Colorado court says Blakely retroactive to Apprendi

Providing perhaps the biggest development in a big week, I just got news that yesterday the Colorado Court of Appeals in People v. Johnson, No. 03CA2339 (col. App. Apr. 7, 2005) (available here) concluded "that Blakely applies retroactively to the date that Apprendi established its new rule."  Here's the court's analysis:

In People v. Bradbury, 68 P.3d 494 (Colo. App. 2002), a division of this court concluded that Apprendi did not apply retroactively because it "established a new rule" and "imposed a new obligation" upon trial courts.  People v. Bradbury, supra, 68 P.3d at 497. 

We adopt the reasoning in Bradbury and conclude that because Apprendi established a new rule which had the effect of overriding a widespread practice of allowing judges to decide facts used to aggravate sentences, Blakely's interpretation of that rule must necessarily apply retroactively to the date the rule was established.  Writing for the majority in Blakely, Justice Scalia clearly limited the holding back to the date of Apprendi when he wrote: "the relevant 'statutory maximum' is not the maximum sentence a judge may impose after finding additional facts, but the maximum he may impose without any additional findings." Blakely, supra, 124 S.Ct. at 2537.

Because Blakely explains and clarifies Apprendi, we apply it retroactively to defendant's sentence, which was imposed after Apprendi was announced.  We note at least two federal cases which have held that Blakely does not apply retroactively to collateral attacks against convictions. See, e.g., In re Dean, 375 F.3d 1287, 1290 (11th Cir. 2004); United States v. Stoltz, 325 F. Supp. 2d 982, 987 (D. Minn. 2004). In concluding that retroactive application should be made here, we nevertheless do not apply Blakely to collateral attacks against convictions unless those convictions postdated Apprendi.  People v. Dunlap (Colo. App. No. 01CA1082, Sept. 9, 2004).