dissenting.
In reading the court’s opinion and the recent Supreme Court case Carr v. United States, this fact cannot be lost: there are seemingly two statutes at issue here. There is § 2250 as we interpreted it in United States v. Dixon, and as the court continues to interpret it, and then there is § 2250 as the Supreme Court interpreted it in Carr. That being said, I have two principal disagreements with the court’s opinion. The first is that it gives Carr too limited a reading; the second is that its interpretation of § 2250 renders the statute constitutionally defective.
I.
After this case was argued, the Supreme Court handed down Carr v. United States, — U.S. -, 130 S.Ct. 2229, — L.Ed.2d - (2010). In it, the Court overturned our previous interpretation of § 2250 in United States v. Dixon, 551 F.3d 578, 581 (7th Cir.2008), rev. sub nom. Carr v. United States, — U.S. -, 130 S.Ct. 2229, — L.Ed.2d - (2010). In that case, the defendant was convicted in Alabama of rape. In 2005, he moved to Indiana but didn’t register as a sex offender, and he stayed under the radar until 2008, when he was arrested in a bar fight. After his arrest, the authorities learned he was a sex offender and wasn’t registered in Indiana as required under SORNA, which was enacted in 2006.
He was charged with and convicted of violating § 2250. On appeal he challenged his conviction on various grounds, including the fact that using his pre-SORNA travel to convict him violated the ex post facto clause. Looking to the statute’s text, we rejected his argument and read § 2250 to apply to a defendant’s travel regardless of when it took place: “the statute does not require that the defendant’s travel postdate the Act, any more than it requires that the conviction of the sex offense that triggers the registration requirement postdate it.” Id. at 582.
The Supreme Court disagreed. It noted that § 2250 has to be read sequentially, meaning the defendant has to have a duty to register under SORNA; he then has to travel; and his violation has to “culminate] in a post-SORNA failure to register.” Carr, supra at 2233. Ultimately, it avoided the ex post facto argument and held that § 2250 doesn’t apply to pre-enactment travel. Id. at 2242. But it didn’t stop there. The Court also gave some additional commentary on § 2250 in the *332form of “considered dicta.” United States v. Bloom, 149 F.3d 649, 653 (7th Cir.1998).
In Dixon we noted that “[t]he evil at which [§ 2250] is aimed is that convicted sex offenders registered in one state might move to another state, fail to register there, and thus leave the public unprotected.” Dixon, 551 F.3d at 582. We also analogized § 2250 to the felon-in-possession statute, noting it doesn’t matter when the firearm passed state lines, the firearm’s travel is not part of the harm; it is simply a jurisdictional hook. Id.
Again, the Supreme Court viewed § 2250 differently: “the proper analogy is not, as the Seventh Circuit suggested, between the travel of a sex offender and the movement of a firearm; it is between the sex offender who ‘travels’ and the convicted felon who ‘possesses.’ ” Carr, supra at 2240. It also disagreed with our position about the defendant’s travel: the travel is not just “a jurisdictional predicate for § 2250, but it is also, like the act of possession, the very conduct at which Congress took aim.” Id. In that way, it is not enough that a defendant has traveled; he has to travel with a specific purpose because Congress has “subjected such offenders to federal criminal liability only when, after SORNA’s enactment, they use the channels of interstate commerce in evading a State’s reach.” Id. at 2239. Of course, if criminal liability only attaches when the travel is for such a purpose, then the showing of purpose and intent that the government must make is pivotal to the prosecution. The Supreme Court rested this reading of § 2250 on both SORNA’s purpose and its structure: “Taking account of SORNA’s overall structure, we have little reason to doubt that Congress intended § 2250 to do exactly what it says: to subject to federal prosecution sex offenders who elude SORNA’s registration requirements by traveling in interstate commerce.” Id. at 2241.
To be clear, no circuit court applying § 2250 has required the prosecution to prove the purpose of the defendant’s interstate travel. This is probably because as it is written, the statute does not have any language to that effect. But the Supreme Court and the dissent saw eye-to-eye on this point: “I agree with the Court that there is a good argument that § 2250(a) should not be read to apply to such a case, where there is little if any connection between the offender’s prior interstate movement and his subsequent failure to register.” Id. at 2248 (Alito, J., dissenting). It is clear that as far as the Supreme Court is concerned, under § 2250 the defendant’s travel is not just a jurisdictional hook but part of the behavior Congress is regulating. And as an inferior court, we have to abide by it.
