Justice KAGAN delivered the opinion of the Court.

In Padilla v. Kentucky, 559 U.S. 356, 130 S.Ct. 1473, 176 L.Ed.2d 284 (2010), this Court held that the Sixth Amendment requires an attorney for a criminal defendant to provide advice about the risk of deportation arising from a guilty plea. We consider here whether that ruling applies retroactively, so that a person whose conviction became final before we decided Padilla can benefit from it. We conclude that, under the principles set out in Teague v. Lane, 489 U.S. 288, 109 S.Ct. 1060, 103 L.Ed.2d 334 (1989), Padilla does not have retroactive effect.

I

Petitioner Roselva Chaidez hails from Mexico, but became a lawful permanent resident of the United States in 1977. About 20 years later, she helped to defraud an automobile insurance company out of $26,000. After federal agents uncovered the scheme, Chaidez pleaded guilty to two counts of mail fraud, in violation of 18 U.S.C. § 1341. The District Court sentenced her to four years of probation and ordered her to pay restitution. Chaidez's conviction became final in 2004.

Under federal immigration law, the offenses to which Chaidez pleaded guilty are "aggravated felonies," subjecting her to mandatory removal from this country. See 8 U.S.C. §§ 1101(a)(43)(M)(i), 1227(a)(2)(A)(iii). But according to Chaidez, her attorney never advised her of that fact, and at the time of her plea she remained ignorant of it.

Immigration officials initiated removal proceedings against Chaidez in 2009, after an application she made for citizenship alerted them to her prior conviction. To avoid removal, Chaidez sought to overturn that conviction by filing a petition for a writ of coram nobis in Federal District Court.1 She argued that her former attorney's failure to advise her of the immigration consequences of pleading guilty constituted ineffective assistance of counsel under the Sixth Amendment.

While Chaidez's petition was pending, this Court decided Padilla. Our ruling vindicated Chaidez's view of the Sixth Amendment: We held that criminal defense attorneys must inform non-citizen clients of the risks of deportation arising from guilty pleas. See 559 U.S., at ___, 130 S.Ct. at 1482. But the Government argued that Chaidez could not benefit from Padilla because it announced a "new rule" and, under Teague, such rules do not apply in collateral challenges to already-final convictions.

The District Court determined that Padilla "did not announce a new rule for Teague purposes," and therefore should apply to Chaidez's case. 730 F.Supp.2d 896, 904 (N.D.Ill.2010). It then found that Chaidez's counsel had performed deficiently under Padilla and that Chaidez suffered prejudice as a result. Accordingly, the court vacated Chaidez's conviction. See No. 03 CR 636-6, 2010 WL 3979664 (N.D.Ill., Oct. 6, 2010).

The United States Court of Appeals for the Seventh Circuit reversed, holding that Padilla had declared a new rule and so should not apply in a challenge to a final conviction. "Before Padilla," the Seventh Circuit reasoned, "the [Supreme] Court had never held that the Sixth Amendment requires a criminal defense attorney to provide advice about matters not directly related to [a] client's criminal prosecution," including the risks of deportation. 655 F.3d 684, 693 (2011). And state and lower federal courts had uniformly concluded that an attorney need not give "advice concerning [such a] collateral (as opposed to direct) consequenc[e] of a guilty plea." Id., at 690. According to the Seventh Circuit, Padilla's holding was new because it ran counter to that widely accepted "distinction between direct and collateral consequences." 655 F.3d, at 691. Judge Williams dissented. Agreeing with the Third Circuit's view, she argued that Padilla "broke no new ground" because it merely applied established law about a lawyer's "duty to consult" with a client. 655 F.3d, at 695 (quoting United States v. Orocio, 645 F.3d 630, 638-639 (C.A.3 2011) (internal quotation marks omitted)).

We granted certiorari, 566 U.S. ___, 132 S.Ct. 2101, 182 L.Ed.2d 867 (2012), to resolve a split among federal and state courts on whether Padilla applies retroactively.2 Holding that it does not, we affirm the Seventh Circuit.

II

Teague makes the retroactivity of our criminal procedure decisions turn on whether they are novel. When we announce a "new rule," a person whose conviction is already final may not benefit from the decision in a habeas or similar proceeding.3 Only when we apply a settled rule may a person avail herself of the decision on collateral review. Here, Chaidez filed her coram nobis petition five years after her guilty plea became final. Her challenge therefore fails if Padilla declared a new rule.

"[A] case announces a new rule," Teague explained, "when it breaks new ground or imposes a new obligation" on the government. 489 U.S., at 301, 109 S.Ct. 1060. "To put it differently," we continued, "a case announces a new rule if the result was not dictated by precedent existing at the time the defendant's conviction became final." Ibid. And a holding is not so dictated, we later stated, unless it would have been "apparent to all reasonable jurists." Lambrix v. Singletary, 520 U.S. 518, 527-528, 117 S.Ct. 1517, 137 L.Ed.2d 771 (1997).

