Amicus Curiae Brief — Lockyer v. Andrade

No. 01-1127 } FILED

IN THE

Supreme Court of the United Pe

BILL LOCKYER, Attorney General of the State of California;

ERNST B. Roe, Warden,

Petitioners,

VS.

LEANDRO ANDRADE,

Respondent.

On Writ of Certiorari to the

United States Court of Appeals for the Ninth Circuit

BRIEF OF THE CRIMINAL JUSTICE LEGAL

FOUNDATION; HONORABLE BILL JONES,

SECRETARY OF STATE OF CALIFORNIA;

CALIFORNIA STATE SENATORS DICK

ACKERMAN, JIM BATTIN, JIM BRULTE, RAY

HAYNES, ROSS JOHNSON, BRUCE McPHERSON,

DICK MONTIETH; AND CALIFORNIA STATE

ASSEMBLYMAN DENNIS HOLLINGSWORTH

AS AMICI CURIAE IN SUPPORT OF PETITIONERS

KENT S. SCHEIDEGGER

CHARLES L. HOBSON

Counsel of Record

Criminal Justice Legal Fdn.

2131 L Street

Sacramento, CA 95816

Phone: (916) 446-0345

Fax: (916) 446-1194

E-mail: cjlf@cjlf.org

Attorneys for Amicus Curiae

Criminal Justice Legal Foundation

:

rs

E

ra

a

E

?

a

A

S

b

:

!

tence so arbitrary as

f the relevant body of

?

prison sentences

ion o

ity of

2. Was the state court’s appli

law to the defendant’s challenge to his

to exist outside the universe of plausible outcomes and there-

fore unreasonable under 28 U.S. C. § 2254(d)?

to the proporti

(i)

TABLE OF CONTENTS

The AEDPA issue should be resolved first, as a

EEE Ee

The California court correctly identified and applied

the few clearly established principles in an unsettled

EPEC eres eseceseccoceccccccoscce

(iii)

The California court’s application of what little clearly

established law there is in this area was reasonable ... .

If this Court subsequently invalidates part of

California’s three strikes scheme, the defendant can

obtain relief in the California courts even if his claim

is rejected in the present case

Conclusion

eeeseeoeereeeeeeeeeeeeeeeeeeeeeeeeeee eee s

22

TABLE OF AUTHORITIES

Cases

Anderson v. Charles, 447 U. S. 404, 65 L. Ed. 2d 222,

Se as Gk. SUP OUUEED oc cccccccescessévccecececess 26

Andrade v. Roe, 270 F. 3d 743 (CA9 2001) .... 2,3, 17, 18

Bell v. Cone, 535 U.S. __

(No. 01-400, May 28, 2002) .................. 3, 6, 22

Bell v. Jarvis, 236 F. 3d 149 (CA4 2000) ............. 23

Caspari v. Bohlen, 510 U. S. 383, 127 L. Ed. 2d 236,

PE, CEE Sc décncacccndcequcncessncts 19

Christianson v. Colt Industries Operating Corp..

486 U.S. 800, 100 L. Ed. 2d 811.

ce cavcacgudasaedeeseeseces 25

Coleman v. Thompson, 501 U. S. 722,

115 L. Ed. 2d 640, 111 S. Ct. 2546(1991) ......... 29

Doyle v. Ohio, 426 U. S. 610, 49 L. Ed. 2d 91,

Fe STEED cavccecndccodécdccucsceesect 25

Edwards v. State, 800 So. 2d 454 (Miss. 2001) ......... 19

Estelle v. Smith, 451 U. S. 454, 68 L. Ed. 2d 359,

Be Gh et, EEE condoncevcescuncescescucnas 6

Francis S. v. Stone, 221 F. 3d 100 (CA2 2000) ......... 23

Gusik v. Schilder, 340 U. S. 128, 95 L. Ed. 146,

Be Ge, CIEE who cncqcccecescccensecess 29, 30

Hall v. Washington, 106 F. 3d 742 (CA7 1997) ........ 25

Harmelin v. Michigan, 501 U. S. 957, 115 L. Ed. 2d 836,

111 S. Ct. 2680 (1991) .............4.. 4, 7, 12-17, 20

Hawkins v. Hargett, 200 F. 3d 1279 (CA10 1999) ... 18,19

Hutto v. Davis, 454 U. S. 370, 70 L. Ed. 2d 556,

PE CE ak ntdue dade dbutatetducndd 9,27

In re Harris, 5 Cal. 4th 813, 21 Cal. Rptr. 2d 373,

855 P. 2d 391 (1993) ...... :6bdebubadedoantedcée 29

In re Lynch, 8 Cal. 3d 410, 105 Cal. Rptr. 217,

SP EEE Suxheedelhes ddutddcadsecwece 21

Kibbe v. Dubois, 269 F. 3d 26 (CAI 2001)......... 25, 26

Marks v. United States, 430 U. S. 188,

51 L. Ed. 2d 260, 97 S. Ct. 990 (1977) ............. 16

McCollough v. Singletary, 967 F. 2d 530

DE Ueiddthéseeebbbacdensnscesunnetes 18, 19

McGruder v. Puckett, 954 F. 2d 313 (CAS 1992) ....... 18

Neal v. Grammer, 975 F. 2d 463 (CA8 1992) .......... 18

O’Brien v. Dubois, 145 F. 3d 16 (CAI 1998) .......... 24

Penry v. Johnson, 532 U. S. 782, 150 L. Ed. 2d 9,

a ED anteatvecaéennes 3, 5, 6, 7, 20, 22

Penry v. Lynaugh, 492 U. S. 302. 106 L. Ed. 2d 256,

PE Gontcacoudccaedddctaedt 7,29

People v. Clark, 5 Cal. 4th 750, 21 Cal. Rptr. 2d 509,

Pee PEED So encuseciesdsheessdedeccdes 29

People v. Cooper, 43 Cal. App. 4th 815,

Se Oe CUED kbodenncdeueedécodcsus 19

People v. Gibson, 90 Cal. App. 4th 371,

108 Cal. Rptr. 2d 809 (2001) ................0008. 18

People v. Hindson, 703 N. E. 2d 956 (Ill. App. 1998)... . 19

vii

People v. King, 16 Cal. App. 4th 567,

BOGE, TR. BE ACESS ccccevcivesesscccccces 21

People v. Superior Court (Romero), 13 Cal.¢th 497,

53 Cal. Rptr. 2d 789, 917 P. 2d 628 (1996) .......... 28

People v. Williams, 17 Cal. 4th 148,

69 Cal. Rptr. 2d 917, 948 P. 2d 429 (1998) .......... 28

Rose v. Lundy, 455 U. S. 509, 71 L. Ed. 2d 379,

Ph CD vchateesecavecauccséecnses 30

Rummel v. Estelle, 445 U. S. 263, 63 L. Ed. 2d 382,

SL a, SPEED bbe wedesccoescensecess Passim

Slack v. McDaniel, 529 U. S. 473, 146 L. Ed. 2d 542.

