dissenting.
With all deference, the present affidavit seems hopelessly inadequate to me as a basis for a magistrate’s *117informed determination that a search warrant should issue.
We deal with the constitutional right of privacy that can be invaded only on a showing of “probable cause” as provided by the Fourth Amendment. That is a strict standard; what the police say does not necessarily carry the day; “probable cause” is in the keeping of the magistrate. Giordenello v. United States, 357 U. S. 480, 486-487; Johnson v. United States, 333 U. S. 10, 14. Yet anything he says does not necessarily go either. He too is bound by the Constitution. His discretion is reviewable. Aguilar v. Texas, 378 U. S. 108, 111. But unless the constitutional standard of “probable cause” is defined in meticulous ways, the discretion of police and of magistrates alike will become absolute. The present case, illustrates how the mere weight of lengthy and vague recitals takes the place of reasonably probative evidence of the existence of crime.
I.
Investigator Mazaka sought a warrant for the purpose of searching the premises at 148% Coburn Avenue, occupied by respondent and his family, because, he averred, he had reason to believe that there was concealed on the premises an illegal still and other material connected with the manufacture of nontax-paid liquors. The grounds for this belief were recited in 12 paragraphs on an attached sheet, as reproduced in the Appendix to the Court’s opinion, ante, p. 112.
The factual recitals comprise 10 paragraphs, each paragraph setting forth the alleged events of a single day, except that August 24, 1961, is dealt with in two paragraphs. Of these factual recitals more will be said in a moment. The first and last paragraphs of the 12 describe the sources from which the affiant has gained the information set forth in the factual paragraphs. These sources are, according to the first paragraph, three in *118number: (1) “observations made by me”; (2) “information received officially from other Investigators”; and (3) “reports orally made to me [by other investigators] describing the results of their observations and investigation.” In the last paragraph the affiant describes the sources of his information slightly differently: “The foregoing information is based upon personal knowledge and information which has been obtained from Investigators . . . .”
Of the 10 factual paragraphs eight describe trips said to have been made to and from the vicinity of 148% Coburn Avenue by one Garry and one Incardone. On these trips, it is said, there were delivered to the vicinity of 148% Coburn Avenue large quantities of sugar (four deliveries) and empty metal cans (two deliveries, on one of which respondent himself is said to have been a passenger in the car); on one occasion it was observed only that the car was “heavily laden.” It is said that on two occasions Garry and Incardone were seen taking apparently filled cans into Garry’s house, 59 Highland Street, from the Pontiac; on one such occasion the Pontiac, it is said, had been at Coburn Avenue earlier in the day, apparently making a sugar delivery. And, finally, it is averred that on one occasion seemingly filled cans were loaded into the Pontiac near 148% Coburn Avenue, shortly after a delivery of empties to that address.
The “facts” recited in these eight paragraphs, it is said, permit the inference that a still was being operated on respondent’s premises. But are these “facts” really facts? A statement of “fact” is only as credible as its source. Investigator Mazaka evidently believes these statements to be correct; but the magistrate must, of course, know something of the basis of that belief. Nathanson v. United States, 290 U. S. 41. Is the belief of this affiant based on personal observation, or on hearsay, or on hearsay on hearsay? Nowhere in the affidavit is the source *119of these eight paragraphs of information revealed. In each paragraph the alleged events are simply described directly, or else it is said that certain events “were observed.” Scarcely a clue is given as to who the observer might have been. It might have been the affiant, though one would not expect that he would so studiously refrain from revealing that he himself witnessed these events. The observers might have been some other investigators, though the affiant does not say so; yet in the two paragraphs next to be discussed the observers are prominently identified as investigators. Perhaps the ultimate source of most of these statements was one or more private citizens, who were interviewed by investigators, whose reports on these interviews came in due course to Investigator Mazaka, who then composed the affidavit. Perhaps many of the “facts” recited in the affidavit were supplied by an unknown informant over the telephone.
In most instances the language of the affidavit suggests that some investigator witnessed the alleged events. For example, the second paragraph begins: “On or about July 28, 1961, about 6:45 P. M., an observation was made covering a Pontiac automobile owned by one Joseph Garry.” But the presumed investigator who may have been “covering” this automobile is in no way identified. There is no way of knowing whether the report of this alleged observation was made directly to the affiant or whether it went through one or more intermediaries.
Turning now to the remaining two “factual” paragraphs, we find it averred that “Investigators” smelled fermenting mash and heard metallic and other noises in the vicinity of 148% Coburn Avenue. On August 18, it is said, investigators twice smelled mash between 4 and 5 a. m. as they walked on the sidewalk in front of and beside the house at 148 Coburn Avenue, which is apparently the house next to respondent’s. The “Investigators” are not further identified. On August 30 at about 4 a. m., it *120is averred, unidentified investigators detected the odor of fermenting mash while they were “walking on the sidewalk in front of 148 Coburn Avenue.” The source of the odor is again not specified; but sounds heard at the same time, similar to the sounds made by “a motor or a pump,” are stated to have come “from the direction of 148% Coburn Avenue.”
Such is the substance of the affidavit. No particular item of information is identified as within the first-hand knowledge of the affiant. Certain smells and sounds are explicitly described as having been directly perceived by unnamed investigators. The sources of all the other information are left to speculation.