But that account has a flipside. Teague also made clear that a case does not "announce a new rule, [when] it `[is] merely an application of the principle that governed'" a prior decision to a different set of facts. 489 U.S., at 307, 109 S.Ct. 1060 (quoting Yates v. Aiken, 484 U.S. 211, 217, 108 S.Ct. 534, 98 L.Ed.2d 546 (1988)). As JUSTICE KENNEDY has explained, "[w]here the beginning point" of our analysis is a rule of "general application, a rule designed for the specific purpose of evaluating a myriad of factual contexts, it will be the infrequent case that yields a result so novel that it forges a new rule, one not dictated by precedent." Wright v. West, 505 U.S. 277, 309, 112 S.Ct. 2482, 120 L.Ed.2d 225 (1992) (concurring in judgment); see also Williams v. Taylor, 529 U.S. 362, 391, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000). Otherwise said, when all we do is apply a general standard to the kind of factual circumstances it was meant to address, we will rarely state a new rule for Teague purposes.

Because that is so, garden-variety applications of the test in Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), for assessing claims of ineffective assistance of counsel do not produce new rules. In Strickland, we held that legal representation violates the Sixth Amendment if it falls "below an objective standard of reasonableness," as indicated by "prevailing professional norms," and the defendant suffers prejudice as a result. Id., at 687-688, 104 S.Ct. 2052. That standard, we later concluded, "provides sufficient guidance for resolving virtually all" claims of ineffective assistance, even though their particular circumstances will differ. Williams, 529 U.S., at 391, 120 S.Ct. 1495. And so we have granted relief under Strickland in diverse contexts without ever suggesting that doing so required a new rule. See, e.g., ibid.; Rompilla v. Beard, 545 U.S. 374, 125 S.Ct. 2456, 162 L.Ed.2d 360 (2005); Wiggins v. Smith, 539 U.S. 510, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003).4 In like manner, Padilla would not have created a new rule had it only applied Strickland's general standard to yet another factual situation — that is, had Padilla merely made clear that a lawyer who neglects to inform a client about the risk of deportation is professionally incompetent.

But Padilla did something more. Before deciding if failing to provide such advice "fell below an objective standard of reasonableness," Padilla considered a threshold question: Was advice about deportation "categorically removed" from the scope of the Sixth Amendment right to counsel because it involved only a "collateral consequence" of a conviction, rather than a component of the criminal sentence? 559 U.S., at ___, 130 S.Ct. at 1482.5 In other words, prior to asking how the Strickland test applied ("Did this attorney act unreasonably?"), Padilla asked whether the Strickland test applied ("Should we even evaluate if this attorney acted unreasonably?"). And as we will describe, that preliminary question about Strickland's ambit came to the Padilla Court unsettled — so that the Court's answer ("Yes, Strickland governs here") required a new rule.

The relevant background begins with our decision in Hill v. Lockhart, 474 U.S. 52, 106 S.Ct. 366, 88 L.Ed.2d 203 (1985), which explicitly left open whether advice concerning a collateral consequence must satisfy Sixth Amendment requirements. Hill pleaded guilty to first-degree murder after his attorney misinformed him about his parole eligibility. In addressing his claim of ineffective assistance, we first held that the Strickland standard extends generally to the plea process. See Hill, 474 U.S., at 57, 106 S.Ct. 366. We then determined, however, that Hill had failed to allege prejudice from the lawyer's error and so could not prevail under that standard. See id., at 60, 106 S.Ct. 366. That conclusion allowed us to avoid another, more categorical question: whether advice about parole (however inadequate and prejudicial) could possibly violate the Sixth Amendment. The Court of Appeals, we noted, had held "that parole eligibility is a collateral rather than a direct consequence of a guilty plea, of which a defendant need not be informed." Id., at 55, 106 S.Ct. 366. But our ruling on prejudice made "it unnecessary to determine whether there may be circumstances under which" advice about a matter deemed collateral violates the Sixth Amendment. Id., at 60, 106 S.Ct. 366.6

