No. 19134.United States Court of Appeals, District of Columbia Circuit.Argued June 16, 1965.
Decided November 4, 1965.
Page 863
Mr. George U. Carneal, Jr., Washington, D.C. (appointed by this court), for appellant.
Mr. Dean W. Determan, Sp. Asst. to Atty. Gen., for appellee. Messrs. David C. Acheson, U.S. Atty. at the time the brief was filed, and Frank Q. Nebeker and Joseph A. Lowther, Asst. U.S. Attys., were on the brief for appellee. Mr. Patrick H. Corcoran, Asst. U.S. Atty. at the time the brief was filed, also entered an appearance for appellee.
Before BAZELON, Chief Judge, and WRIGHT and McGOWAN, Circuit Judges.
WRIGHT, Circuit Judge:
This case is before us for the second time following appellant’s conviction on two counts of narcotics violations. On the appeal from the conviction, we held that the trial judge had erred in refusing to allow appellant to cross-examine the police officer concerning the reliability of the informant who had given the information which led to appellant’s arrest.
Page 864
The case was remanded for a hearing on the issue of the informant’s reliability. Jackson v. United States, 118 U.S.App. D.C. 341, 336 F.2d 579 (1964). On remand, after hearing the testimony of the police officer and the informant, the trial court concluded that “probable cause for the defendant’s arrest existed and that he suffered no prejudice because of the arrest and therefore no new trial is required.”
Officer Bello testified that on November 15, 1962, he and another officer were approached by one Ethel Gaskins who informed them that “a Negro male, five foot nine, 25 to 27 years, wearing a brown cap, a tan zipper waist-length jacket, green corduroy trousers, dark complexion,” was in the Franklin Delicatessen and had heroin in his possession. On this information alone — the officers had never heard of appellant — the policemen entered the store, found appellant and another Negro male, and asked both of them to step outside. Once outside, a search first of the person arrested with appellant was unproductive. A subsequent search of appellant disclosed he was carrying contraband narcotics.
The District Court concluded that the quality of the information the officer allegedly received was clearly such as would normally lead directly to the suspect, and that the informant was reliable as she had previously given accurate information to Officer Bello. We agree that the information was sufficiently detailed so that the officers should have been led directly to the suspect. But we reject as clearly erroneous the District Court’s assumption or implied finding that the officers in fact received this detailed information. It follows that the conclusion of probable cause is unacceptable. See Jackson v. United States, supra, 118 U.S.App.D.C. at 342, 336 F.2d at 580.
I
A distinction seems to exist in reviewing judge-made findings in criminal cases between cases where the judge sits in the place of the jury, with the appellate court reviewing his finding of guilt, and where the judge sits and decides matters which traditionally or by statute have been allocated to him. In the former situations, this court applies the same rule it applies in reviewing criminal jury cases: The conviction must be reversed if it is clear “that upon the evidence a reasonable mind could not find guilt beyond a reasonable doubt.”[1]
When dealing with findings of a judge on issues other than guilt, however, the test for review, as indicated by the Supreme Court, varies. In cases where the lower court acts concerning matters peculiarly within its discretion, a narrow test is applied.[2] But fact findings unrelated to the exercise of trial court discretion are subjected to a broader test. While the Federal Rules of Criminal Procedure establish no particular test for review, Rule 57(b), FED.R. CRIM.P., authorizes courts, “if no procedure is specifically prescribed by rule, * * * [to] proceed in any lawful manner not inconsistent with these rules or with any applicable statute.” Thus, in reviewing facts in such cases, courts apply the “clearly erroneous” standard of
Page 865
Rule 52(a), FED.R.CIV.P.[3] For example, the Supreme Court, in Campbell v. United States, 373 U.S. 487, 493, 83 S.Ct. 1356, 10 L.Ed.2d 501 (1963), applied the “clearly erroneous” standard in reviewing a finding of a trial court, pursuant to a motion under the Jencks Act, 18 U.S.C. § 3500(e)(1), that a certain report was actually a copy of the notes of an interview and therefore was a “statement” under the Act. See also Davis v. United States, 328 U.S. 582, 593, 66 S.Ct. 1256, 90 L.Ed. 1453 (1946). Other courts have applied the same test in criminal proceedings.[4]
We apply it here.
