I dissent with reluctance, since I realize fully that in the long run it will make little difference whether this prosecution aborts because of a defective arrest warrant or the painfully obvious Miranda-Honeycutt violation disclosed by the record.
In brief, I believe that the court’s opinion draws too heavily on the law relating to declarations against interest as an exception to the hearsay rule, while giving insufficient weight to the facts of this particular case and the very limited purpose for which Martinez’ identification of Campa was used. People v. Leach (1975) 15 Cal.3d 419, 439 [124 Cal.Rptr. 752, 541 P.2d 296] condemned the use of nondisserving portions of declarations against interest at trial, on the issue of guilt. Here we are concerned with probable cause justifying an arrest and our only point of contact with the exception to the hearsay rule is the commonsense notion that when people make statements which are clearly against their interest, the chance that they are telling the truth is comparatively good.
What the majority ignores when talking in the abstract about declarants in police custody who seek to exculpate themselves by implicating others, is that in this case the material furnished the magistrate permitted him to infer that Martinez had admitted everything that was his to admit. The police report, which was incorporated in the affidavit in support of the warrant, contained a statement from a “citizen” informant that it was the passenger in the van, not the driver, who had done the shooting. The magistrate was also informed that a previous warrant, issued by another magistrate, had *887resulted in the recovery of the van, as well as the arrest of Martinez. From these facts alone it is reasonably inferrable that Martinez was the driver of the van—which is precisely what Martinez admitted after his arrest. Thus his reliability does not depend solely on the incriminating nature of his admission but rests also on the fact that the extent of the admission corresponds to the apparent degree of his actual involvement in the shooting, as reflected in independent information obtained by the police.
There is no question that under Leach, Martinez’ extrajudicial identification of Campa as the triggerman would not be admissible because it was not “specifically disserving to the interest of the declarant.” (People v. Leach, supra, 15 Cal.3d at p. 441.) As already noted, however, we are not concerned with admitting an incriminating statement at trial which, if believed, could send a man to prison: our sole concern is whether the statement, viewed in context, is sufficient to give rise to probable cause to make an arrest. Thus I would grant that if nothing submitted to the magistrate negatived the possibility that Martinez was trying to exculpate himself by implicating Campa as the triggerman, the warrant should not have issued.1 That, however, was not the case. The magistrate could infer from all that was submitted to him that Martinez was, indeed, only the driver, precisely as he had admitted. His reliability as an informant was at least as great as that of the typical narcotics user who, for whatever reason, has correctly fingered other users or dealers in the past: while few would stake their lives on the accuracy of his next bit of news about one of his fellow junkies, we have held time after time that as long as he purports to speak with personal knowledge, his say-so is sufficient to authorize the issuance of a warrant. (People v. De Santiago (1969) 71 Cal.2d 18, 21-22 [76 Cal.Rptr. 809, 453 P.2d 353].)
*888I would reverse.
Mosk, J., concurred.
In making this concession, I appear to differ with Professor LaFave whose views concerning the veracity of persons in Martinez’ position are a good deal more sanguine than those of the court: “It is true, of course, that these individuals typically provide information after they have been apprehended by the police and after it is apparent to them that the police already know of their own involvement in the serious offense. It might be said, as the Supreme Court said of accomplice accusations in another context, that this information is ‘inevitably suspect, . . . given the recognized motivation to shift blame onto others.’ [Bruton v. United States (1968) 391 U.S. 123 (20 L.Ed.2d 476, 88 S.Ct. 1620).] This is a legitimate cause for concern, but it is well to note that the danger is less here than in the at-trial setting about which the Court was speaking. For one thing, a fabrication of false accusations purporting to involve others in one’s criminal activities is less likely to occur on the spur of the moment following arrest than upon that later occasion. (Indeed, it may be stated as a general proposition that a ‘spontaneous’ admission against penal interest is more deserving of reliance.) For another, the effort to ‘shift blame’ is more likely to amount to an overstatement of the influence and control wielded by actual accomplices than the naming of persons who were in fact not involved at all, and these misstatements, even if they occur, are not likely to distort the facts bearing directly upon the probable cause determination.” (Fns. omitted.) (LaFave, Search and Seizure (1978) § 3.3, pp. 527-528.)