That non-decision left the state and lower federal courts to deal with the issue; and they almost unanimously concluded that the Sixth Amendment does not require attorneys to inform their clients of a conviction's collateral consequences, including deportation. All 10 federal appellate courts to consider the question decided, in the words of one, that "counsel's failure to inform a defendant of the collateral consequences of a guilty plea is never" a violation of the Sixth Amendment. Santos-Sanchez v. United States, 548 F.3d 327, 334 (C.A.5 2008).7 That constitutional guarantee, another typical decision expounded, "assures an accused of effective assistance of counsel in `criminal prosecutions'"; accordingly, advice about matters like deportation, which are "not a part of or enmeshed in the criminal proceeding," does not fall within the Amendment's scope. United States v. George, 869 F.2d 333, 337 (C.A.7 1989). Appellate courts in almost 30 States agreed.8 By contrast, only two state courts held that an attorney could violate the Sixth Amendment by failing to inform a client about deportation risks or other collateral consequences of a guilty plea.9 That imbalance led the authors of the principal scholarly article on the subject to call the exclusion of advice about collateral consequences from the Sixth Amendment's scope one of "the most widely recognized rules of American law." Chin & Holmes, Effective Assistance of Counsel and the Consequences of Guilty Pleas, 87 Cornell L.Rev. 697, 706 (2002).10

So when we decided Padilla, we answered a question about the Sixth Amendment's reach that we had left open, in a way that altered the law of most jurisdictions — and our reasoning reflected that we were doing as much. In the normal Strickland case, a court begins by evaluating the reasonableness of an attorney's conduct in light of professional norms, and then assesses prejudice. But as earlier indicated, see supra, at 1107-1108, Padilla had a different starting point. Before asking whether the performance of Padilla's attorney was deficient under Strickland, we considered (in a separately numbered part of the opinion) whether Strickland applied at all. See 559 U.S., at ___, 130 S.Ct., at 1480-1482. Many courts, we acknowledged, had excluded advice about collateral matters from the Sixth Amendment's ambit; and deportation, because the consequence of a distinct civil proceeding, could well be viewed as such a matter. See id., at ___, 130 S.Ct., at 1480-1481. But, we continued, no decision of our own committed us to "appl[y] a distinction between direct and collateral consequences to define the scope" of the right to counsel. Id., at ___, 130 S.Ct., at 1481. And however apt that distinction might be in other contexts, it should not exempt from Sixth Amendment scrutiny a lawyer's advice (or non-advice) about a plea's deportation risk. Deportation, we stated, is "unique." Ibid. It is a "particularly severe" penalty, and one "intimately related to the criminal process"; indeed, immigration statutes make it "nearly an automatic result" of some convictions. Ibid. We thus resolved the threshold question before us by breaching the previously chink-free wall between direct and collateral consequences: Notwithstanding the then-dominant view, "Strickland applies to Padilla's claim." Id., at ___, 130 S.Ct., at 1482.

If that does not count as "break[ing] new ground" or "impos[ing] a new obligation," we are hard pressed to know what would. Teague, 489 U.S., at 301, 109 S.Ct. 1060. Before Padilla, we had declined to decide whether the Sixth Amendment had any relevance to a lawyer's advice about matters not part of a criminal proceeding. Perhaps some advice of that kind would have to meet Strickland's reasonableness standard — but then again, perhaps not: No precedent of our own "dictated" the answer. Teague, 489 U.S., at 301, 109 S.Ct. 1060. And as the lower courts filled the vacuum, they almost uniformly insisted on what Padilla called the "categorica[l] remov[al]" of advice about a conviction's non-criminal consequences — including deportation — from the Sixth Amendment's scope. 559 U.S., at ___, 130 S.Ct., at 1482. It was Padilla that first rejected that categorical approach — and so made the Strickland test operative — when a criminal lawyer gives (or fails to give) advice about immigration consequences.11 In acknowledging that fact, we do not cast doubt on, or at all denigrate, Padilla. Courts often need to, and do, break new ground; it is the very premise of Teague that a decision can be right and also be novel. All we say here is that Padilla's holding that the failure to advise about a non-criminal consequence could violate the Sixth Amendment would not have been — in fact, was not — "apparent to all reasonable jurists" prior to our decision. Lambrix, 520 U.S., at 527-528, 117 S.Ct. 1517. Padilla thus announced a "new rule."

III

Chaidez offers, and the dissent largely adopts, a different account of Padilla, in which we did no more than apply Strickland to a new set of facts. On Chaidez's view, Strickland insisted "[f]rom its inception" that all aspects of a criminal lawyer's performance pass a test of "`reasonableness under prevailing professional norms'": The decision thus foreclosed any "categorical distinction between direct and collateral consequences." Brief for Petitioner 21-22 (emphasis deleted) (quoting Strickland, 466 U.S., at 688, 104 S.Ct. 2052). Indeed, Chaidez contends, courts prior to Padilla recognized Strickland's all-encompassing scope and so applied its reasonableness standard to advice concerning deportation. See Brief for Petitioner 25-26; Reply Brief 10-12. She here points to caselaw in three federal appeals courts allowing ineffective assistance claims when attorneys affirmatively misled their clients about the deportation consequences of guilty pleas.12 The only question left for Padilla to resolve, Chaidez claims, was whether professional norms also require criminal lawyers to volunteer advice about the risk of deportation. In addressing that issue, she continues, Padilla did a run-of-the-mill Strickland analysis. And more: It did an especially easy Strickland analysis. We had earlier noted in INS v. St. Cyr, 533 U.S. 289, 121 S.Ct. 2271, 150 L.Ed.2d 347 (2001) — a case raising an issue of immigration law unrelated to the Sixth Amendment — that a "competent defense counsel" would inform his client about a guilty plea's deportation consequences. Id., at 323, n. 50, 121 S.Ct. 2271. All Padilla had to do, Chaidez concludes, was recite that prior finding.