~ ) eR. Tree Jeeccvecccess 30

Smallwood v. Scott, 73 F. 3d 1343 (CAS 1996) ........ 19

Solem v. Helm, 463 U. S. 277, 77 L. Ed. 2d 637,

SP Ge, Fee GED ccccccccccccecpecceses Passim

State v. Bacon, 702 A. 2d 116 (Vt. 1997) ............. 18

State v. Bonner, 577 N. W. 2d 575 (SD 1998) ...... 18, 20

State v. Brown, 825 P. 2d 482 (Idaho 1992) ........ 18, 19

State v. Four Jugs of Intoxicating Liquors,

FR SR PPT TTTTITIT TTT Tri ee 27

State v. Harris, 844 S. W. 2d 601 (Tenn. 1992)......... 18

State v. Jones, 543 S. E. 2d 541 (SC 2001) ............ 18

State v. Lara, 580 N. W. 2d 783 (lowa 1998) .......... 19

State v. Lee, 841 S. W. 2d 648 (Mo. 1992) ............ 19

State v. Lindsey, 770 So. 2d 339 (La. 2000) ........... 19

State v. Oliver, 745 A. 2d 1165 (NJ 2000) ............ 18

Vili

State v. Riley, 497 N. W. 2d 23 (Neb. 1993) ........... 19

State v. Scott, 961 P. 2d 667 (Kan. 1998) ............. 19

State v. Thorp, 2 P. 3d 903 (Or. App. 2000) ........... 18

Stone v. Powell, 428 U. S. 465, 49 L. Ed. 2d 1067,

a SG bea dnvedee seedacnesesnceds 24

Strickland v. Washington, 466 U. S. 668,

80 L. Ed. 2d 674, 104 S. Ct. 2052 (1984) ............ 6

Teague v. Lane, 489 U. S. 288, 103 L. Ed. 2d 334,

ee SEED ovuadeeusesiecadseseus 5, 6, 29

Terry v. Ohie 392 U.S. 1, 20 L. Ed. 2d 889,

ee 24

Thompson v. Keohane, 516 U. S. 99, .

133 L. Ed. 2d 383, 116 S. Ct. 457(1995) ........... 20

Tuan Van Tran v. Lindsey, 212 F. 3d 1143

DE ndkenebaseeedaeéeckdaceansscceune 5, 23

United States v. Bucuvalas, 970 F. 2d 937 (CA1 1992) .. 19

United States v. Kratsas, 45 F. 3d 63 (CA4 1995)....... 19

United States v. Rutherford, 54 F. 3d 370 (CA7 1995) ... 27

United States v. Sarbello, 985 F. 2d 716

Pt OPE kc bac cedcenvancedsedessocesosses 17,19

Vernonia Sch. Dist. 47J v. Acton, 515 U. S. 646,

132 L. Ed. 2d 564, 115 S. Ct. 2386 (1995) .......... 24

Weems v. United States, 217 U. S. 349, 54 L. Ed. 793,

Pe DOP EIU onincdeensdhdencccuesecneves 7

Williams v. Taylor, 529 U. S. 362, 146 L. Ed. 2d 389,

120 S. Ct. 1495 (2000) ............... 6, 19, 22, 23, 24

ix

United States Constitution

U. S. Const... Amdt. 4 0.0... cece cece eee eens

United States Statutes

I . ., cseundconsoueuses

Ne i deacuaeks

ne

28 U.S.C. §2244(dM INC)... ec cece cece cece ees

State Statutes

Cal. Penal Code § 484 .......... 06 cee eee e eens

Cal. Penal Code § 487 .........:c cece eeeeencees

Cal. Penal Code § 490 ........50 cee ee eee eeeeee

Cal. Penal Code § 666 .........5- 5-0 e cece cence

Cal. Penal Code §1170.12(b)(1) .... 6. - eee eee

Cal. Penal Code § 1192.7(c)(18) ..... 0-606 e cress

Cal. Penal Code § 1385 .........-0 cece ee eeeees

Miscellaneous

American Heritage Dictionary (3d ed. 1992).......

Scheidegger, Habeas Corpus, Relitigation, and the

Legislative Power, 98 Colum. L. Rev. 888 (1998)...

U.S. Sentencing Guidelines §4B1 ........---+-.

IN THE

Supreme Court of the United States

BILL LOCKYER, Attorney General of the State of California;

ERNST B. Roe, Warden,

Petitioners,

VS.

LEANDRO ANDRADE,

Respondent.

BRIEF OF THE CRIMINAL JUSTICE LEGAL

FOUNDATION, ET AL., AS AMICI CURIAE

IN SUPPORT OF PETITIONERS

INTEREST OF AMICI CURIAE

The Criminal Justice Legal Foundation (CJLF)' is a non-

profit California corporation organized to participate in

litigation relating to the criminal justice system as it affects the

public interest. CJLF seeks to bring the due process protection

of the accused into balance with the rights of the victim and of

society to rapid, efficient, and reliable determination of guilt

and swift execution of punishment.

The Ninth Circuit’s attack on California’s “three strikes”

law is an unwarranted expansion of habeas review which

cripples an important recidivist sentencing scheme, and is thus

This brief was written entirely by counsel for amicus, as listed on the

cover, and not by counsel for any party. No outside contributions were

made to the preparation or submission of this brief.

Both parties have given written consent to the filing of this brief.

—_

.

contrary to the interest of victims and society that the CJLF was

formed to protect.

