(concurring in part and dissenting in part). I agree with my colleagues that the trial court’s refusal to grant a severance constituted an abuse of discretion requiring reversal and a new trial. However, I would go further and hold that the facts presented were not legally sufficient to establish the crime of attempted grand larceny. In my view, upholding an attempt prosecution on these facts requires a drastic departure from our prior case law and, in effect, makes our State’s law virtually indistinguishable from the law of those jurisdictions that have adopted the Model Penal Code. Such a dilution in our State’s requirements for establishing an attempt cannot be justified in light of our strong recent statements in People v Di Stefano (38 NY2d 640) and People v Warren (66 NY2d 831) that the drafters of our Penal Law intended to adhere to the more demanding test established in People v Rizzo (246 NY 334). Accordingly, I dissent from the majority’s ruling to the extent that it permits a retrial on the attempted grand larceny counts.1
*195I. The Legal Standard
To establish an attempt, the prosecution must prove both the requisite intent to commit a specific object crime and an act "which tends to effect the commission of such crime” (Penal Law § 110.00). The standard for determining whether a particular act rose to the level of an attempt is well settled. "The act need not be 'the final one towards the completion of the offense’ * * * but it must 'carry the project forward within dangerous proximity to the criminal end to be attained’ ” (People v Bracey, 41 NY2d 296, 300; see, e.g., People v Sobieskoda, 235 NY 411; People v Werblow, 241 NY 55; People v Collins, 234 NY 355). The requirement of "dangerous proximity” means that "[t]he act or acts must come or advance very near to the accomplishment of the intended crime” (People v Rizzo, 246 NY 334, 337, supra [emphasis supplied]).
In contrast, the drafters of the Model Penal Code have formulated the standard for attempts in terms of taking a "substantial step” toward the completion of the crime. Their purpose in doing so was to "shift the emphasis from what remains to be done — the chief concern of the proximity tests— to what the actor has already done” (ALI Model Penal Code § 5.01, Tent Draft No. 10, comments, at 47 [emphasis in original]). The underlying goal was to "broaden the scope of attempt liability” by permitting prosecution where "the steps already undertaken are substantial”, notwithstanding that "major steps must be taken before the crime can be completed” (id.). However, as we have very recently reaffirmed, New York has not adopted the Model Penal Code drafters’ approach, but has instead elected to adhere to the proximity analysis articulated in People v Rizzo (246 NY 334, 337, supra) and its predecessors (People v Warren, 66 NY2d 831, 833, supra; accord, People v Di Stefano, 38 NY2d 640, 652, supra). Thus, the focus on the actor’s proximity to the completion of the object crime is not merely archaic verbiage. To the contrary, it represents the current state of the law in New York.
Although, as the majority notes, the proper application of the proximity analysis is very much dependent on the facts of the particular case and the manner in which the intended crime is to be carried out, some general principles may be *196discerned. The cases have repeatedly stated, for example, that "dangerous proximity to the criminal end” exists when the defendant’s acts have set in motion a chain of events that are likely to lead to the completion of the crime unless some external force intervenes. Thus, the court stated in Rizzo that "[t]he law * * * considers those acts only as tending to the commission of the crime which are so near to its accomplishment that in all reasonable probability the crime itself would have been committed but for timely interference” (246 NY, at 337). In another formulation, the court indicated that an attempt could not be found where "[t]he force set in motion is neither continuous nor mechanical, and its operation may be broken before the stage of attempt has been attained by the withdrawal or repentance of the guilty intermediary” (People v Werblow, 241 NY 55, 65, supra). To the same effect are People v Mills (178 NY 274, 285 [act "must be such as would naturally effect that (criminal) result, unless prevented by some extraneous cause”]), People v Collins (234 NY 355, 359-360, supra [same]), People v Sullivan (173 NY 122, 135 [attempt may be found where defendants’ acts would have effected object crime "had their design not been frustrated by the presence or interference of (a third party)”]), and People v Sobieskoda (235 NY 411, 419, supra [act constituted attempt if it had "gone to the extent of placing it in (defendants’) power to commit offense unless interrupted, and nothing but such interruption prevented the present commission of the offense”]; see generally, Annotation, Attempts to Commit Offenses of Larceny by Trick, Confidence Game, False Pretenses, and the Like, 6 ALR3d 241, 246, § 2 [a]). Other factors that have been considered are the temporal and geographic proximity between the act and the object crime, the existence of "several contingencies” standing between the act and the object crime (see, e.g., People v Warren, supra, at 833) and the remoteness or immediacy of the act in relation to that crime (People v Rizzo, supra, at 337). Finally, the courts have often relied on the distinction between acts of "mere preparation” and those tending to effect commission of the crime (see, e.g., People v Bracey, supra, at 300; People v Collins, supra, at 359-361). The concept of "mere preparation” has been persuasively defined by other courts as "the devising or arranging the means or measures necessary for the commission of the offense” (e.g., People v Von Hecht, 133 Cal App 2d 25, 38, 283 P2d 764, 773; People v Murray, 14 Cal 159; State v Pollard, 215 La 655, 41 So 2d 465; State v Block, 333 Mo 127, 62 SW2d 428; see, *197Annotation, op. cit, 6 ALR3d, at 246). Moreover, our own court has indicated that to be guilty of an attempt, the accused must have both acquired the wherewithal to commit the object crime and made some direct movement toward the ultimate object (People v Collins, supra, at 360; People v Sullivan, supra, at 135-136). Absent the latter step, the accused is guilty of no more than "mere preparation.”