With that in mind, I have two points of disagreement with the court’s application of Car)". First, even if Carr is limited to its basic holding, the facts we have here do not satisfy the statute. Second, if we give due deference to Cards “considered dicta,” the facts we have here do not satisfy the statute because there is nothing in the record about why he traveled.
A.
The court and I agree that under Carr § 2250 has to be read sequentially. We just disagree on what that means. I think Carr’s sequential requirement means that the defendant has to have a duty to register under SORNA; he then has to travel; and his violation has to “culminat[e] in a post-SORNA failure to register.” Carr, supra at 2235. Cards sequential reading is not just a checklist for courts. It has a purpose: it assures that there is “a nexus between a defendant’s interstate travel and his failure to register as a sex offend*333er.” Id. at 2235. Thus, “[o]nce a person becomes subject to SORNA’s registration requirements, ... that person can be convicted under § 2250 if he thereafter travels and then fails to register.” Id. at 2235 (emphasis added).
That is not what happened here. From the stipulated record, Vasquez’s only duty to register as a sex offender arose in Illinois, which he failed to do. He didn’t have a duty to register in California — under the sparse stipulated facts, we don’t know how long he was there or that he had changed his residence or any status that would compel him to register in California. 42 U.S.C. § 16913(c). So, his interstate travel did not culminate in his failure to register, nor was it in any way connected to his failure to register. Thus, it was not part of the harm that Congress was addressing, but a mere jurisdictional hook for making this a federal crime.
B.
I also disagree with the court’s treatment of Carr. From the discussion above, it should be clear that the Supreme Court views § 2250 as requiring that some purpose to avoid, evade or elude registering attach to the defendant’s travel; it is not enough that the defendant travels across state lines to run an errand or visit a friend. Here there is nothing in the record about why Vasquez traveled; all we know is that he did. Thus, without any proof concerning why Vasquez traveled to California, his conviction should be overturned.
II
This leads to my second principal disagreement with the court: interpreting the statute the way we did in Dixon and the way the court does here, without § 2250 regulating the defendant’s travel, it is unconstitutional. Granted, that is a significant statement, in light of the fact that our sister circuits have applied the same analysis as the court and found that § 2250 is a legitimate exercise of the Commerce Clause. But against the backdrop of the traditional boundaries that have marked Congress’s power under the Commerce Clause and our interpretation of Lopez, it is clear that § 2250 is not a legitimate exercise of congressional power. And in an effort to uphold it, the court endorses a significant expansion of congressional power.
The plain language of § 2250, without applying Carr’s considered dicta, establishes that the statute only requires that a defendant have traveled interstate at some time. And limiting Carr only to its narrow holding, the time for the travel merely has to be after the statute was enacted. Under either application, the defendant’s travel is not connected to him evading his duty to register under SORNA, and it is not what Congress is regulating.
While the distinction between a person who travels to evade registering and a person who travels and fails to register is semantically slight, it is constitutionally significant. To appreciate the significance of this distinction and understand the error in the court’s Commerce Clause analysis, it is necessary to sketch the traditional limits of Congress’s commerce power. For the past fifteen years, courts have based much of their understanding of the commerce power on the three categories articulated in United States v. Lopez, 514 U.S. 549, 558-59, 115 S.Ct. 1624, 131 L.Ed.2d 626 (1995):
First, Congress may regulate the use of the channels of interstate commerce. See, e.g., Darby, 312 U.S. at 114, 61 S.Ct. 451; Heart of Atlanta Motel, supra, at 256, 85 S.Ct. 348 (“ ‘[T]he authority of Congress to keep the channels of inter*334state commerce free from immoral and injurious uses has been frequently sustained, and is no longer open to question.’ ” (quoting Caminetti v. United States, 242 U.S. 470, 491, 37 S.Ct. 192, 61 L.Ed. 442 (1917))). Second, Congress is empowered to regulate and protect the instrumentalities of interstate commerce, or persons or things in interstate commerce, even though the threat may come only from intrastate activities. See, e.g., Shreveport Rate Cases, 234 U.S. 342, 34 S.Ct. 833, 58 L.Ed. 1341 (1914); Southern R. Co. v. United States, 222 U.S. 20, 32 S.Ct. 2, 56 L.Ed. 72 (1911) (upholding amendments to Safety Appliance Act as applied to vehicles used in intrastate commerce); Perez, supra, at 150, 91 S.Ct. 1357 (“[F]or example, the destruction of an aircraft (18 U.S.C. § 32), or ... thefts from interstate shipments (18 U.S.C. § 659)”).
