379 U.S. 89 (1964)
On the afternoon of November 10, 1961, the petitioner, William Beck, was driving his automobile in the vicinity of East 115th Street and Beulah Avenue in Cleveland, Ohio. Cleveland police officers accosted him, identified themselves, and ordered him to pull over to the curb. The officers possessed neither an arrest warrant nor a search warrant. Placing him under arrest, they searched his car but found nothing of interest. They then took him to a nearby police station where they searched his person and found an envelope containing a number of clearing house slips “beneath the sock of his leg.” The petitioner was subsequently charged in the Cleveland Municipal Court with possession of clearing house slips in violation of a state criminal statute.1
He filed a motion to suppress as evidence the clearing house slips in question, upon the ground that the police had obtained them by means of an unreasonable search and seizure in violation of the Fourth and Fourteenth Amendments. After a hearing the motion was overruled, the clearing house slips were admitted in evidence, and the petitioner was convicted. His conviction was affirmed by an Ohio Court of Appeals, and ultimately by the Supreme Court of Ohio, with two judges dissenting. 175 Ohio St. 73, 191 N. E. 2d 825. We granted certiorari to consider the petitioner’s claim that, under the rule of Mapp v. Ohio, 367 U. S. 643, the clearing house slips were wrongly admitted in evidence against him because they had been seized by the Cleveland police in violation of the Fourth and Fourteenth Amendments. 376 U. S. 905.
Although the police officers did not obtain a warrant before arresting the petitioner and searching his automobile and his person, the Supreme Court of Ohio found the search nonetheless constitutionally valid as a search incident to a lawful arrest. And it is upon that basis that the Ohio decision has been supported by the respondent here. See Draper v. United States, 358 U. S. 307; Ker v. California, 374 U. S. 23.
There are limits to the permissible scope of a warrantless search incident to a lawful arrest, but we proceed on the premise that, if the arrest itself was lawful, those limits were not exceeded here. See Harris v. United States, 331 U. S. 145; United States v. Rabinowitz, 339 U. S. 56; cf. Preston v. United States, 376 U. S. 364. The constitutional validity of the search in this case, then, must depend upon the constitutional validity of the petitioner’s arrest. Whether that arrest was constitutionally valid depends in turn upon whether, at the moment the arrest was made, the officers had probable cause to make it — whether at that moment the facts and circumstances within their knowledge and of which they had reasonably trustworthy information were sufficient to warrant a prudent man in believing that the petitioner had committed or was committing an offense. Brinegar v. United States, 338 U. S. 160, 175-176; Henry v. United States, 361 U. S. 98, 102. “The rule of probable cause is a practical, nontechnical conception affording the best compromise that has been found for accommodating . . . often opposing interests. Requiring more would unduly hamper law enforcement. To allow less would be to leave law-abiding citizens at the mercy of the officers’ whim or caprice.” Brinegar v. United States, supra, at 176.
In turning to the question of whether or not the record in the case before us can support a finding of probable cause for the petitioner's arrest, it may be well to repeat what was said by Mr. Justice Clark, speaking for eight members of the Court, in Ker v. California:
“While this Court does not sit as in nisi prius to appraise contradictory factual questions, it will, where necessary to the determination of constitutional rights, make an independent examination of the facts, the findings, and the record so that it can determine for itself whether in the decision as to reasonableness the fundamental — i. e., constitutional — criteria established by this Court have been respected. The States are not thereby precluded from developing workable rules governing arrests, searches and seizures to meet ‘the practical demands of effective criminal investigation and law enforcement’ in the States, provided that those rules do not violate the constitutional proscription of unreasonable searches and seizures and the concomitant command that evidence so seized is inadmissible against one who has standing to complain. See Jones v. United States, 362 U. S. 257 (1960). Such a standard implies no derogation of uniformity in applying federal constitutional guarantees but is only a recognition that conditions and circumstances vary just as do investigative and enforcement techniques.” 374 U. S. 23, at 34.