But Chaidez's (and the dissent's) story line is wrong, for reasons we have mostly already noted: Padilla had to develop new law, establishing that the Sixth Amendment applied at all, before it could assess the performance of Padilla's lawyer under Strickland. See supra, at 1107-1108, 1110. Our first order of business was thus to consider whether the widely accepted distinction between direct and collateral consequences categorically foreclosed Padilla's claim, whatever the level of his attorney's performance. We did not think, as Chaidez argues, that Strickland barred resort to that distinction. Far from it: Even in Padilla we did not eschew the direct-collateral divide across the board. See 559 U.S., at ___, 130 S.Ct., at 1481 ("Whether that distinction is [generally] appropriate is a question we need not consider in this case"). Rather, we relied on the special "nature of deportation" — the severity of the penalty and the "automatic" way it follows from conviction — to show that "[t]he collateral versus direct distinction [was] ill-suited" to dispose of Padilla's claim. Id., at ___, 130 S.Ct., at 1482. All that reasoning came before we conducted a Strickland analysis (by examining professional norms and so forth), and none of it followed ineluctably from prior law.13

Predictably, then, the caselaw Chaidez and the dissent cite fails to support their claim that lower courts "accepted that Strickland applied to deportation advice." Brief for Petitioner 25; see post, at 1118-1120. True enough, three federal circuits (and a handful of state courts) held before Padilla that misstatements about deportation could support an ineffective assistance claim. But those decisions reasoned only that a lawyer may not affirmatively misrepresent his expertise or otherwise actively mislead his client on any important matter, however related to a criminal prosecution. See, e.g., United States v. Kwan, 407 F.3d 1005, 1015-1017 (C.A.9 2005). They co-existed happily with precedent, from the same jurisdictions (and almost all others), holding that deportation is not "so unique as to warrant an exception to the general rule that a defendant need not be advised of the [collateral] consequences of a guilty plea." United States v. Campbell, 778 F.2d 764, 769 (C.A.11 1985).14 So at most, Chaidez has shown that a minority of courts recognized a separate rule for material misrepresentations, regardless whether they concerned deportation or another collateral matter. That limited rule does not apply to Chaidez's case. And because it lived in harmony with the exclusion of claims like hers from the Sixth Amendment, it does not establish what she needs to — that all reasonable judges, prior to Padilla, thought they were living in a Padilla-like world.

Nor, finally, does St. Cyr have any relevance here. That decision stated what is common sense (and what we again recognized in Padilla): A reasonably competent lawyer will tell a non-citizen client about a guilty plea's deportation consequences because "`[p]reserving the client's right to remain in the United States may be more important to the client than any potential jail sentence.'" Padilla, 559 U.S., at ___, 130 S.Ct., at 1483 (quoting St. Cyr, 533 U.S., at 322, 121 S.Ct. 2271). But in saying that much, St. Cyr did not determine that the Sixth Amendment requires a lawyer to provide such information. Courts had held to the contrary not because advice about deportation was insignificant to a client — really, who could think that, whether before or after St. Cyr? — but because it concerned a matter collateral to the criminal prosecution.15 On those courts' view, the Sixth Amendment no more demanded competent advice about a plea's deportation consequences than it demanded competent representation in the deportation process itself. Padilla decided that view was wrong. But to repeat: It was Padilla that did so. In the years following St. Cyr, not a single state or lower federal court considering a lawyer's failure to provide deportation advice abandoned the distinction between direct and collateral consequences, and several courts reaffirmed that divide. See, e.g., Santos-Sanchez, 548 F.3d, at 335-336; Broomes v. Ashcroft, 358 F.3d 1251, 1256-1257 (C.A.10 2004); United States v. Fry, 322 F.3d 1198, 1200-1201 (C.A.9 2003). It took Padilla to decide that in assessing such a lawyer's performance, the Sixth Amendment sets the standard.16

IV

This Court announced a new rule in Padilla. Under Teague, defendants whose convictions became final prior to Padilla therefore cannot benefit from its holding. We accordingly affirm the judgment of the Court of Appeals for the Seventh Circuit.

It is so ordered.

Justice THOMAS, concurring in the judgment.