We have interpreted each category with reference to the citations used. United States v. Wilson, 73 F.3d 675, 686-87 (7th Cir.1995).
A.
In support of the first Lopez category that “Congress may regulate the use of the channels of interstate commerce,” Lopez cites three cases: United States v. Darby, 312 U.S. 100, 113-15, 61 S.Ct. 451, 85 L.Ed. 609 (1941); Heart of Atlanta Motel, Inc. v. United States, 379 U.S. 241, 256, 85 S.Ct. 348, 13 L.Ed.2d 258 (1964); Caminetti v. United States, 242 U.S. 470, 491, 37 S.Ct. 192, 61 L.Ed. 442 (1917). The pertinent discussion in each case cited focuses on Congress’s ability to regulate the misuse of the channels of interstate commerce. In Darby, it was the power to ban goods produced without minimum labor standards from traveling on the channels of interstate commerce. 312 U.S. at 113-15, 61 S.Ct. 451. In Heart of Atlanta, the cited portion concerned Congress’s power to keep the channels of interstate commerce “free from immoral or injurious uses.” 379 U.S. at 256, 85 S.Ct. 348 (quoting Caminetti, 242 U.S. at 491, 37 S.Ct. 192). And in Caminetti, the Court upheld the Mann Act, which “seeks to reach and punish the movement in interstate commerce of women and girls with a view to the accomplishment of the unlawful purposes prohibited.” 242 U.S. at 491, 37 S.Ct. 192 (emphasis added). These cites illustrate Congress’s traditional power to keep the channels of interstate commerce free from misuse. In effect, when Congress does this, it is “excluding] from commerce articles whose use in the states for which they are destined it may conceive to be injurious to the public health, morals or welfare.” Darby, 312 U.S. at 114, 61 S.Ct. 451.
Consistent with this understanding of Congress’s power, we have noted that under the first Lopez category, it can proscribe shipments of stolen goods, kidnaped persons, and prostitutes from traveling on the channels. Wilson, 73 F.3d at 680 n. 5. And we have upheld a child pornography statute because “Congress ha[d] set out to prohibit the interstate movement of a commodity through the channels of interstate commerce.” United States v. Schaffner, 258 F.3d 675, 680-81 (7th Cir.2001) (citing Lopez, 514 U.S. at 559, 115 S.Ct. 1624); see also United States v. Kenney, 91 F.3d 884, 889 (7th Cir.1996). These regulations focus on the movement across state lines with an illicit purpose. Hoke v. United States, 227 U.S. 308, 322, 33 S.Ct. 281, 57 L.Ed. 523 (1913). That is, the crime is complete once the offending person or good has moved interstate. Caminetti, 242 U.S. at 491, 37 S.Ct. 192.
It is not, as the court interprets Lopez, a matter of the defendant having traveled for some innocent purpose and then later committing the crime. See Mortensen v. *335United States, 322 U.S. 369, 374, 64 S.Ct. 1037, 88 L.Ed. 1331 (1944) (“To constitute a violation of the Act, it is essential that the interstate transportation have for its object or be the means of effecting or facilitating the proscribed activities.”). A person’s mere travel across state lines does not give Congress authority to later regulate all of his future conduct — or in this case, make his previous failure to register in Illinois a federal crime. The Tenth Circuit made this same point in United States v. Patton, where it observed that the first Lopez “category is confined to statutes that regulate interstate transportation itself, not manufacture before shipment or use after shipment.” 451 F.3d 615, 621 (10th Cir.2006) (McConnell, J.).