“A finding is `clearly erroneous’ when although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed.”[5] It is clear now that the rule applies whether or not there were conflicts in testimony which the lower court resolved. See generally 2B BARRON HOLTZOFF, FEDERAL PRACTICE AND PROCEDURE § 1132 (Wright ed. 1961). The test simply becomes narrower when credibility is involved, a result consistent with the language of Rule 52(a), FED.R. CIV.P. So, while courts applying the rule have rejected findings based upon credibility in both civil, e.g., Riddell v. Guggenheim, 9 Cir., 281 F.2d 836 (1960), and criminal cases, e.g., Maxwell v. United States, supra Note 1, this power should be exercised with great caution. See Marsh v. United States, 2 Cir., 29 F.2d 172, cert. denied, 279 U.S. 849, 49 S.Ct. 346, 73 L.Ed. 992
(1928) (L. Hand, J.). The one thing that seems certain in this area is that the “truth” is elusive. To a large extent, therefore, we must rely upon the trial court’s judgment based upon its observation of the witnesses. See generally FRANK, COURTS ON TRIAL (1949).
II
It remains our responsibility, however, to review fact findings and to reject them when we are firmly convinced they are wrong, when the probability of error is too great to tolerate. This involves, in some contexts at least, an evaluation
Page 866
of the credibility of the witness or witnesses upon whose testimony the finding is based. While the trial judge’s observation of demeanor must be given appropriate weight, it must be remembered that “[c]redibility involves more than demeanor. It apprehends the overall evaluation of testimony in light of its rationality or internal consistency and the manner in which it hangs together with other evidence.” Carbo v. United States, 9 Cir., 314 F.2d 718, 749 (1963), cert. denied, 377 U.S. 953, 84 S.Ct. 1625, 12 L.Ed.2d 498 (1964). Thus, to the extent that the credibility of the police officers in this case turns on their demeanor, we must defer. But that the officers seemed to be telling the truth does not end the matter. A number of other factors often considered in judging credibility must be examined, such as whether the witness was interested in the outcome, his reputation, his degree of recall, the internal inconsistencies in his testimony, and the likelihood of his story.[6]
A close scrutiny of the record in this case is required because of circumstances which seriously hampered the opportunity of the judge at the hearing on mandate fully to evaluate credibility. The testimony in this case is spread out in three transcripts which were compiled over a nearly two-year period.[7] The serious in consistencies in the officers’ testimony become clear only by studying the first transcript and comparing it with the second and third. Significantly, the judge who presided at the trial and at the hearing on mandate did not preside at the first hearing; nor is there adequate proof that he familiarized himself with or considered the transcript of the first hearing.[8]
Apparently the trial judge did not consider the possibility that Bello’s testimony should be rejected. Had he done so specifically, he probably would have dealt with some of the matters we discuss below. As it is, we cannot assume that he gave Bello’s credibility the attention it deserved.
Normally, as an appellate court, we accept the testimony of police officers and other witnesses credited by the trial court. But we are not compelled to accept the testimony of any witness. See, e.g., Kelly v. United States, 90 U.S. App.D.C. 125, 194 F.2d 150 (1952); Herter v. United States, 9 Cir., 27 F.2d 521
(1928), both rejecting police testimony. Officer Bello’s testimony was corroborated by Ralls, his fellow officer,[9]
but, as the Supreme Court has stated, the doctrine that if witnesses concur in proof of a material fact they ought to be believed “can be received only under many qualifications, and with great caution.” The Santissima Trinidad, 20 U.S. (7
Page 867
Wheat.) 283, 338, 5 L.Ed. 454 (1822) (Story, J.). In some cases police testimony, like other testimony, will simply be too weak and too incredible, under the circumstances, to accept.
