); United States v. Barron, 575 F.2d 752, 755 (9th Cir.1978), here the differences were either strikingly apparent, such as a swollen eye, or they related to an important component of S.S.'s description of her assailant, his hair style. Furthermore, the suggestiveness of these differences was compounded by the fact that the police informed S.S. prior to her identification of the appellant that they had a suspect. S.S. admitted that this information caused her to assume that one of the individuals in the lineup was the suspect, and such coaching has long been recognized by the courts as an important factor in determining whether a pre-trial lineup procedure was impermissibly suggestive, see United States v. Monks, 774 F.2d 945, 956 (9th Cir.1985) (photo-lineup was not unnecessarily suggestive even though all the other persons in the lineup were "very dissimilar to the [defendant]" since the witnesses were not informed that they were viewing the suspect); United States v. Coppola, 486 F.2d 882, 887 (10th Cir.1973) (photo-lineup was not impermissibly suggestive even though photograph of defendant differed from the other photographs since "there was no suggestion that one picture should be picked out over the others"), cert. denied, 415 U.S. 948, 94 S.Ct. 1469, 39 L.Ed.2d 563 (1974).

35

Having found that the identification procedures were unnecessarily suggestive, we must next determine whether the reliability of the identification outweighs the corrupting effect of these suggestive procedures. An examination of the five Biggers factors reveals that S.S.'s pre-trial identification contained sufficient indicia of reliability that it did not create "a very substantial likelihood of irreparable misidentification." Brathwaite, 432 U.S. at 116, 97 S.Ct. at 2254 (citation omitted). The confrontation occurred within a week of the crime and S.S. immediately identified the appellant upon viewing the photo-lineup. Cf. Biggers, 409 U.S. 188, 93 S.Ct. 375 (pre-trial identification found reliable despite seven-month lapse before identification). The crime itself occurred in a lighted lobby, and S.S. had a good opportunity to view her assailant both prior to the rape, as he stood in front of her desk, and during the rape, when his face was less than a foot away from hers. Cf. O'Brien v. Wainwright, 738 F.2d 1139, 1141 (11th Cir.1984) (upholding reliability of close-up view lasting only seconds), cert. denied, 469 U.S. 1162, 105 S.Ct. 918, 83 L.Ed.2d 931 (1985). Additionally, S.S. testified that she specifically concentrated on her assailant's face so as to be able to remember what he looked like. Finally, the appellant concedes that S.S.'s description of her assailant closely resembles the physical appearance of the appellant himself.

36

On the strength of these factors, we conclude that the reliability of S.S.'s pre-trial identification outweighed the corruptive influence generated by the suggestive pre-trial identification procedures. Accordingly, we reject the appellant's claim that the admission of S.S.'s identification testimony deprived him of due process.

37

AFFIRMED.

1

The appellant also challenges his conviction for raping and robbing S.S. on the ground that the trial judge failed to give a limiting instruction to the jury on the use of evidence that he raped and robbed T.J. as "other crimes" evidence in assessing his guilt for the crimes involving S.S. We decline to address this argument, however, because it was not raised below. See Farmers Ins. Co. v. Hubbard, 869 F.2d 565, 570 (10th Cir.1989); Burnette v. Dresser Indus., Inc., 849 F.2d 1277, 1282 (10th Cir.1988)

2

Apparently, the state called Scholl as a witness in order to take the "sting" out of her testimony

3

In Jackson v. Virginia, the Supreme Court held that a sufficiency of the evidence claim implicates a defendant's constitutional right to due process of law. 443 U.S. 307, 320-21, 99 S.Ct. 2781, 2789-90 (1979). Thus, such a claim is reviewable on a petition for habeas corpus. See 28 U.S.C. § 2254(a). The ultimate legal issue of the sufficiency of the evidence is reviewed de novo. Case v. Mondragon, 887 F.2d 1388, 1393 (10th Cir.1989), cert. denied, 494 U.S. 1035, 110 S.Ct. 1490, 108 L.Ed.2d 626 (1990)

4

The court in Robinson ultimately afforded relief to the defendant by granting his untimely request for a new trial. Robinson, 462 So.2d at 477. In contrast to Robinson, however, the instant case is a habeas action rather than a direct criminal appeal. Consequently, the only issue before us is whether there is sufficient evidence of guilt on the record to satisfy the due process standard adopted by Jackson v. Virginia, 443 U.S. at 319, 99 S.Ct. at 2789. On this issue, our holding is consistent with Robinson

5

The ultimate question of whether the admission of pre-trial identification testimony violates due process is reviewed de novo on appeal, although the underlying facts found by the state court are entitled to a statutory presumption of correctness pursuant to 28 U.S.C. 2254(d). Sumner v. Mata, 455 U.S. 591, 597, 102 S.Ct. 1303, 1306, 71 L.Ed.2d 480 (1982) (per curiam)