SUMMARY OF FACTS AND CASE

The prisoner, Leandro Andrade, was convicted in a Califor-

nia court of two counts of petty theft. Cal. Penal Code §§ 484,

487. See Andrade v. Roe, 270 F. 3d 743, 749 (CA9 2001).

Petty theft is a misdemeanor, see Cal. Penal Code § 490, but if

the defendant was previously convicted of and served a term for

burglary or other specified crimes, then the theft may be treated

as either a misdemeanor or a felony. See Cal. Penal Code

§ 666. Because the prisoner had three prior residential burglary

convictions, his thefts were enhanced to felonies. See Andrade,

supra, at 749. Andrade also had priof convictions for misde-

meanor theft in 1982, a federal felony transportation of mari-

juana offense in 1988, a state petty theft offense in 1990, and a

1991 parole violation for escape from federal prison. /d., at

748-749. California law classifies residential burglary as a

serious felony, see Cal. Penal Code § 1192.7(c)(18), making

Andrade subject to California’s “three strikes” law. See Cal.

Penal Code § 1170.12(b)(1). Under the three strikes statute.

Andrade received two consecutive 25-years-to-life sentences

for his two felony theft convictions. See Andrade, supra, at

749.

A California Court of Appeal affirmed Andrade’s convic-

tion, rejecting an Eighth Amendment attack on his sentence in

an unpublished opinion issued on May 13, 1997. App. to Pet.

for Cert. 69. Relying on the California precedent, the Court of

Appeal applied a three-part test that is almost identical to the

test utilized in Solem v. Helm, 463 U. S. 277 (1983). Compare

App. to Pet. for Cert. 77, with Solem, supra, at 292. The

_ California Supreme Court denied the prisoner’s petition for

review on July 23, 1997. App. to Pet. for Cert. 81. Andrade

filed a pro se habeas petition in the federal District Court for the

Central District of California attacking his sentence on Eighth

Amendment grounds. The District Court denied the petition

and subsequently denied Andrade a certificate of appealability.

See id., at 67.

The Ninth Circuit Court of Appeals granted Andrade a

certificate of appealability and appointed counsel. See Andra-

de, 270 F. 3d, at 750. The Court of Appeals reversed the

District Court, finding that Andrade’s sentence was dispropor-

tionate to his crime in violation of the Eighth Amendment, and

that the California Court of Appeal’s decision disregarded the

three-part test of Solem, supra, making it an unreasonable

application of Supreme Court precedent and therefore not

subject to the deferential review standards of 28 U. S. C.

§ 2254(d)(1). See id., at 766-767. This Court granted Califor-

nia’s certiorari petition on April 1, 2002.

SUMMARY OF ARGUMENT

Federal courts should address the Antiterrorism and

Effective Death Penalty Act of 1996 (“AEDPA”) issues before

assessing the correctness of the state court decision. This Court

has implicitly followed this practice in Penry v. Johnson, 532

U. S. 782 (2001) and Bell v. Cone, 535 U.S. __ (No. 01-400,

May 28, 2002). This Court should now make its practice an

express rule.

The present case differs from the Court’s prior AEDPA

cases because there are few clearly established principles in the

unsettled body of law governing Eighth Amendment attacks on

prison sentences. While the relevant decisions generally

establish a deferential standard for reviewing sentences, there

is a continuing conflict between two of the leading precedents

in this area, Rummel v. Estelle, 445 U.S. 263 (1980) and Solem

v. Helm, 463 U. S. 277 (1983). Rummel established a very

deferential standard for reviewing propcertionality claims

against prison sentences. Although it indicated that some

extraordinarily harsh sentences may not pass constitutional

muster, such cases were necessarily rare.

\

Although Solem proclaimed to be similarly deferential, the

execution of that standard differed significantly from Rummel.

The second and third parts of its three-part proportionality test

cannot be reconciled with Rummel. The Solem Court also

demonstrated a willingness to draw conclusions based upon

sentencing philosophy that was anathema to the deference to

legislative policy decisions in Rummel.

Harmelin v. Michigan, 501 U.S. 957 (1991) did not end the

confusion. As a split opinion, it is a weak candidate for

clarifying the law, although Justice Kennedy’s narrower

concurrence can be seen as the controlling precedent. That

opinion does not conclusively resolve the conflict between

Solem and Rummel, but does move the law more towards

Rummel’s broadly deferential standard. It does not attempt to

provide adequate guidance for how to determine when a

sentence is disproportionate because this area of the law is not

amenable to such standards.

The treatment of proportionality challenges in other

jurisdictions reflects the unsettled state of the law, as many state

and federal courts decry the lack of standards in this field.

Since Harmelin, some courts now claim that Solem has been

overruled, and many more claim that it has been weakened.

Under these circumstances, there is broad room for reasonable

disagreement.

The state court did not fail to apply the proper precedent.

It applied a California constitutional standard that is the

equivalent of Solem’s three-part test. Since Andrade’s sentence

is less severe and his culpability greater than the defendant in

Solem, his case is not materially indistinguishable from Solem.

Therefore, the state court satisfied the clearly established

portion of 28 U.S. C. §2254(d).

The California court’s rejection of Andrade’ s rroportional-

ity claim was a reasonable application of the little clearly

established law there is in this area. The Ninth Circuit applied

a clear error standard to assess the reasonableness of the state

court decision. A better approach is to focus on finding reasons

to support the state court decision, rather than assessing its

alleged error. The decision to uphold Andrade’s sentence 1s

reasonable under this approach.

If this Court decides that the three strikes law is unconstitu-

tional in Ewing v. California, No. 01-6978, the rejection of

Andrade’s federal habeas claim will not prevent him from

getting relief in the California courts. California allows

successive habeas petitions to raise claims made retroactive to

final judgments. Since Andrade’s claim_would be retroactive

under the first exception to Teague v. Lane, 489 U. S. 288

(1989) and under California’s more generous retroactivity rule,

he would be able to have his day in court.

ARGUMENT

I. The AEDPA issue should be resolved first,

as a “threshold matter.”

Ninth Circuit precedent takes a backwards approach to

federal habeas review under the Antiterrorism and Effective

Death Penalty Act of 1996 (“AEDPA”). Van Tran v. Lindsey,

212 F. 3d 1143 (CA9 2000), held that a federal habeas court

should “first determine whether the state court’s decision was

erroneous prior to considering whether it was contrary to or

involved an unreasonable application of the law under the

AEDPA....” /d., at 1155.

Amicus CJLF has previously argued that the AEDPA issue

should be addressed first, and adopts those arguments in this

case. See Brief for Criminal Justice Legal Foundation as

Amicus Curiae in Bell v. Cone, No. 01-400, pp. 8-14; Brief for

Criminal Justice Legal Foundation as Amicus Curiae in Penry

v. Johnson, No. 00-6677, pp. 3-9. This Court addresses the

AEDPA issues first in its habeas cases. In Penry v. Johnson,

2. These briefs are available on our Web site, www.cjlf.org/briefs/

briefmain.htm.