The trial court made no findings of fact in this case. The trial judge simply made a conclusory statement: “A lawful arrest has been made, and this was a search incidental to that lawful arrest.” The Court of Appeals merely found “no error prejudicial to the appellant.” In the Supreme Court of Ohio, Judge Zimmerman’s opinion contained a narrative recital which is accurately excerpted in the dissenting opinions filed today. But, putting aside the question of whether this opinion can fairly be called the opinion of the court,2 such a recital in an appellate opinion is hardly the equivalent of findings made by the trier of the facts. In any event, after giving full scope to the flexibility demanded by “a recognition that conditions and circumstances vary just as do investigative and enforcement techniques,” we hold that the arrest of the petitioner cannot on the record before us be squared with the demands of the Fourth and Fourteenth Amendments.
The record is meager, consisting only of the testimony of one of the arresting officers, given at the hearing on the motion to suppress. As to the officer’s own knowledge of the petitioner before the arrest, the record shows no more than that the officer “had a police picture of him and knew what he looked like,” and that the officer knew that the petitioner had “a record in connection with clearing house and scheme of chance.”3 Beyond that, the officer testified only that he had “information” that he had “heard reports,” that “someone specifically did relate that information,” and that he “knew who that person was.” There is nowhere in the record any indication of what “information” or “reports” the officer had received, or, beyond what has been set out above, from what source the “information” and “reports” had come. The officer testified that when he left the station house, “I had in mind looking for [the petitioner] in the area of East 115th Street and Beulah, stopping him if I did see him make a stop in that area.” But the officer testified to nothing that would indicate that any informer had said that the petitioner could be found at that time and place. Cf. Draper v. United States, 358 U. S. 307. And the record does not show that the officers saw the petitioner “stop” before they arrested him, or that they saw, heard, smelled, or otherwise perceived anything else to give them ground for belief that the petitioner had acted or was then acting unlawfully.4
No decision of this Court has upheld the constitutional validity of a warrantless arrest with support so scant as this record presents. The respondent relies upon Draper v. United States, 358 U. S. 307. But in that case the record showed that a named special employee of narcotics agents who had on numerous occasions given reliable information had told the arresting officer that the defendant, whom he described minutely, had taken up residence at a stated address and was selling narcotics to addicts in Denver. The informer further had told the officer that the defendant was going to Chicago to obtain narcotics and would be returning to Denver on one of two trains from Chicago, which event in fact took place. In complete contrast, the record in this case does not contain a single objective fact to support a belief by the officers that the petitioner was engaged in criminal activity at the time they arrested him.
An arrest without a warrant bypasses the safeguards provided by an objective predetermination of probable cause, and substitutes instead the far less reliable procedure of an after-the-event justification for the arrest or search, too likely to be subtly influenced by the familiar shortcomings of hindsight judgment. “Whether or not the requirements of reliability and particularity of the information on which an officer may act are more stringent where an arrest warrant is absent, they surely cannot be less stringent than where an arrest warrant is obtained. Otherwise, a principal incentive now existing for the procurement of arrest warrants would be destroyed.” Wong Sun v. United States, 371 U. S. 471, 479-480. Yet even in cases where warrants were obtained, the Court has held that the Constitution demands a greater showing of probable cause than can be found in the present record. Aguilar v. Texas, 378 U. S. 108; Giordenello v. United States, 357 U. S. 480;5 Nathanson v. United States, 290 U. S. 41.6
When the constitutional validity of an arrest is challenged, it is the function of a court to determine whether the facts available to the officers at the moment of the arrest would “warrant a man of reasonable caution in the belief” that an offense has been committed. Carroll v. United States, 267 U. S. 132, 162. If the court is not informed of the facts upon which the arresting officers acted, it cannot properly discharge that function. All that the trial court was told in this case was that the officers knew what the petitioner looked like and knew that he had a previous record of arrests or convictions for violations of the clearing house law. Beyond that, the arresting officer who testified said no more than that someone (he did not say who) had told him something (he did not say what) about the petitioner. We do not hold that the officer’s knowledge of the petitioner’s physical appearance and previous record was either inadmissible or entirely irrelevant upon the issue of probable cause. See Brinegar v. United States, 338 U. S. 160, 172-174. But to hold that knowledge of either or both of these facts constituted probable cause would be to hold that anyone with a previous criminal record could be arrested at will.