In contrast to the statutes that properly regulate a person’s travel across the channels of interstate commerce, under § 2250 the court separates the defendant’s travel from the crime of failing to register. And that renders it constitutionally problematic. The Supreme Court may have tacitly recognized this in Carr when it noted that under § 2250 Congress “subjected such offenders to federal criminal liability only when, after SORNA’s enactment, they use the channels of interstate commerce in evading a State’s reach.” Carr, supra at 2238. Under Carr the focus is, as it should be, on the sex offender’s misuse of the channels of interstate commerce: using them to evade registration. And unless we interpret the statute as Carr did, § 2250 is not a permissible use of congressional power over the channels of interstate commerce.1
B.
The opinion goes beyond the first Lopez category and also upholds § 2250 under the second category, noting that Vasquez “was undeniably ‘a person ... in interstate commerce’ when he moved from Illinois to California, and traveled to California via the ‘channels of interstate commerce.’ ” Op. at 330. Under that category, “Congress is empowered to regulate and protect the instrumentalities of interstate commerce, or persons -or things in interstate commerce, even though the threat may come only from intrastate activities.” Lopez, 514 U.S. at 558, 115 S.Ct. 1624. The court interprets this category to mean that because a person travels across state lines, he is “a person in interstate commerce” and Congress can regulate him thereafter.
But that literal and expansive interpretation of the second Lopez category is contrary to our circuit’s precedent. There are two parts to that category: “instrumentalities of interstate commerce” and “persons or things in interstate commerce, even though the threat may come only from intrastate activities.” The first part is selfexplanatory.2 It is the second part concerning “persons or things” that is problematic for courts.
*336In the midst of this rather arcane area of the law, it is important to remember that Lopez doesn’t stand for a radical enlargement of Congress’s power under the first two categories, but rather an enforcement of limits under the third — the activities that substantially affect interstate commerce. And its three categories should be interpreted as a convenient rhetorical formulation for summarizing Congress’s traditional power over commerce. See United States v. Rybar, 103 F.3d 273, 286-89 (3d Cir.1996) (Alito, J., dissenting).
In defining that clause previously, we noted that the “inclusion of the language ‘persons and things’ was likely based on precedent — not happenstance.” Wilson, 73 F.3d at 687. And the key to understanding that language is the Supreme Court’s citation to Perez v. United States; the pertinent language in Perez is where the Court notes: “The commerce clause reaches .... protection of the instrumentalities of interstate commerce, as for example, the destruction of an aircraft (18 U.S.C. § 32), or persons or things in commerce, as, for example, thefts from interstate shipments (18 U.S.C. § 659).” 402 U.S. 146, 150, 91 S.Ct. 1357, 28 L.Ed.2d 686 (1971). Lopez cites to these same statutes. 514 U.S. at 558, 115 S.Ct. 1624 (citing Perez, 402 U.S. at 150, 91 S.Ct. 1357 (citing 18 U.S.C. §§ 32, 659)). Essentially, what Lopez did was define this power by looking to the explication given in Perez. Understood in this way, the phrase “persons or things in interstate commerce” clearly refers to and must be defined by the laws that Congress can pass to protect the persons or things that the instrumentalities are moving. Wilson, 73 F.3d at 687.
With that understanding, we have expressed reservation that videotape cassettes that have moved across state lines are “things in interstate commerce.” United States v. Angle, 234 F.3d 326, 337 n. 12 (7th Cir.2000). We have also avoided using the second category to uphold legislation that criminalized interfering with an abortion facility simply because the pregnant women have traveled there. In doing so, we noted that “[hjolding that the Access Act qualifies as a regulation of an instrumentality of interstate commerce based on a literal reading of one sentence in Lopez ... is unnecessary without further guidance from the Supreme Court.” Wilson, 73 F.3d at 687 n. 12. This makes sense because the second Lopez category involves “things actually being moved in interstate commerce, not all people and things that have ever moved across state lines.” Patton, 451 F.3d at 622. Thus, it is wholly inconsistent with our precedent to uphold the constitutionality of § 2250 as regulating sex-offenders as a “person ... in interstate commerce.”
C.
Unlike the other circuits to address this question, the court also cites Scarborough v. United States, 431 U.S. 563, 577, 97 S.Ct. 1963, 52 L.Ed.2d 582 (1977) and the “minimal nexus” reasoning as a basis to uphold § 2250 under the Commerce Clause.3 Notably, none of the other circuits has directly relied on Scarborough and its minimal nexus test to uphold § 2250 as it applies to persons who have traveled interstate. In doing so, the court is recognizing a power Congress never had, and doing so without giving deference to the reasoning in Cair.