532 U.S. 782 (2001), this Court declined to decide “the merits

of Penry’s Fifth Amendment claim. Rather, the question is

whether the Texas court’s decision was contrary to or an

unreasonable application of our precedent.” /d., at 794-795. In

Bell v. Cone, 535 U.S. __ (No. 01-400, May 28, 2002), this

Court again determined that the state court decision was

“reasonable” without deciding whether it was correct. See id.

(slip op., at 12, 14-15). What is implicit in Penry and Bell

should be made explicit. As soon as a court determines that the

moving party is not entitled to the relief he sceks, the court can

and should stop, both for efficiency and to avoid rendering

advisory opinions. Cf. Teague v. Lane, 489 U. S. 288, 316

(1989) (plurality opinion).

When confronted with the murky area of disproportionality

challenges to prison sentences, the California Court of Appeal

recognized the few clearly established legal principles in this

field and applied them reasonably. The inquiry should begin

and end there. The Ninth Circuit’s attempt to edify the Califor-

nia courts was unnecessary and deflected it from the proper

analytical path.

II. The California court correctly identified

and applied the few clearly established principles

in an unsettled body of law.

In this Court’s previous applications of the “clearly

established” portion of 28 U. S. C. § 2254(d), identification of

the controlling precedent was straightforward. In Williams v.

Taylor, 529 U. S. 362 (2000), the landmark aecision of Strick-

land v. Washington, 466 U. S. 668 (1984) was easily identifi-

able as the relevant, clearly established Supreme Court prece-

dent. See Williams, supra, at 390-391; id., at 413-414

(O’Connor, J., concurring). In Penry v. Johnson, 332 U:S. 782

(2001), analysis of Penry’s Fifth Amendment claim centered on

whether Estelle v. Smith, 451 U. S. 454 (1981) could be

distinguished, see Penry, supra, at 793-795, while his Eighth

Amendment claim centered on Texas’ compliance with the first

Penry case, Penry v. Lynaugh, 492 U. S. 302 (1989). See

Penry v. Johnson, 532 U.S., at 786.

This case raises difficult issues regarding what is clearly

established under § 2254(d). An examination of this Court's

cases governing the proportionality of prison sentences under

the Eighth Amendment uncovers a body of law that is consi-

derably less settled than those that confronted the Williams and

Penry courts. Under these difficult circumstances, the Califor-

nia Court of Appeal cannot be said to be contrary to the clearly

established Supreme Court precedent that is in this field.

A. The Unsettled Law.

An understanding of the unsettled body of law governing

Eighth Amendment attacks on the proportionality of prison

sentences requires a brief survey of the cases in this field.

Some assert that precedent for an Eighth Amendment propor-

tionality requirement traces back to Weems v. United States,

217 U. S. 349 (1910). See Harmelin v. Michigan, 501 U. S.

957, 1012 (1991) (White, J., dissenting). Although historically

interesting, Weems’ unusual set of facts seriously limits its

relevance to modern prison sentences. The relevant precedent

is much more recent, a quartet of cases beginning with Rumme/

v. Estelle, 445 U. S. 263 (1980).

]. Rumncel.

Rummel addressed a life sentence for a comparatively minor

theft under Texas’ recidivist scheme. See id., at 264-265. In

upholding the sentence, this Court established that while prison

might be subject to Eighth Amendment proportional-

ity review, the standard of review was very deferential. It noted

that “{t}his Court has on occasion stated that the Eighth

Amendment prohibits its imposition of a sentence that is

grossly disproportionate to the severity of the crime.” /d., at

271. The largest source of such precedent came from capital

cases, which were readily distinguished “[b]ecause a sentence

of death differs in kind from any sentence of imprisonment

; Noncapital cases raising successful prop<rtionality

claims were “exceedingly rare.” /bid. Weems, the most notable

example, was limited to its “peculiar facts,” including the harsh

and alien cadena temporal. See id., at 274. The Rummel Court

concluded that “one could areve, without fear of contradiction

by any decision of this Cour that for crimes concededly

classified and classifiable as felonies . . . the length of the

sentence actually imposed is purely a matter of legislative

prerogative.” /bid.

This Court recognized a narrow exception to its deferential

rule. “This is not to say that a proportionality principle does not

come into play in the extreme example mentioned by the

dissent, post, at 288, if a legislature made overtime parking a

felony punishable by life imprisonment.” /d., at 274, n. 11. It

addressed proportionality in such narrow terms because

“a more extensive intrusion into the basic line-drawing

process that is pre-eminently the province of the legislature

when it makes an act criminal would be difficult to square

with the view expressed in Coker [v. Georgia, 433 UV. S.

584 (1977)] that the Court’s Eighth Amendment judgments

should neither be nor appear to be merely the subjective

views of individual Justices.” /d., at 275.

Rummel was convicted of obtaining $120.75 by false

pretenses, a felony. See id., at 266. He also had prior felony

convictions for $80 in credit card fraud and forging a $28.36

check. See id., at 265. Under Texas law the two prior felony

convictions elevated Rummel’s sentence to life. See id., at 264.

The Court dismissed Rummel’s attempt to use the small

amounts stolen and his lack of violence as objective examples

of disproportionality. It held that drawing such distinctions

are indeed ‘subjective’ and therefore properly within the

province of legislatures, not courts.” /d., at 275-276. Other

attempts by Rummel to distinguish his case from this Court’s

nearly total deference to sentencing statutes were similarly

unsuccessful. See id., at 280-282. Therefore, Texas’ manda-

tory life sentence for Rummei’s minor theft was upheld. See

id., at 285.

2. Hutto.

Hutto v. Davis, 454 U. S. 370 (1982) (per curiam) rein-

forced Rummel’s deference to the legislative prerogative to

define crimes and set punishments. The Court addressed a

Fourth Circuit decision holding that a 40-year sentence for

possession of less than nine ounces of marijuana was cruel and

unusual punishment. See id., at 371. The Fourth Circuit was

reversed in unusually harsh language.