In Padilla v. Kentucky, 559 U.S. 356, 130 S.Ct. 1473, 176 L.Ed.2d 284 (2010), this Court held that the Sixth Amendment requires an attorney for a criminal defendant to apprise his client of the risk of deportation created by a guilty plea. I dissented. The Sixth Amendment provides that "[i]n all criminal prosecutions," an accused enjoys the right "to have the Assistance of Counsel for his defence." By its terms, this right extends "to legal advice directly related to defense against prosecution of the charged offense," and "[t]here is no basis in text or in principle" to expand the reach of this guarantee to guidance concerning the collateral consequences of a guilty plea. Id., at ___, 130 S.Ct., at 1495 (SCALIA, J., dissenting). Today, the Court finds that Padilla announced a new rule of constitutional law and that, under our decision in Teague v. Lane, 489 U.S. 288, 109 S.Ct. 1060, 103 L.Ed.2d 334 (1989), "defendants whose convictions became final prior to Padilla therefore cannot benefit from its holding." Ante, at 1113. I continue to believe that Padilla was wrongly decided and that the Sixth Amendment does not extend — either prospectively or retrospectively — to advice concerning the collateral consequences arising from a guilty plea. I, therefore, believe that the Teague analysis is unnecessary and thus concur only in the judgment.

Justice SOTOMAYOR, with whom Justice GINSBURG joins, dissenting.

The Court holds today that Padilla v. Kentucky, 559 U.S. 356, 130 S.Ct. 1473, 176 L.Ed.2d 284 (2010), announced a "new" rule within the meaning of Teague v. Lane, 489 U.S. 288, 301, 109 S.Ct. 1060, 103 L.Ed.2d 334 (1989), and so does not apply to convictions that became final before its announcement. That is wrong, because Padilla did nothing more than apply the existing rule of Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), in a new setting, the same way the Court has done repeatedly in the past: by surveying the relevant professional norms and concluding that they unequivocally required attorneys to provide advice about the immigration consequences of a guilty plea. Because Padilla fell squarely within the metes and bounds established by Strickland, I respectfully dissent.

I

A

The majority correctly sets forth the governing legal principles under Teague and Strickland. Ante, at 1107-1108. The Teague inquiry turns centrally on the "nature of the rule" in question, and for that reason, "[w]here the beginning point is a rule of ... general application, ... it will be the infrequent case that yields a result so novel that it forges a new rule." Wright v. West, 505 U.S. 277, 308-309, 112 S.Ct. 2482, 120 L.Ed.2d 225 (1992) (KENNEDY, J., concurring in judgment); see ante, at 1107-1108. The majority makes the important observation that "when all we do is apply a general standard to the kind of factual circumstances it was meant to address, we will rarely state a new rule." Ibid. It makes sense, then, that "garden-variety applications of ... Strickland... do not produce new rules." Ante, at 1107.

In Strickland, we did not provide a comprehensive definition of deficient performance, and instead held that "[t]he proper measure of attorney performance remains simply reasonableness under prevailing professional norms." 466 U.S., at 688, 104 S.Ct. 2052. Strickland's reasonableness prong therefore takes its content from the standards by which lawyers judge their professional obligations, ibid., and those standards are subject to change. That is why, despite the many different settings in which it has been applied, we have never found that an application of Strickland resulted in a new rule.1

Significantly, we have previously found that applications of Strickland to new factual scenarios are not barred under 28 U.S.C. § 2254(d)(1) of the Antiterrorism and Effective Death Penalty Act (AEDPA). Section 2254(d)(1) precludes habeas relief unless a state court decision violates "clearly established Federal law," which, as relevant here, largely overlaps with the inquiry under Teague of whether a decision was "dictated by precedent." 489 U.S., at 301, 109 S.Ct. 1060 (plurality opinion).2 In Wiggins v. Smith, 539 U.S. 510, 522, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003), for example, we found that Williams v. Taylor, 529 U.S. 362, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000), "made no new law" when it held that Strickland extended to an attorney's responsibility to conduct a background investigation in a capital case. Rather, we explained that "in referring to the ABA Standards for Criminal Justice as guides, [Williams] applied the same `clearly established' precedent of Strickland we apply today." 539 U.S., at 522, 123 S.Ct. 2527. Similarly, in Lafler v. Cooper, 566 U.S. ___, ___, ___-___, 132 S.Ct. 1376, 1385-1386, 1390-1391, 182 L.Ed.2d 398 (2012), we rejected the argument advanced by the Solicitor General that the Sixth Amendment did not extend to advice about a plea offer because it did not impact the fairness of the trial. Instead, we simply held that Strickland applied to this form of attorney misconduct.

In short, where we merely apply Strickland in a way that corresponds to an evolution in professional norms, we make no new law.