It is possible that an informer did in fact relate information to the police officer in this case which constituted probable cause for the petitioner’s arrest. But when the constitutional validity of that arrest was challenged, it was incumbent upon the prosecution to show with considerably more specificity than was shown in this case what the informer actually said, and why the officer thought the information was credible. We may assume that the officers acted in good faith in arresting the petitioner. But “good faith on the part of the arresting officers is not enough.” Henry v. United States, 361 U. S. 98, 102. If subjective good faith alone were the test, the protections of the Fourth Amendment would evaporate, and the people would be “secure in their persons, houses, papers, and effects,” only in the discretion of the police.
Reversed.
The Supreme Court of Ohio, 175 Ohio St. 73, 74, 191 N.E.2d 825, 827, “determined” the following facts in this case:
“The Cleveland police had good reason to believe that defendant was regularly engaged in carrying on a scheme of chance involving clearinghouse slips. There was testimony that he had previously been convicted on that score. Information was given to the police by an informer that defendant would be in a certain locality at a certain time pursuing his unlawful activities. He was found in that locality, as predicted, driving an automobile. Police officers stopped the car and searched it, without result. Defendant was then arrested and taken to a police station, and his clothing was examined, resulting in the discovery and seizure of the illegal clearinghouse slips, which formed the basis of the charge against him and his subsequent conviction.”
These are the facts upon which Ohio’s highest court based its opinion and they have support in the record.
The syllabus rule, Rule VI, peculiar to that State and of which the majority speaks, was promulgated in 1858, 5 Ohio St. vii, and provides:
“A syllabus of the points decided by the Court in each case, shall be stated, in writing, by the Judge assigned to deliver the opinion of the Court, which shall be confined to the points of law, arising from the facts of the case, that have been determined by the Court. . .” (Emphasis supplied.)
As my late Brother of revered memory, Mr. Justice Burton of Ohio, said in the Ohio case of Perkins v. Benguet Consol. Mining Co., 342 U.S. 437, 442, n. 3 (1952), “[a] syllabus must be read in the light of the facts in the case, even where brought out in the accompanying opinion rather than in the syllabus itself.” The good Justice was only following Ohio’s own cases. See Williamson Heater Co. v. Radich, 128 Ohio St. 124, 190 N.E. 403 (1934); Perkins v. Bright, 109 Ohio St. 14, 19-20, 141 N.E. 689, 690 (1923); In re Poage, 87 Ohio St. 72, 82-83, 100 N.E. 125, 127-128 (1912).
The Court ignores these findings entirely. Where the highest court of a State after detailed and earnest consideration determines the facts and they are reasonably supportable, I would let them stand. And I would, of course, give the same respect to findings of probable cause by United States district courts when approved by United States courts of appeals. Otherwise, this Court will be continually disputing with state and federal courts over the minutiae of facts in every search and seizure case. Especially is this true if the Court disputes the findings sua sponte where, as here, no attack is leveled at them.
Believing that the Ohio Supreme Court’s findings, set out above, fully support its conclusion that probable cause existed, in this case in support of the arrest and the search incident thereto, I would affirm.
Mr. Justice Harlan, dissenting.
Judge Zimmerman of the Supreme Court of Ohio stated as a fact,7 “Information was given to the police by an informer that defendant would be in a certain locality at a certain time pursuing his unlawful activities. He was found in that locality as predicted, driving an automobile.” I regard this as the crucial point in the case, for if the informant did give the police that information, the fact of its occurrence would sufficiently indicate the informant’s reliability to provide a basis for petitioner’s arrest, Draper v. United States, 358 U. S. 307. It is this Court’s function, therefore, to determine whether the State’s finding is adequately supportable. In doing so it is essential to consider what are the appropriate standards of appellate review.