“[{T]he Court of Appeals sanctioned an intrusion into the

basic linedrawing process that is ‘properly within the

province of legislatures, not courts.’ More importantly,

however, the Court of Appeals could be viewed as having

ignored, consciously or unconsciously, the hierarchy of the

federal court system created by the Constitution and

Congress. ... But unless we wish anarchy to prevail within

the federal judicial system, a precedent of this Court must

be followed by the lower federal courts no matter how

misguided the judges of those courts may think it to be.”

Id., at 374-375 (citation omitted) (quoting Rummel, 445

U. S., at 275-276).

In other words, this Court meant what it said in Rummel. A

proportionality challenge to a prison sentence was next to

impossible. Barring the extreme example of life for overtime

parking, see Rummel, 445 U. S., at 274, n. 11, prison sentences

were likely to survive an Eighth Amendment attack.

3. Solem.

Solem v. Helm, 463 U.S. 277 (1983) addressed whether life

without the possibility of parole was an unconstitutionally

disproportionate punishment for Helm’s seventh nonviolent

felony conviction. See id., at 279. Helm’s prior felony

convictions were three third-degree burglaries, theft by false

10

pretenses, grand larceny, and third-offense driving under the

influence. See id., at 279-280. His current conviction was for

“uttering a ‘no account’ check for $100.” /d., at 281. The

Solem Court began its analysis with the history of the prohibi-

tion against cruel and unusual punishment. It found that a

proportionality requirement was “deeply rooted and frequently

repeated in common-law jurisprudence.” /d., at 284. Propor-

tionality had been recognized “in this Court for almost a

century.” Id., at 286. Therefore it rejected the state’s conten-

tion that proportionality did not apply to felony prison sen-

tences. /d., at 289. While courts must be deferential to

legislative sentencing policies, “no penalty is per se constitu-

tional.” /d., at 290.

Solem took an objective approach to proportionality review.

It applied a three-part test, examining “i) the gravity of the

offense and the harshness of the penalty; ii) the sentences

imposed on other criminals in the same jurisdiction; and iii) the

sentences imposed for the commission of the same crime in

other jurisdictions.” /d., at 292. Applying this test the Solem

Court found that Helm’s sentence violated the Eighth Amend-

ment’s proportionality requirement. See id., at 303.

Its holding that this punishment violated the Eighth Amend-

ment is difficult to square with Rummel and Hutto. It is true

that the facts of Solem do differ from Rummel. Solem involved

a potentially harsher punishment, as the life sentence in Rummel

was mitigated by the possibility of parole within 12 years. See

id., at 297. Yet this would seem to have made little difference

to the Rummel Court. The Rummel decision accorded nearly

complete deference to prison sentences. See, ante, at 8. While

the Rummel Court did note the “however slim” possibility of

parole in 12 years, see 445 U. S., at 280-281, its holding did not

turn on this point. Rummel’s discussion of the possibility of

parole was a small part of a larger analysis dealing with

Rummel’s claim that no state would have punished him as

severely as Texas. See id., at 279-281. The fact that the

possibility of parole ameliorated Rummel’s sentence did not

11

support the argument’s dismissal. Instead, it showed that

Rummel’s claim involved making subtle distinctions between

gradations of crime and punishment that courts must not make.

“We offer these additional considerations not as inherent flaws

in Rummel’s suggested interjurisdictional analysis but as

illustrations of the complexities confronting any court that

would attempt such a comparison.” /d., at 281.

Solem also contradicted the spirit of Rummel. Instead of

Rummel 's declaration that prison sentences sould be attacked

only in the most extreme cases, the Solem Court warned that all

sentences were subject to scrutiny. Thus, even “a single day in

prison may be unconstitutional in some circumstances.” Solem,

463 U. S.. at 290. Solem’s three-part test is also contrary to

Rummel. Solem’s first part, comparing the gravity of the

offense to the severity of the penalty, see id., at 290-291, was

calculated differently in Rummel. The only sentence it explic-

itly recognized as disproportionate was the extreme circum-

stance of life for a parking offense. This conflict was made

more prominent by Solem’s application of this standard. The

Court minimized the gravity of Helm’s offenses because they

were not violent and involved small sums, an analysis that the

Rummel Court flatly rejected as inherently subjective. Compare

Solem, supra, at 296-297, with Rummel, 445 U. S., at 275-276.

The second Solem factor, comparison to sentences imposed

on other criminals in the jurisdiction, see 463 U. S. at 291, was

dismissed by Rummel in a footnote. See 445 U. S., at 282,

n. 27. Rummel also thoroughly dismissed Solem’s third factor,

comparing the sentences to sentences for similarly situated

criminals in other jurisdictions. Compare Solem, supra, at 291,

with Rummel. supra, at 281-282. “Absent a constitutionally

imposed uniformity inimical to traditional notions of federal-

ism, some State will always bear the distinction of treating

particular offenders more severely than any other State.”

Rummel, supra, at 282.

The two decisions even disagreed on the relevance of

sentencing philosophy to proportionality claims. While the

12

Solem Court attacked the North Dakota sentence for precluding

any efforts to rehabilitate Helm, see 463 U. S., at 297, n. 22, the

Rummel Court refused to adopt any theory of sentencing. See

445 U. S., at 283-284.

In spite of this conflict, Solem disclaimed any intent to

overrule Rummel. See 463 U. S., at 303, n. 32. The Solem

Court showed at least some affinity for Rummel’s analysis by

reiterating that successful proportionality challenges to noncapi-

tal sentences should be “ ‘exceedingly rare.” ” /d., at 289

(quoting Rummel, 445 U. S., at 272). Similarly, the Solem

Court also denied that it was creating any general authority for

appellate review of sentences. Courts should still accord

“substantial deference” to “legislatures and sentencing courts

.... Solem, 463 U. S., at 290, n. 16. It asserted that it merely

supplied standards for reviewing Eighth Amendment propor-

tionality claims that were absent from Rummel. See id., at 304,

n. 32. Since Helm’s crimes were nonviolent and minor, and his

sentence was comparatively harsh to similarly situated crimi-

nals in North Dakota and other states, his sentence was struck

down om being “significantly disproportionate to his crime

..” Id., at 303.

4. Harmelin.

The tensions underlying Eighth Amendment proportionality

review of prison sentences exploded into full view in Harmelin

v. Michigan, 501 U.S. 957 (1991). Harmelin was sentenced to

life without the possibility of parole for possessing 672 grams

of cocaine. See id., at 961 (opinion of Scalia, J.). Except for

rejecting an individualized sentencing requirement for noncapi-

tal sentences and upholding Harmelin’s sentence, see id., at 996

(majority), there was no majority opinion in this 5-4 decision.