*337Scarborough created the legal fiction that once a gun has crossed state lines it is forever “in or affecting” commerce and Congress can prohibit felons from possessing them — this is described as “a minimal nexus.” 431 U.S. at 575, 97 S.Ct. 1963. We have also used the logic of a “minimal nexus” or “limited nexus” to uphold the constitutionality of the ear-jacking statute. United States v. Taylor, 226 F.3d 593, 600 (7th Cir.2000). Although this test seems to work when applied to things, such as guns and cars, there are four problems with extending the minimal or limited nexus rationale to persons.
First, while Scarborough is still good law as far as its reasoning goes in felon-in-possession cases, it has been implicitly criticized by the Supreme Court in the commercial arson context in Jones v. United States, 529 U.S. 848, 857, 120 S.Ct. 1904, 146 L.Ed.2d 902 (2000). We have recognized that criticism and refused to extend the minimal nexus test to materials. United States v. Craft, 484 F.3d 922, 927 (7th Cir.2007). And other courts and scholars have noted the problems inherent in Scarborough’s reasoning and extending it to other circumstances. United States v. Bishop, 66 F.3d 569, 593-600 & n. 13 (3d Cir.1995) (Becker, J., concurring in part and dissenting in part); see also Dean A. Strang, Felons, Guns, and the Limits of Federal Power, 39 J. Marshall L.Rev. 385 (2006); United States v. Chesney, 86 F.3d 564, 577-82 (6th Cir.1996) (Batchelder, J., concurring).
Second, there is a logical distinction between guns and persons that can’t be lost in applying Scarborough here. The cases endorsing the “minimal nexus” test concerned things — commodities that were included in the actual makeup of commerce. But persons are different: we are not inherently commercial; we cannot be bought or sold; and our participation in commerce is limited to our decision to engage in it. Consistent with this distinction, in the felon-in-possession context it is the gun that has crossed state lines; it is not enough that the felon has crossed state lines and subsequently possesses a gun that has remained intrastate. See United States v. Travisano, 724 F.2d 341, 347-48 (2d Cir.1983).
Third, the Supreme Court in Carr looked at Scarborough and the minimal nexus rubric and noted that § 2250 is distinguishable: “Understanding the act of travel as an aspect of the harm Congress sought to punish serves to distinguish § 2250 from the felon-in-possession statute to which the Seventh Circuit analogized.” Carr, supra at 2239. Indeed, the Court went on to note that analogizing this to Scarborough is inappropriate: “In this case, the proper analogy is not, as the Seventh Circuit suggested, between the travel of a sex offender and the movement of a firearm; it is between the sex offender who ‘travels’ and the convicted felon who ‘possesses.’ ” Id.
Fourth, expanding Scarborough will obliterate the limits between what is local and what is national. In striking down the Violence Against Women Act in United States v. Morrison, the Supreme Court observed that if the “aggregated impact” rationale under the third Lopez category were adopted, it would allow Congress to “regulate murder or any other type of violence” and even reach issues including “family law and other areas of traditional state regulation.” 529 U.S. 598, 616, 120 S.Ct. 1740, 146 L.Ed.2d 658 (2000). While Congress could not regulate those areas under its broadest power in Morrison, it could under the court’s rationale here. By applying the minimal nexus to a person’s travel, Congress could take over the states’ ability to punish domestic crimes.
For instance, under 18 U.S.C. § 2262 (interstate violation of a protective order), *338the government would no longer have to prove the defendant traveled in interstate commerce with the intent to violate a protective order. It would only have to show that the defendant had at some time traveled across state lines, regardless of his purpose, and that at some time later he violated a protective order. If this were true, Congress could effectively take over the monitoring and control of local, domestic crime, by making an element of the crime that the person has traveled interstate at some time. That, however, stands in complete contradiction to Morrison. As the Supreme Court aptly noted: “The regulation and punishment of intrastate violence that is not directed at the instrumentalities, channels, or goods involved in interstate commerce has always been the province of the States.” 529 U.S. at 618-19, 120 S.Ct. 1740 (citation omitted).