Generally “our inquiry clearly is limited to a study of the undisputed portions of the record.” Thomas v. Arizona, 356 U. S. 390, 402. “[T]here has been complete agreement that any conflict in testimony as to what actually led to a contested confession [or to a contested arrest] is not this Court’s concern. Such conflict comes here authoritatively resolved by the State’s adjudication.” Watts v. Indiana, 338 U. S. 49, 51-52. See also, Gallegos v. Nebraska, 342 U. S. 55, 60-61; Haley v. Ohio, 332 U. S. 596, 597-598. It is equally clear that in cases involving asserted violations of constitutional rights the Court is free to draw its own inferences from established facts, giving due weight to the conclusions of the state court, but not being conclusively bound by them, Ker v. California, 374 U. S. 23; Spano v. New York, 360 U. S. 315.
A distinction between facts and inferences may often be difficult to draw, but the guiding principle for this Court should be that when a question is in doubt and demeanor and credibility of witnesses, or contemporaneous understandings of the parties, have a part to play in its resolution, this Court should be extremely slow to upset a state court’s inferential findings. The impetus for our exercising de novo review of the facts comes from the attitude that unless this Court can fully redetermine the facts of each case for itself, it will be unable to afford complete protection for constitutional rights. But when the “feel” of the trial may have been a proper element in resolving an issue which is unclear on the record, our independent judgment should give way to the greater capability of the state trial court8 in determining whether a constitutional right has been infringed.9 Proper regard for the duality of the American judicial system demands no less.
Federal habeas corpus, which allows a federal court in appropriate circumstances to develop a fresh record, Townsend v. Sain, 372 U. S. 293, provides a far more satisfactory vehicle for resolving such unclear issues, for the judge can evaluate for himself the on-the-spot considerations which no appellate court can estimate with assurance on a cold record. Those considerations are important to the case at bar.
While I agree that the record is not free from all doubt, I believe that the following selected portions of the testimony of one of the arresting officers are sufficient to carry the day for the State’s judgment:
“Q. Did you have reasonable and probable cause to stop this man?
“A. Yes, I did.
“Q. Based on his previous record?
“A. Information and previous record and observation. [Emphasis added.]
“Q. When you left the Station, did you have in mind stopping Mr. Beck?
“A. I had in mind looking for him in the area of East 115th Street and Beulah, stopping him if I did see him make a stop in that area.
“Q. You indicated that you were operating on information?
“A. Yes.
“Q. From whom did you get this information?
“A. I couldn’t divulge that information.
“Q. But someone specifically did relate that information to you?
“A. Yes.
“Q. And you knew who that person was?
“A. Yes.”
It is true that the officer never specifically said “The informant told me that Beck was operating in the area of East 115th Street and Beulah,” but he did testify that he went looking for Beck in that specific area, that he was acting in part on information, and that his information had been related to him by some specific person whose name he felt privileged not to divulge. I find the state court inference reasonable, even on the basis of the admittedly sparse record before us, that the informant told the officer that Beck was operating in the mentioned area.
Furthermore, in reaching this inference, on-the-spot considerations might well have come into play. There appears to have been no lack of common understanding at trial that the informant had given the officer the crucial information. Petitioner argued in the Ohio Supreme Court, “the pattern is obvious, an officer testifies he had information from a confidential source that a particular person is ‘picking up’ numbers in a given area and based on that information they arrest such person ‘on sight’ without a warrant.”10 Judge Zimmerman of the Supreme Court of Ohio found it to be the fact without seeing any need for elaboration. Respondent, in its brief in this Court, assumed it to be the fact.11 And petitioner raised no question as to this inference in either his petition or brief. Indeed the question is raised for the first time, sua sponte, by the Court’s opinion.
On this basis I vote to affirm.
Although it was Judge Zimmerman’s opinion for the Supreme Court of Ohio which articulated the specific finding in question here, that finding must be attributed to the trial court, for we must presume that its conclusion that the arrest was constitutionally permissible was based on the factual findings necessary to support it. If the Court is unwilling to accept this presumption, it should, at least, remand the case to the Ohio courts in order that any question on this score may be set at rest.
See note 1, supra.