Justice Scalia, joined by the Chief Justice, rejected Eighth

Amendment proportionality review except in capital cases. See

id., at 994. The opinion denounced Solem as simply wrong and

“scarcely the expression of clear and well accepted constitu-

tional law.” See id., at 965. Rummel and Hutto were cited

13

approvingly in Justice Scalia’s opinion although both cases

accepted the possibility of a noncapital proportionality guaran-

tee. See id., at 962-964. Harmelin’s sentence was necessarily

lawful under this approach. See id., at 996.

Justice Kennedy, joined by Justice O’Connor and Justice

Souter, agreed that Harmelin’s sentence should be upheld, but

that stare decisis counseled against rejecting “the narrow

proportionality principle that has existed in our Eighth Amend-

ment jurisprudence for 80 years.” /d., at 996 (opinion of

Kennedy, J.). However, this opinion recognized that it was

upholding a fractured line of authority. “Although our propor-

tionality decisions have not been clear or consistent in all

respects, they can be reconciled, and they require us to uphold

petitioner's sentence.” /d., at 996-997. After briefly surveying

the Eighth Amendment proportionality cases since Weems, see

id., at 997-998, Justice Kennedy found fault with the current

state of proportionality analysis in noncapital cases. “Though

our decisions recognize a proportionality principle, its precise

contours are unclear. This is so in part because we have applied

the rule in few cases and even then to sentences of different

types.” Jd., at 998. Thus Solem “appeared to apply a different

analysis than in Rummel and Davis” even though it purported

to uphold those decisions. /bid. What could be harvested from

these conflicting authorities was “some common principles that

give content to the uses and limits of proportionality review.”

Ibid.

These principles are very general, and lean more towards

Rummel than Solem. First, “the fixing of prison terms for

specific crimes involves a substantive peneological judgment

that as a general matter is ‘properly within the province of

legislatures, not courts.” /bid. (quoting Rummel, 445 U. S., at

275-276). Determining the proper punishment involves

difficult policy questions containing strong political and moral

elements. See ibid. Appellate courts should not make such

distinctions. This was the central theme of Rummel, that the

14

appellate courts should not second-guess these policy decisions.

See Rummel, 445 U. S., at 279.

“The second principle is that the Eighth Amendment does

not mandate adoption of any peneological theory.” Harmelin,

501 U. S., at 999 (opinion of Kennedy, J.). Once again, this is

closest to Rummel’s reasoning. As noted earlier, while the

Rummel decision deferred to the peneological theories support-

ing Texas’ recidivism statute, the So/em Court criticized the life

without possibility of parole sentence as foreclosing any

attempts at Helm’s rehabilitation. See ante, at 10. For the third

principle, that “marked divergences both in underlying theories

of sentencing and in the length of prescribed prison terms are

the inevitable, often beneficial, result of the federal structure,”

Justice Kennedy’s opinion cited Solem. See ibid. This princi-

ple is also closer to Rummel than Solem. The Rummel Court

strongly criticized interjurisdiction comparison of sentences,

see Rummel, 445 U. S., at 281; Harmelin, supra, at 999-1000

(opinion of Kennedy, J.), and was willing to uphold a penalty

even if it was the most severe in the nation for that crime. See

Rummel, at 209-282. While the Solem decision acknowledged

difficulties in making sch comparisons, see 463 U. S., at 294-

295; Harmelin, supra, at 1000 (opinion of Kennedy, J.). it still

made interjurisdictional comparison a mandatory part of its

three-part test. See Solem, supra, at 291-292.

The fourth principle, that proportionality review should be

driven by objective factors, is derived from both Solem and

Rummel. See Harmelin, 501 U. S., at 1000 (opinion of Ken-

nedy, J.). The deference in Rummel was driven by the need to

avoid substituting subjective judicial sentiments for legislative

policy choices. See Rummel, 445 U. S., at 275. In Solem, the

quest for objectivity led to its three-part test. See 463 U. S.., at

290-291. Justice Kennedy’s opinion also recognized that “we

lack clear objective standards to distinguish between sentences

for different terms of years” making successful proportionality

challenges to prison sentences “ ‘ “exceedingly rare.” ° ”

15

Harmelin, supra, at 1001 (quoting Solem, supra, at 290

(quoting Rummel, supra, at 272)).

The remaining important feature of the concurrence was the

decision to forego the last two parts of the Solem test. Solem

was now “understood as holding that comparative analysis

within and between jurisdictions is not always relevant to

proportionality review.” /d., at 1004-1005. Thus, the last two

Solem factors would only be referred to when “a threshold

comparison of the crime committed and the sentence imposed

leads to an inference of gross disproportionality.” /d., at 1005.

Harmelin is a clear step back from Solem. Two justices

voted to overrule it, while the remaining members of the

majority would all but abandon two-thirds of its test. Justice

Kennedy's opinion asserted it “neither ‘eviscerate|s]’ Solem,

nor ‘abandon{s]’ its second and third factors ....” /d., at 1005

(quoting id., at 1018, 1020 (White, J., dissenting)). It did,

however, represent a revival of Rummel and Hutto. See id., at

1005 (opinion of Kennedy, J.) (“full account”). Given the

conflict between Solem and Rummel, any return to Rummel is

necessarily a step back from Solem.

This is the body of precedent that confronted the California

appellate court when it addressed Andrade’s Eighth Amend-

ment claim: conflicting decisions providing comparatively

little concrete guidance outside of a strong deference to

noncapital sentences, and this Court stepping back from the

only explicit test it had promulgated. The question is not

simply whether the California court correctly identified and

applied clearly established law, but rather whether there was

any clearly established law for the court to apply?

B. Searching for Clarity.

Any attempt to find a clearly established precedent for the

California court to follow begins in Harmelin v. Michigan, 501

U. S. 957 (1991). In addition to being the last word on the

subject, Harmelin reflects the fissures and uncertainty running

16

through the Eighth Amendment’s proportionality requirement

for noncapital sentences. As a split opinion in an already

conflicted body of precedent, Harmelin is not a strong candi-

date for clearly establishing any precedent. However, a handful

of principles can be refined from the Harmelin opinion. In a

fragmented opinion, the holding is taken from the opinion

concurring on the narrowest grounds. See Marks v. United

States, 430 U. S. 188, 193 (1977). Since Justice Kennedy’s

concurrence did not call for overruling any of this Court’s

precedents, it is the narrower of the two Harmelin concurrences,

and therefore is F' .nelin’s holding.