Thus, I believe that consistent with a government of limited powers and in light of the Supreme Court’s precedent in this area, we should not extend Scarborough’s “minimal nexus” beyond firearms to reach persons in an effort to find that this statute comes under the Commerce Clause.
D.
When § 2250 is applied in the way it was in Dixon and by the court here, it emphasizes the need to apply Carr’s considered dicta to the statute and require a showing that the defendant’s travel was with an illicit intent to evade, elude, or avoid registering. The alternative is an unconstitutional statute.
III.
This leads to my final point of disagreement: this is a specific-intent crime. That is true applying the statute either as we did in Dixon or as the Supreme Court did in Ccmt. If we follow Carr’s reasoning and the purpose of the travel is vital to the statute that naturally forces the government to prove that the defendant had an elevated intent. It is not enough to travel and negligently or through ignorance fail to register.
To address the court’s position on this point, interpreting the statute apart from Carr this is a specific-intent crime. The court looks to United States v. Cox, 577 F.3d 833, 836 (7th Cir.2009), and its reasoning in support of interpreting § 2250 as a general-intent crime. In Cox the statute at issue proscribed conduct that was already unlawful: transporting someone across state lines to become a prostitute. But it added the element that the person be under 18. We held that under the statute the government doesn’t have to prove that the defendant knew he was transporting a minor, which makes sense, given the strict liability that normally attaches to sexual acts with minors.
But the statute at issue here does not proscribe inherently unlawful conduct; rather, it requires that the defendant must register. He has an affirmative, administrative duty — one that he must perform or be imprisoned. Thus, it is reasonable that he would have to know about the duty before he is held accountable. Nothing suggests that Congress intended to hold someone responsible for knowingly failing to do something without any evidence that he knew what he was supposed to do.
Moreover, the court’s position that we can transfer knowledge between a state-imposed duty and a federal duty is difficult to reconcile with the basic concepts of justice and our precedent. See United States v. Pulungan, 569 F.3d 326, 331 (7th Cir.2009). Nothing in the statute defines the § 2250 obligation with reference to the Illinois obligation. They are distinct. And there is no reason to think that Vasquez’s known legal duty under Illinois law should transfer to his federal obligation. In short, just because Vasquez knew about his state duty to register we cannot uphold *339his conviction because we assume he was aware of his federal duty to register.
IV.
In sum, there are two statutes here: § 2250 as it is written and as we have interpreted it pre-Carr, and as the Supreme Court has interpreted it in Carr. Taking § 2250 as it is written, the statute is unconstitutional because it does not require interstate travel with the intent to avoid or evade registration under SORNA. Under the Supreme Court’s reasoning in Carr, however, the statute passes muster constitutionally because it regulates the defendant’s travel, by attaching criminal liability to sex offenders who travel interstate to evade registration. And applying the reasoning in Carr, we would have to overturn Vasquez’s conviction because there is no proof of why he traveled. I also believe that both the grammatical structure of § 2250 and its context counsel reading this as a specific-intent crime. For these reasons, I must respectfully dissent.
. In effect, the Supreme Court’s opinion in Carr verifies the defect I’ve identified in § 2250 by incorporating this additional requirement into the statute and placing the statute’s focus beyond its text and onto the defendant’s travel on the channels of interstate commerce.
. In support of the second category, Lopez cited three cases, one of which, Perez, cited two statutes. The first two cases concerned railroads, which are actual instrumentalities of interstate commerce. Houston, E. & W. Tex. Ry. v. United States (Shreveport Rate Cases), 234 U.S. 342, 34 S.Ct. 833, 58 L.Ed. 1341 (1914); Southern Ry. Co. v. United States, 222 U.S. 20, 32 S.Ct. 2, 56 L.Ed. 72 (1911). Specifically, they concerned Congress's ability to set rates and standards for railroads. Shreveport Rate Co., 234 U.S. at 351-53, 34 S.Ct. 833; see also Southern Ry. Co., 222 U.S. at 26, 32 S.Ct. 2.
. The opinion does not suggest that § 2250 can be upheld under the third Lopez category, which only comes into play with economic activity that substantially affects interstate commerce. That is not at issue here.