The Court’s unconcern over the failure of the affidavit to identify the sources of the information recited seems based in part on the detailed, lengthy nature of the factual recitals. The Court seems to say that even if we assume that only some small part of the information is. trustworthy, still enough remains to establish probable cause. But I would direct attention to the fact that only one of the 12 paragraphs in this affidavit definitely points the finger of suspicion at 148% Coburn Avenue: that is the paragraph describing the alleged events of August 28, 1961. In every other paragraph the recitals point no more to /^5% Coburn Avenue than they do to 1/+8 Co-burn Avenue. The August 28 paragraph is critical to the finding of the existence of probable cause for the search of 148% Coburn Avenue. Yet the source of the information contained in that paragraph is in no way identified and it is therefore impossible to determine the trustworthiness of that crucial information.
II.
A discussion of the legal principles governing the sufficiency of this affidavit must, unhappily, begin with Draper v. United States, 358 U. S. 307. There an officer *121had been told by an informer, known to the officer to be reliable, that a man of a certain description would get off a certain train with heroin in his possession. The officer met the train, observed a man of that description getting off, and arrested him. The Court held that there was probable cause for the arrest. In Jones v. United States, 362 U. S. 257, the Court applied the holding in Draper to find an affidavit sufficient to establish probable cause for the issuance of a search warrant, even though the facts stated in the affidavit did not rest on the affiant’s personal observations but rather on the observations of another. The Court held that an affidavit could rest on hearsay, “so long as a substantial basis for crediting the hearsay is presented.” Id., at 269. (Emphasis supplied.) In Jones the basis for crediting the informant’s hearsay was: (1) the affiant swore that the informant had previously given information to him which was correct; (2) the affiant had been given corroborating information by other informants; and (3) the affiant was independently familiar with the persons claimed by the informants to be concealing narcotics in their apartment, and he knew them to have admitted to the use of narcotics.
I dissented from the decisions of the Court in these two cases, for the reasons which I set forth most fully in Draper, supra, at 314 et seq. But though I regard these decisions* as taking a view destructive of the guarantees of the Fourth Amendment, they are in any event clearly not dispositive of the present case. As I have already shown, the affidavit here does not set forth a single corroborating *122fact that is sworn to be within the personal knowledge of the affiant. Moreover, there is not a single statement in the affidavit that could not well be hearsay on hearsay or some other multiple form of hearsay.
We are told, however, that it is at least clear that “Investigators” detected the smell of mash in the vicinity of 148% Coburn Avenue. And the Court says: “Observations of fellow officers of the Government engaged in a common investigation are plainly a reliable basis for a warrant applied for by one of their number,” ante, p. 111. But I would make Taylor v. United States, 286 U. S. 1, 6, my starting point, where the Court stated: “Prohibition officers may rely on a distinctive odor as a physical fact indicative of possible crime; but its presence alone does not strip the owner of a building of constitutional guarantees against unreasonable search.” In Johnson v. United States, 333 U. S. 10, 13, the Court explained what the decision in Taylor meant: “That decision held only that odors alone do not authorize a search without warrant. If the presence of odors is testified to before a magistrate and he finds the affiant qualified to know the odor, and it is one sufficiently distinctive to identify a forbidden substance, this Court has never held such a basis insufficient to justify issuance of a search warrant.” (Emphasis supplied.) It is hardly necessary to point out that a magistrate cannot begin to assess the odor-identifying qualifications of persons whose identity is unknown to him. Nor is it necessary to belabor the point that these odors of mash are not ever stated in the affidavit to have emanated from 148% Coburn Avenue.
III.
The Court of Appeals was surely correct when it observed that “the affidavit leaves as a complete mystery the manner in which the Investigators discovered their information.” 324 F. 2d 864, 869. Such being the case, *123I see no way to avoid the conclusion of the majority below: “If hearsay evidence is to be relied upon in the preparation of an affidavit for a search warrant, the officer or attorney preparing such an affidavit should keep in mind that hearsay statements are only as credible as their source and only as strong as their corroboration. And where the source of the information is in doubt and the corroboration by the affiant is unclear, the affidavit is insufficient.” Id., at 869-870. That conclusion states a relatively clear standard of probable cause and is in sharp contrast to the amorphous one upon which today’s decision rests.
In Jones v. United States, supra, this Court forgot, as it forgets again today, that the duty of the magistrate is not delegable to the police. Nathanson v. United States, 290 U. S. 41. It is for the magistrate, not the police, to decide whether there is probable cause for the issuance of the warrant. That function cannot be discharged by the magistrate unless the police first discharge their own, different responsibility: “to evidence what is reliable and why, and not to introduce a hodge-podge under some general formalistic coverall.” 324 F. 2d, at 870. And see Masiello v. United States, 304 F. 2d 399, 401-402. That is the duty of the police — the rest is not for them.
I Avould affirm the decision below.
Tn these cases we might have drawn a clear, unmistakable line and held that hearsay evidence could not support a search warrant. But we did not so hold; instead we held that hearsay was competent for this purpose if there was “a substantial basis” for crediting it, thereby muddying the waters with considerations of corroboration and informer’s reliability. Thus, by forsaking precise standards, the discretion of police and magistrates became less subject to judicial control.