Officer Bello’s ability to recall dates and events was unimpressive. [See, e.g., 3 Tr. 29, 39.] The only thing he claimed to remember in detail was the description of Jackson. Furthermore, there are important inconsistencies in his testimony. For example, Bello testified at the first hearing that the officers arrested the man who was in the delicatessen with Jackson because “[w]e wanted to be sure of our identification. They resembled each other.” [1 Tr. 19.] At the trial he claimed the arrest was made because the officers did not “want to make it too obvious that we had information that he had the narcotics on him” [2 Tr. 39], and at the hearing on mandate he said the description tallied as soon as he saw Jackson [3 Tr. 25-26]. This and other inconsistencies contribute to our conviction that the credibility of Bello’s testimony was questionable at best.[10]
Our conviction that Bello’s testimony should have been discredited, however, is primarily based on the fact that it contains internal contradictions and is contrary to human experience. The doctrine that appellate courts must reverse findings based upon “inherently incredible” testimony has long been accepted in this jurisdiction.[11] Sometimes, it is possible to disprove testimony as a matter of logic by the uncontradicted facts or by scientific evidence. E.g., The Telephone Cases (Dolbear v. American Bell Tel. Co.), 126 U.S. 1, 567, 8 S.Ct. 778, 31 L.Ed. 863 (1888). But the doctrine of inherent incredibility does not require such positive proof. It is enough to invoke the doctrine if the person whose testimony is under scrutiny made allegations which seem highly questionable in the light of common experience and knowledge,[12] or behaved in a manner strongly at variance with the way in which we would normally expect a similarly situated person to behave.[13]
Page 868
These elements are present in this case. It is difficult to believe that a prostitute, probably an addict as well, in a Negro slum would walk up to an officer and recite so detailed a description of the suspect.[14] Gaskins flatly denied giving the information. Moreover, the uncontradicted testimony indicates that the officers did not behave as we would expect officers with such a detailed description of the suspect to behave. They arrested both Jackson and the other man in the delicatessen, and searched the other man first; and their behavior was inadequately explained, since Bello gave completely inconsistent reasons for arresting the other man. This indeed was a course of conduct pursued by persons with special training and knowledge who “cannot be presumed liable to mistake,” The Santissima Trinidad supra, 20 U.S. (7 Wheat.) at 339, which was “utterly at variance with” their testimony, Atlantic Works v. Brady, supra
Note 13, 107 U.S. (17 Otto) at 203. Even if some consistent explanation were offered, the admission of these officers to one illegal search tends to prove they were involved in mere investigative arrests and thus furnishes strong reason to reject their testimony. We have “the definite and firm conviction” that if the officers had the detailed description of the suspect they claimed to have, they would have arrested but one man. Their action, in the circumstances of this case, in arresting two men, and searching the wrong one first, makes their testimony as to the description of the suspect allegedly received from the informant inherently incredible.[15]
In rejecting under the clearly erroneous test the trial court’s finding that the officers had the description they claimed, we need not reject any of the trial court’s other findings. Thus it is possible that Gaskins had given information in the past, and did tell Bello that there was a Negro man in Franklin’s Delicatessen who had heroin in his possession. We reject only the implied finding that the officers received the detailed description they claimed. And in doing this it should be emphasized that we are not necessarily finding the officers’ testimony untrue. Our task is a technical one. It has nothing to do with truth in the abstract. The officers’ testimony “may be true; but we cannot give it effect against what [they themselves] did, and did not do, without disregarding the ordinary laws that govern human conduct.” Atlantic Works v. Brady, supra Note 13, 107 U.S. (17 Otto) at 203.
Page 869
In any new trial, if the defense of insanity is raised, the court must assure that its presentation is more adequate on behalf of this indigent defendant than its presentation at the trial appealed here. See Jackson v. United States, supra,
118 U.S.App.D.C. at 343, 336 F.2d at 581 (concurring opinion).
Reversed.
(1964), vacated on other grounds, 380 U.S. 527, 85 S.Ct. 1345, 14 L.Ed.2d 266 (1965); United States v. Page, 9 Cir., 302 F.2d 81, 85 (1962) (en banc); United States v. Lawrenson, 4 Cir., 298 F.2d 880, cert. denied, 370 U.S. 947, 82 S.Ct. 1594, 8 L.Ed.2d 812 (1962) (allegation that prosecutor improperly withheld information).