B

Contrary to the majority's reconstruction, Padilla is built squarely on the foundation laid out by Strickland. Padilla relied upon controlling precedent. It began by reciting the basic rule that "[u]nder Strickland, we first determine whether counsel's representation `fell below an objective standard of reasonableness.'" Padilla, 559 U.S., at ___, 130 S.Ct., at 1482 (quoting Strickland, 466 U.S., at 688, 104 S.Ct. 2052). We recognized that "[t]he first prong — constitutional deficiency — is necessarily linked to the practice and expectations of the legal community: `[t]he proper measure of attorney performance remains reasonableness under prevailing professional norms.'" Padilla, 559 U.S., at ___, 130 S.Ct., at 1482 (quoting Strickland, 466 U.S., at 688, 104 S.Ct. 2052).

We therefore examined the substantial changes in federal immigration law that provided the backdrop to the relevant professional standards. Padilla, 559 U.S., at ___, 130 S.Ct., at 1478-1480. Pursuant to the Immigration Act of 1917, 39 Stat. 889-890, a judge could recommend that a defendant who had committed a deportable offense not be removed from the country. Congress entirely eliminated this procedure in 1990. 104 Stat. 5050. Then the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (IIRIRA), 110 Stat. 3009-596, abolished the Attorney General's authority to grant discretionary relief from removal for all but a small number of offenses. Padilla, 559 U.S., at ___, 130 S.Ct., at 1480. These changes in immigration law meant that for a noncitizen who committed a removable offense, "removal [had become] practically inevitable." Ibid.

In parallel with these developments, the standards of professional responsibility relating to immigration had become more demanding. "For at least the past 15 years," we observed in Padilla, "professional norms have generally imposed an obligation on counsel to provide advice on the deportation consequences of a client's plea." Id., at ___, 130 S.Ct., at 1485. Citing an array of practice guides and professional responsibility manuals, we noted that "[t]he weight of prevailing professional norms supports the view that counsel must advise her client regarding the risk of deportation." Id., at ___, 130 S.Ct., at 1482. Indeed, "authorities of every stripe — including the American Bar Association, criminal defense and public defender organizations, authoritative treatises, and state and city bar publications — universally require defense attorneys to advise as to the risk of deportation consequences for non-citizen clients." Id., at ___, 130 S.Ct., at 1482 (internal quotation marks omitted).

We drew further support for our conclusion that professional standards required advice about deportation consequences from our decision in INS v. St. Cyr, 533 U.S. 289, 121 S.Ct. 2271, 150 L.Ed.2d 347 (2001). See Padilla, 559 U.S., at ___, 130 S.Ct., at 1482-1483 (citing St. Cyr, 533 U.S., at 323, 121 S.Ct. 2271). In St. Cyr, we had explained that the availability of discretionary relief from removal was critical to a noncitizen's decision to accept a plea offer, and expected counsel to follow the instructions of "numerous practice guides," such as the ABA's Standards for Criminal Justice, to inform themselves of the possible immigration consequences of a plea. Padilla, 559 U.S., at ___, 130 S.Ct., at 1483 (citing St. Cyr, 533 U.S., at 323, n. 50, 121 S.Ct. 2271); see id., at 322, n. 48, 121 S.Ct. 2271. And we there found that many States already required that a trial judge advise defendants of the same. Ibid. St. Cyr thus "recognized that `preserving the client's right to remain in the United States may be more important to the client than any potential jail sentence.'" Padilla, 559 U.S., at ___, 130 S.Ct., at 1483 (quoting St. Cyr, 533 U.S., at 322, 121 S.Ct. 2271).

Our application of Strickland in Padilla followed naturally from these earlier observations about changes in immigration law and the accompanying evolution of professional norms. When we decided St. Cyr and Padilla, nothing about Strickland's substance or applicability had changed. The only difference from prior law was that the underlying professional norms had changed such that counsel's failure to give this advice now amounted to constitutionally deficient performance.3 Both before Padilla and after, counsel was obligated to follow the relevant professional norms. It was only because those norms reflected changes in immigration law that Padilla reached the result it did, not because the Sixth Amendment right had changed at all.

II

A

Accepting that routine applications of Strickland do not result in new rules, the majority nevertheless holds that Padilla went a step further. In its view, Padilla "`br[oke] new ground'" by addressing the threshold question of whether advice about deportation is a collateral consequence of a criminal conviction that falls within the scope of the Sixth Amendment. Ante, at 1110-1111. But that is wrong, because Padilla declined to embrace the very distinction between collateral and direct consequences of a criminal conviction that the majority says it did. In fact, the Court stated very clearly that it found the distinction irrelevant for the purposes of determining a defense lawyer's obligation to provide advice about the immigration consequences of a plea. 559 U.S., at ___, n. 8, 130 S.Ct., at 1480-1481, n. 8. We asserted that we had "never applied a distinction between direct and collateral consequences to define the scope of constitutionally `reasonable professional assistance' required under Strickland," and concluded that "[w]hether that distinction is appropriate is a question we need not consider in this case." Id., at ___, 130 S.Ct., at 1481 (emphasis added). The distinction was "ill suited" to the task at hand, we explained, because deportation has a "close connection to the criminal process," and is "uniquely difficult to classify as either a direct or a collateral consequence." Id., at ___, 130 S.Ct., at 1482). Indeed, "[o]ur law ha[d] enmeshed criminal convictions and the penalty of deportation for nearly a century," and we had "long recognized" that deportation is "particularly severe." Id., at ___, 130 S.Ct., at 1481.4