Harmelin does stand for more than deference to noncapital

sentences. The reiteration of the principle that sentences must

be analyzed by objective factors, see Harmelin, 501 U. S.., at

1000 (opinion of Kennedy, J.), is not expressly deferential. But

see Rummel, 445 U. S., at 275(extensive judicial intrusion into

legislative punishment decisions threatens objectivity princi-

ple). However, the objectivity principle provides limited

guidance. Justice Kennedy’s concurrence reiterates the

difficulty of finding any objective factors for analyzing the

constitutionality of prison sentences. See Harmelin, supra, at

1001 (opinion of Kennedy. J.). Once again. this supports

deferring to the constitutionality of prison sentences. The

“relative lack of objective standards” makes successful Eighth

Amendment attacks on prison sentences “ * “exceedingly

rare.” ’” Ibid. (quoting Solem, 463 U. S., at 290 (quoting

Rummel, supra, at 272)).

The final principle, that the Eighth Amendment only forb:us

grossly disproportionate’ ” sentences, see ibid. (quoting

Solem, 463 U. S., at 288), is the culmination of the deference

that is the central theme of Harmelin. This drove the final

principle of Justice Kennedy’s Harmelin concurrence, dispens-

ing with Solem’s comparative analysis absent an inference of

gross disproportionality. See id., at 1005. Like the rest of

Justice Kennedy’s opinion, this principle is most likely to be

violated by a court improperly striking down a sentence.

“ 6

17

While Justice Kennedy’s opinion instructed reviewing

courts to defer to all but the grossly disproportionate sentences,

it understandably provides little real guidance on how to

determine when a prison sentence raises such an inference. The

law is unclear in this area because there are comparatively few

proportionality cases, and they address “sentences of different

types.” See id., at 998. The analysis of Harmelin’s sentence of

life without possibility of parole a first-time drug offense has

minimal relevance for Andrade ’s two consecutive 25-years-to-

life sentences under a recidivist sentencing statute. This lack of

ready standards means that the scope of reasonable disagree-

ment is necessarily very broad. Therefore very few applications

of these precedents can be classified as “unreasonable.” _

Although Solem and Rummel present closer facts, their

guidance is limited. A final principle of Justice Kennedy’s

Harmelin concurrence is that the law governing Eighth Amend-

ment attacks on prison sentences is unclear and conflicted. See

ibid. This opinion did not resolve the conflict between Solem

and Rummel, see part | A 3, ante, that afflicts this part of the

law. The common principles it drew from these cases boil

down to instruction to the appellate courts to defer to most

prison sentences presented to them. While stare decisis kept it

from accepting Justice Scalia’s invitation to overrule Solem, see

id., at 996, Justice Kennedy’s concurrence does represent a

significant move away from Solem towards Rummel. See part

I A 4, ante. Contrary to the Ninth Circuit’s assertion, see

Andrade v. Roe, 270 F. 3d 743, 766-767 (CA9 2001), Solem

does not compel the result in this case because its exact strength

as precedent is uncertain. While So/em may have survived

Harmelin, it has been diminished. Precisely how diminished is

not yet known.

The confused state of the law after Harmelin is reflected by

the reaction of the state and federal courts. Many decisions in

the state and lower federal courts comment on the uncertain

state of the law since Harmelin. See, e.g., United States v.

Sarbello, 985 F. 2d 716, 723 (CA3 1994) (“lack of clear

18

directive from the Supreme Court”); Neal v. Grammer, 975

F. 2d 463, 465 (CA8 1992) (“the future of the proportionality

test is uncertain”); Hawkins v. Hargett, 200 F. 3d 1279, 1281

(CA10 1999) (Harmelin “fractured” leaving the meaning of

Solem “less than clear”); McCollough v. Singletary, 967 F. 2d

$30, 535 (CA11 1992) (viability of Solem called into doubt by

Harmelin), State v. Brown, 825 P. 2d 482, 490-491 (Idaho

1992) (Harmelin “fractured”); People v. Gibson, 90

Cal. App. 4th 371, 387, 108 Cal. Rptr. 2d 809, 820 (2001)

(continued existence of proportionality review questionable

after Harmelin). State v. Oliver, 745 A. 2d 1165, 1169 (NJ

2000) (“we have generally avoided entering the debate among

the several members of the Supreme Court concerning the

Eighth Amendment’s proscription against cruel and unusual

punishment”); State v. Thorp, 2 P. 3d 903, 906 (Or. App. 2000)

(Harmelin was “severely fractured”); State v. Bonner, 577

N. W. 2d 575, 579 (SD 1998) (regretting lack of majority

opinion in Harmelin); State v. Jones, 543 S. E. 2d 541, 545,

n. 11 (SC 2001) (“questionable” whether Solem test still

mandatory after Harmelin); State v. Harris, 844 S. W. 2d 601,

602 (Tenn. 1992) (precise contours of the federal proportional-

ity guarantee unclear); State v. Bacon, 702 A. 2d 116, 122, n. 7

(Vt. 1997) (Harmelin is a “fractured opinion” that casts doubt

on Solem).

The Ninth Circuit’s opinion in this case reflects the con-

fused state of the law. It casually dismissed the California

Court of Appeal’s decision for failing to follow clearly estab-

lished law because it questioned So/em in light of Harmelin.

See Andrade, 270 F. 3d, at 766-767. This ignores both the

conflicts within this Court’s precedents, see ante, at 10-12, and

the many state and federal opinions questioning Solem after

Harmelin.

Some courts have gone so far as to claim that Solem was

effectively overruled by Harmelin. Not long after Harmelin

was decided, the Fifth Circuit stated that “disproportionality

survives; Solem does not.” McGruder v. Puckett, 954 F. 2d

Oe ae re Se

19

313, 316 (CAS 1992); see also Smallwood v. Scott, 73 F. 3d

1343, 1346, n. 4 (CAS 1996) (“Solem was overruled to the

extent that it found in the Eighth Amendment a guarantee of

proportionality); Edwards v. State, 800 So. 2d 454, 469 (Miss.