The application of these standards to the fact patterns in prior cases is instructive. In People v Sullivan (173 NY 122, supra), for example, the court upheld a conviction for murder committed in the course of an attempted burglary, but noted that an attempt might not be found where defendants had merely procured tools for the break-in. Further, the court declined to find an attempted larceny in People v Werblow (241 NY 55, supra), where the defendants had only planned the crime and sent and received cablegrams in furtherance of the scheme. In People v Rizzo (246 NY 334, supra), the defendants had armed themselves, set out in car looking for individual who they knew was carrying payroll money, and stopped at sites where the messenger was expected, but were intercepted by police before they actually encountered the messenger. Despite all of these concrete steps toward the commission of the crime, this court held them insufficient to constitute the crime of attempted robbery because the defendants never came dangerously close to the completion of the object crime (see also, People v Di Stefano, 38 NY2d 640, modfg 45 AD2d 56, supra [same result on similar facts]). Finally, in a very recent decision, People v Warren (66 NY2d 831, supra), we held that an attempted drug-possession prosecution did not lie where the defendants had held two meetings with the police informant, had reached agreement regarding the terms of the sale, and had set up a third meeting for a later time "in a distant parking lot” before they were intercepted by officers who had been secretly watching. In reversing the conviction, we held that the defendants had not come "very near to the accomplishment of the intended crime” (66 NY2d, at 833), since the planned purchase was to take place several hours later at a distant location and several additional steps needed to be taken before the transaction could be consummated.
II. Analysis
As is evident from the foregoing, the set of facts presented here is unlike any other in which a conviction for attempt has *198been upheld.2 Defendants Mahboubian and Sakhai were charged with attempting to commit larceny by obtaining insurance proceeds under false pretenses. The act which formed the basis for this charge was the abortive staged theft of Mahboubian’s collection from the Regency warehouse. However, this act constituted no more than "mere preparation” analogous to the gathering of equipment for a burglary or the obtaining of weapons for the effectuation of a planned robbery. In other words, by staging the warehouse break-in and theft, defendants were, in effect, merely laying the foundation for the crime they planned ultimately to commit: defrauding Lloyd’s of London. They had not yet taken "a step in the direct movement towards th[at] crime”, as the case law requires (People v Collins, supra, at 360 [emphasis supplied]; see, People v Sullivan, supra). Further, the "dangerous proximity” test outlined in People v Rizzo (supra) is unsatisfied because (1) the act on which the People relied, the warehouse break-in, was remote, both in time and place, from the ultimate goal (see, People v Warren, supra; People v Rizzo, supra, at 337); (2) there remained "several contingencies” standing between the act and the effectuation of the crime (see, People v Warren, supra); (3) even after the break-in, defendants still had complete freedom of action and could therefore have easily changed their minds before taking the next step toward their goal; and (4) the act was simply not "such as would naturally effect th[e intended] result [i.e., the wrongful obtaining of insurance money], unless prevented by some extraneous *199cause” (People v Mills, supra, at 285; see, e.g., People v Rizzo, supra, at 337).
The majority’s contrary conclusion seems to ignore these factors, particularly the last. Although the majority stresses, quite correctly, that the last act before interruption need not be the final step toward completion of the crime, the majority fails to address that the last act before interruption here, the break-in, did not even begin to set in motion the chain of events that would inevitably have led to defendants’ fraudulent acquisition of the insurance proceeds if law enforcement authorities had not intervened. To the contrary, far more was required, including the safe disposition of the purportedly stolen goods, the collection of documents to establish the ownership and value of those goods, the filing of a loss claim with the insurance company and, finally, the successful avoidance of detection through the investigation that the insurer would unquestionably have undertaken. Hence, notwithstanding the majority’s assertion to the contrary (majority opn, at 191, quoting People v Sobieskoda, 235 NY 411, 419, supra), defendants’ acts had not " 'gone to the extent of placing it in their power to commit the offense unless interrupted,’ ” except in the sense that any step taken in the direction of their goal would have increased the likelihood of its accomplishment.3
Further, the facts on which the majority relies — that defendants’ completed acts "encompassed the most hazardous and difficult portion of their criminal scheme” and that the nighttime break-in was itself "potentially and immediately dangerous” (majority opn, at 191) — are, in fact, irrelevant in establish*200ing "dangerous proximity” to the object crime. At best, these circumstances support the premise that the steps taken toward the completion of the crime were indeed "substantial.” However, as noted above, the New York view of attempt, in contrast to the Model Penal Code view, does not accept the substantiality of the steps actually taken as the dispositive criterion. Instead, the analysis must focus on the relationship between the already completed acts and the object crime.