At bottom, then, the majority's argument hinges upon a distinction the Court has never embraced and that Padilla found irrelevant to the issue it ultimately decided. Without this revision to our recent decisional history, the majority's analysis unravels.

B

The majority finds that the "legal landscape," Graham v. Collins, 506 U.S. 461, 468, 113 S.Ct. 892, 122 L.Ed.2d 260 (1993), before Padilla was nearly uniform in its rejection of Strickland's application to the deportation consequences of a plea. Ante, at 1108-1111. It concludes that the lower courts were generally in agreement that the Sixth Amendment did not require attorneys to inform clients of the collateral consequences of a plea, and that this weighs heavily in favor of finding that Padilla announced a new rule. Ante, at 1109, nn. 7, 8. But the majority's discussion of these precedents operates at too high a level of generality and fails to account for the development of professional standards over time. St. Cyr noted the importance of advising clients about immigration consequences was of recent vintage, indeed more recent than some of the cases the majority cites. See 533 U.S., at 322-323, 121 S.Ct. 2271. The Court relies upon decisions issued over a period that spans more than 30 years. See ante, at 1109, nn. 7, 8. Nearly half of them (17) were decided before the enactment of IIRIRA. See ibid. And all but two of the Federal Court of Appeals cases were decided before St. Cyr. See ante, at 1109, nn. 7, 8. These earlier decisions show nothing more than that the underlying professional norms had not yet evolved to require attorneys to provide advice about deportation consequences.

Cases from the period following IIRIRA and St. Cyr undermine the majority's generalizations about the state of the law before Padilla. Deportation had long been understood by lower courts to present "the most difficult" penalty to classify as either a collateral or direct consequence. United States v. Russell, 686 F.2d 35, 38 (C.A.D.C. 1982); cf. Janvier v. United States, 793 F.2d 449, 455 (C.A.2 1986) (holding that Strickland applied to advice about a judicial recommendation against deportation). Eventually, and in parallel with changes in federal immigration law and the corresponding professional norms, the lower courts had acknowledged an important qualification to the collateral consequences rule. After the passage of IIRIRA and this Court's decision in St. Cyr, many courts concluded that a lawyer's affirmative misstatements about the immigration consequences of a guilty plea can constitute deficient performance under Strickland. Indeed, each Federal Court of Appeals to address the question after St. Cyr so held. See United States v. Couto, 311 F.3d 179, 188 (C.A.2 2002); United States v. Kwan, 407 F.3d 1005, 1015 (C.A.9 2005); cf. Downs-Morgan v. United States, 765 F.2d 1534, 1540-1541 (C.A.11 1985).5 State-court decisions from this period were in accord and relied upon similar reasoning.6

These decisions created an important exception to the collateral/direct consequences distinction. They also foreshadowed the Court's reasoning in Padilla by basing their analysis of the relevant professional norms on the special nature of deportation, the ABA standards governing immigration practice, and the Court's assessment of those standards in St Cyr. See Kwan, 407 F.3d, at 1016 ("That counsel may have misled [the defendant] out of ignorance is no excuse. It is a basic rule of professional conduct that a lawyer must... [remain] abreast of changes in the law and its practice.... Counsel's performance... fell below the [ABA]'s ethical standard for criminal defense attorneys with respect to immigration consequences. The Supreme Court noted this standard in [St. Cyr]"); Couto, 311 F.3d, at 187-191 (citing St. Cyr and the relevant ABA standards, and concluding that "recent Supreme Court authority supports [a] broader view of attorney responsibility" that encompasses affirmative misrepresentations about deportation consequences); see also Downs-Morgan, 765 F.2d, at 1541 ("[D]eportation and exclusion [are] harsh consequences").

The majority believes that these decisions did not meaningfully alter the state of the law in the lower courts before Padilla, because they merely applied the age-old principle that a lawyer may not affirmatively mislead a client. Ante, at 1111-1112. But, as explained, the reasoning of these cases renders that characterization at best incomplete. See, e.g., Kwan, 407 F.3d, at 1016. While these lower court precedents are consistent with the general principle that attorneys should not mislead clients by providing incorrect advice, they did not rest primarily on that rule. Rather, they recognized the significant changes in professional norms that predated Padilla and that we had noted in St. Cyr. As a consequence, the "wall between direct and collateral consequences" that the lower courts had erected, ante, at 1110, had already been dealt a serious blow by the time the Court decided Padilla.