The “clearly erroneous” test is also followed on appeals i habeas corpus proceedings. E.g., Carroll v. Boles, 4 Cir., 347 F.2d 96, 98 (1965). Of course, habeas corpus is considered a civil proceeding, so it is normal that the Rules of Civil Procedure should apply. But the use of the clearly erroneous standard in habeas corpus is significant in that those proceedings generally deal with federal rights, including the question of illegal search and seizure. See also the many cases applying this test in appeals from proceedings under 28 U.S.C. § 2255. E.g., Nichols v. United States, 5 Cir., 325 F.2d 716
(1963), cert. denied, 377 U.S. 970, 84 S.Ct. 1651, 12 L.Ed.2d 739
(1964); Johnston v. United States,, 10 Cir., 292 F.2d 51, cert. denied, 368 U.S. 906, 82 S.Ct. 186, 7 L.Ed.2d 100 (1961); Bayken v. United States, 6 Cir., 272 F.2d 186 (1959); Hearn v. United States, 7 Cir., 194 F.2d 647, cert. denied, 343 U.S. 968, 72 S.Ct. 1064, 96 L.Ed. 1364 (1952).
“* * * It is idle to try to define the meaning of the phrase, `clearly erroneous’; all that can be profitably said is that an appellate court, though it will hesitate less to reverse the finding of a judge than that of an administrative tribunal or of a jury, will nevertheless reverse it most reluctantly and only when well persuaded. * * *” United States v. Aluminum Co. of America, 2 Cir., 148 F.2d 416, 433
(1945).
“THE DEFENDANT: Can you give me a transcript of the pretrial motion?
THE COURT: Is that all you want to ask him?
THE DEFENDANT: May I have a transcript?
THE COURT: Is that all you want to ask him?” [2 Tr. 53.]
This court in effect used the inherently incredible doctrine relatively recently in reversing a criminal conviction rendered by a judge: “It is nearly or quite incredible that appellant could have used a knife as extensively as the girl said he did without her ever seeing it.” Farrar v. United States, supra
Note 1, 107 U.S.App.D.C. at 205, 275 F.2d at 869.
(1956).
(1965); see Wong Sun v. United States, 371 U.S. 471, 479, 83 S.Ct. 407, 9 L.Ed.2d 441 (1963). One reason for this may well be the possibility that officers will fabricate evidence after the arrest to justify their acts, and the difficulty of proving such fabrication. It is true that we do not guard against after-the-fact fabrication in all cases by requiring that a warrant be obtained whenever practicable. But this does not preclude us from attaching significance to the failure to obtain a warrant in appropriate cases. If anything, our decision not to impose a blanket rule requires us to scrutinize more closely the testimony in cases involving warrantless arrests.
McGOWAN, Circuit Judge (dissenting):
With all respect for the views of the majority as to the disposition to be made of this admittedly close case, my own feeling is that reversal is not required. Upon the facts as the trial judge took them to be, the question is solely one of the reasonableness of the police officer’s decision to act on the basis of what the informant had told him. Under the circumstances, that action, if it was to be effective, apparently had to be immediate. I cannot say that, viewing the matter from one situated as was the officer, the decision to act promptly to apprehend appellant was unreasonable. The informant was personally known to the officer as one who had given him reliable leads in the past. That those happened to be in other areas of criminality than narcotics does not seem to me to be controlling. Those very areas suggest that she was living in that unfortunate world where narcotics make their appearance, and where an asserted failure to share them with an associate provides a vengeful and sordid, but highly credible, motivation to inform the police. Assuming that the motivation in question was signified by the informant under the circumstances related by the police, this record has for me a flavor of credibility; or, at the least, I cannot say that it may not reasonably have seemed so to the officer on the beat.
The majority opinion expressly grounds its reversal, however, in its refusal to accept the facts as they appeared to the trial judge. Applying the “clearly erroneous” standard of Rule 52(a), FED.R.CIV.P., it nonetheless rejects the hearing judge’s “assumption or implied finding of fact” that the police officers received any detailed description of appellant from the alleged informant.[1] There being a direct conflict in the testimony on this point between the informant, on the one hand, and the two police officers, on the other, this reversal may be said to come about because the trial judge chose to believe the latter. From the remoter vantage point I occupy — a remoteness in contemplation of which Rule 52(a) was consciously framed — I cannot say that this choice was “clearly erroneous.”
Page 870