Finally, the weight that the majority places on the fact that defendants’ "activities had reached the point where police intervention was called for, lest the burglars escape or the collection disappear” (id., at 191) is puzzling. While that observation would undoubtedly be helpful if the defendants had been charged with attempted theft of the art collection, its relevance in this context, where the ultimate object of the charged attempt crime was the theft of insurance proceeds through fraud, is difficult to discern.
In sum, by upholding an attempt prosecution in these circumstances, the court has significantly diluted the well-established requirements for proving an act rising to the level of an attempt. Moreover, by blurring the important distinction between a "substantial step” toward the completion of the object crime and an act "tend[ing] to effect commission of such crime” (Penal Law § 110.00), the court has brought the law of New York a giant step closer to the less stringent analysis adopted in the Model Penal Code, an analysis which our Legislature has deliberately rejected.
For all of these reasons, I cannot join in the majority’s decision to permit a retrial on the attempted grand larceny counts. Accordingly, I dissent on this aspect of the majority’s holding.
Chief Judge Wachtler and Judges Simons, Hancock, Jr., and Bellacosa concur with Judge Kaye; Judge Titone concurs in part and dissents in part and votes to reverse, dismiss the attempted grand larceny count and order a new trial on the burglary and conspiracy counts in a separate opinion in which Judge Alexander concurs.
In each case: Order reversed, etc.
. I also disagree with the majority’s analysis in relation to the burglary counts, since I agree with defendant Mahboubian that he should not be held accountable for an intent to facilitate his own future crime. However, I do not dissent on the majority’s decision to uphold the burglary counts because *195I conclude that the evidence was sufficient to support the inference that both Mahboubian and Sakhai intended that the premises of the Regency warehouse be damaged in the course of the staged theft, thereby establishing an intent to commit the crime of criminal mischief.
. The only New York case that I have found which seems to be directly on point is People v Rappaport (207 Misc 604), in which a staged theft was discovered before the defendant had the opportunity to make a false insurance claim and obtain the insurance proceeds. In that case, the court held that the crime of attempted larceny had not been established because, at the time he was caught, the defendant’s acts had not "so irrevocably committed him to the scheme of defrauding his insurance company that his purpose would have been accomplished save for his arrest” (id., at 605; see also, People v Trepanier, 84 AD2d 374). Other jurisdictions which have considered similar fact patterns have rejected liability for attempt (see, e.g., In re Schurman, 40 Kan 533, 20 P 277; Commonwealth v Prius, 75 Mass 127; State v Block, 333 Mo 127, 62 SW2d 428; State v Fraker, 148 Mo 143, 49 SW 1017; Nemecek v State, 72 Okla Crim 195, 114 P2d 492; Rex v Robinson, [1915] 2 KB 342). Finally, the cases on which the People rely (People v Vastano, 117 AD2d 637; Steiner v Commissioner of Correction, 490 F Supp 204), as well as cases from other jurisdictions in which liability has been upheld for attempted larceny by insurance fraud (Galbraith v State, 468 NE2d 575 [Ind App]; State v Grubbs, 657 SW2d 380 [Mo Ct App]), involved circumstances in which some step had been taken toward actually filing a loss claim with the insurer.
. Contrary to the majority’s view (majority opn, at 190, n 2), the language in some of the cases requiring forces which are "continuous” and "mechanical” does not, if applied literally, lead to the conclusion that the final step before completion must be accomplished. Rather, the point of this language is to provide a useful framework for analyzing cases where there remained several important steps between the defendants and their criminal goal. Thus, in People v Trepanier (84 AD2d 374, 377), People v Vastano (117 AD2d 637) and Steiner v Commissioner of Correction (490 F Supp 204), three attempted insurance fraud cases on which the People have placed heavy reliance, the defendants were held to have committed the crime of attempt even though no false claim had been filed because they had done everything within their power toward the accomplishment of their goal by placing the means to complete this final step in the hands of their confederates. It is in this context, where the defendants were not "very near to the accomplishment of the intended crime” (People v Rizzo, supra, at 347), that the notion of setting a "continuous” and "mechanical” force in motion comes into play. Of course, no such circumstance is present fhere.