As the majority points out, these misrepresentation cases stopped short of imposing an affirmative obligation on lawyers to consult with clients about the consequences of deportation. Ante, at 1111-1112. But the majority places too much emphasis on the absence of lower court authority finding that an attorney's omissions with respect to deportation resulted in ineffective assistance. The distinction between omissions and affirmative misrepresentations on which these lower court cases depended cannot be reconciled with Strickland. In Padilla itself, we rejected the Solicitor General's suggestion that Strickland should apply to advice about the immigration consequences of a plea only in cases where defense counsel makes an affirmative misstatement. Padilla, 559 U.S., at ___, 130 S.Ct., at 1483-1484. We did so because we found that Strickland was incompatible with the distinction between an obligation to give advice and a prohibition on affirmative misstatements. 559 U.S., at ___, 130 S.Ct., at 1483-1484 (citing Strickland, 466 U.S., at 690, 104 S.Ct. 2052). Strickland made clear that its standard of attorney performance applied to both "acts" and "omissions," and that a rule limiting the performance inquiry to one or the other was too narrow. 466 U.S., at 690, 104 S.Ct. 2052. Thus, the distinction between misrepresentations and omissions, on which the majority relies in classifying lower court precedent, implies a categorical rule that is inconsistent with Strickland's requirement of a case-by-case assessment of an attorney's performance.7 Id., at 688-689, 104 S.Ct. 2052; see, e.g., Roe v. Flores-Ortega, 528 U.S. 470, 479, 120 S.Ct. 1029, 145 L.Ed.2d 985 (2000). In short, that some courts have differentiated between misleading by silence and affirmative misrepresentation hardly establishes the rationality of the distinction. Notably, the Court offers no reasoned basis for believing that such a distinction can be extracted from Strickland.

To be sure, lower courts did continue to apply the distinction between collateral and direct consequences after St. Cyr. See ante, at 1112-1113; see, e.g., Broomes v. Ashcroft, 358 F.3d 1251, 1256-1257 (C.A.10 2004). Even so, and even assuming the misrepresentation cases did not call the distinction into question, the existence of these lower court decisions is not dispositive. "[T]he standard for determining when a case establishes a new rule is `objective,' and the mere existence of conflicting authority does not necessarily mean a rule is new." Wright, 505 U.S., at 304, 112 S.Ct. 2482 (O'Connor, J., concurring in judgment) (citing Stringer v. Black, 503 U.S. 222, 237, 112 S.Ct. 1130, 117 L.Ed.2d 367 (1992)); see Graham v. Collins, 506 U.S. 461, 506, 113 S.Ct. 892, 122 L.Ed.2d 260 (1993) (Souter, J., dissenting).

Where the application of Strickland was straightforward, rooted in 15 years of professional standards and the Court's prior St. Cyr decision, there is no reason to put these lower court cases, many from more than a decade earlier, ahead of this Court's simple and clear reasoning in Padilla. Nevertheless, the majority reaches the paradoxical conclusion that by declining to apply a collateral-consequence doctrine the Court had never adopted, Padilla announced a new rule.

III

What truly appears to drive the majority's analysis is its sense that Padilla occasioned a serious disruption in lower court decisional reasoning. See, e.g., ante, at 1110 ("If that does not count as `break[ing] new ground' ... we are hard pressed to know what would" (quoting Teague, 489 U.S., at 301, 109 S.Ct. 1060)). The concurring and dissenting opinions in Padilla similarly reflected the impression that it was a significant and destabilizing decision. See 559 U.S., at ___, 130 S.Ct., at 1488 (ALITO, J., concurring in judgment); id., at ___, 130 S.Ct., at 1497 (SCALIA, J., dissenting) (describing the majority opinion as a "sledge hammer"); ante, at 1110, n. 10. But the fact that a decision was perceived as momentous or consequential, particularly by those who disagreed with it, does not control in the Teague analysis. Faithfully applying the Teague rule depends instead on an examination of this Court's reasoning and an objective assessment of the precedent at issue. Stringer, 503 U.S., at 237, 112 S.Ct. 1130. In Padilla, we did nothing more than apply Strickland. By holding to the contrary, today's decision deprives defendants of the fundamental protection of Strickland, which requires that lawyers comply with professional norms with respect to any advice they provide to clients.

* * *

Accordingly, I would reverse the judgment of the Seventh Circuit and hold that Padilla applies retroactively on collateral review to convictions that became final before its announcement. With respect, I dissent.