2001) (Solem overruled to the extent that it guaranteed

proportionality”). Others have been more circumspect,

concluding. as the California Court of Appeal did. that Solem

was called into question by Harmelin. See, e.g., Sarbello, 985

F. 2d, at 723; United States v. Kratsas, 45 F. 3d 63, 66 (CA4

1995) (“cast some doubt”); McCollough, 967 F. 2d, at 535

(viability of Solem called into doubt); Brown, 825 P. 2d, at 491-

492 (“seriously erodes”); State v. Lara, 580 N. W. 2d 783, 785

(Iowa 1998) (“called into question”); State v. Scott, 961 P. 2d

667, 672 (Kan. 1998) (“discredited”); State v. Lindsey, 770

So. 2d 339, 344, n. 2 (La. 2000) (“called into question”); State

v. Riley, 497 N. W. 2d 23, 26 (Neb. 1993) (“extremely

doubtful” precedent). Finally, another group of courts found

that Solem has been limited by Harmelin. See, e.g., United

States v. Bucuvalas, 970 F. 2d 937, 946 n. 15 (CA1 1992)

(“significant curtailment”), overruled on other grounds in

Cleveland v. United States, 531 U.S. 1218 (2000); Hawkins,

200 F. 3d, at 1282 (“narrows Solem”); People v. Cooper, 43

Cal. App. 4th 815, 820, 51 Cal. Rptr. 2d 106, 108 (1996)

{weakened substantially”); People v. Hindson, 703 N. E. 2d

956, 965 (Ill. App. 1998) (“narrowed”); State v. Lee, 841

S. W. 2d 648, 654 (Mo. 1992) (“altered”). When the California

appellate court questioned So/em, it had plenty of company.

This is relevant to the “clearly established federal law”

inquiry under §2254(d). Although the source of clearly

established law is restricted to this Court’s decisions, Williams

v. Taylor, 529 U. S. 362, 412 (2000), the fact that the lower

courts have difficulty in deriving a clear rule from this Court’s \

precedents is relevant to determining whether the purported rule

is clearly established. See Caspari v. Bohlen, 510 U. S. 383,

395 (1994) (conflicting, reasonable views of lower courts

precluded “old-rule” status under Teague); Williams, supra, at

- 412 (equivalence of Teague “old rule” with AEDPA “clearly

20

established”). If questioning Solem in light of Harmelin

constitutes a failure to apply established Supreme Court

precedent, then why have so many other courts followed the

same path?

A state court decision is:

“ ‘contrary to’ our clearly established precedent if the state

court either ‘applies a rule that contradicts the governing

law set forth in our cases,’ or ‘confronts a set of facts that

are materially indistinguishable from a decision of this

court and nevertheless arrives at a result different from our

precedent.” ~ Penry v. Johnson, 532 U. S. 782, 792 (2001)

(quoting Williams, 529 U. S., at 405-406).

This is a matter of rule selection. Because this Court’s prece-

dents are not always clear or consistent, see, e.g., Harmelin v.

Michigan, 501 U.S. 957, 996-997 (1991) (opinion of Kennedy,

J.); Thompson v. Keohane, 516 U.S. 99, 111 (1995), it may not

be easy to discern the governing law from this Court’s cases.

While confusion in the lower federal and state courts is not

dispositive, it is a symptom of an unclear or conflicted body of

law, i.e., not “clearly established.”

If this Court could not agree on the proper standard for

assessing proportionality challenges to prison sentences in

Harmelin, then there can be very little “clearly established

federal law” in this field. At most, appellate courts presented

with proportionality attacks on prison sentences should com-

pare the severity of the sentence to the magnitude of the crime

and the defendant’s criminal history to determine whether there

is an inference of gross disproportionality. This determination

is made with “utmost deference to the Legislature and the

sentencing court.” See Bonner, 577 N. W. 2d, at 580. If an

inference of gross disproportionality is found, only then may

the reviewing court turn to other objective factors. These may

or may not include the Solem factors, or other factors not yet

identified by this Court.

a ee

21

What is not clearly established is any particular method for

comparing punishment and culpability. Because the conflict

between Rummel and Solem is unresolved, neither of these

different approaches can be said to clearly govern proportional-

ity claims. While Justice Kennedy’s Harmelin opinion pro-

vides some guidance, the life without possibility of parole

sentence for a drug conviction gives Harmelin little relevance

to Andrade’s lesser sentence under a recidivism statute. Since

each defendant is unique, and it is difficult to draw a line

separating discrete prison sentences, see Rummel, 445 U. S., at

275, it may be impossible to give more concrete guidance. So

long as a state court compares the culpability of the defendant's

crime and criminal history to the severity of the sentence, then

its decision is not contrary to any principle of federal law which

has been clearly established by this Court up to this time.

The California Court of Appeal applied an independent state

constitutional standard to the defendant’s disproportionality

claim that was almost identical to Solem’s three-part test.

“A tripartite test has been established to determine whether

a penalty offends the prohibition against cruel and unusual

punishment. First, courts examine the nature of the offense

and the offender, with particular regard to the degree of

danger both present to society. Second a comparison is

made of the challenged penalty with those imposed in the

same jurisdiction for more serious crimes. Third, the

challenged penalty is compared with those imposed for the

same offense in other jurisdictions.” App. to Pet. for Cert.

77 (internal quotations omitted); accord People v. King, 16

Cal. App. 4th 567, 571, 20 Cal. Rptr. 2d 220, 222 (1993);

In re Lynch, 8 Cal. 3d 410, 425-427, 503 P. 2d 921, 930-

932 (1972).

Since this standard is as favorable to the defendant as any

articulated by this Court, including So/em, the California court

did not choose the incorrect rule. Andrade’s case is not

“materially indistinguishable” from Solem. His sentence is less

severe than Helm’s and his culpability is greater. Therefore, the

22

-

California court at least identified and applied the proper rule.

This was not a rogue state court ignoring clearly established -

precedent. Instead, when confronted with a confusing body of

law, it applied a state constitutional standard more generous

than the defendant was entitled to under this Court’s cases. The

state court’s decision cannot be said to be “contrary to clearly

established Federal law” as that phrase is interpreted in Wil-

liams.

I!l. The California court’s application of

what little clearly established law there is in this area

was reasonable.

The Ninth Circuit’s analysis of the reasonable application

prong under 28 U. S. C. § 2254(d) suffers from several flaws.

Its misidentification of Solem v. Helm, 463 U.S. 277 (1983) as

the controlling precedent to the near exclusion of the other

precedents, see ante, at 17, distorts its analysis of whether the

California court’s decision was reasonable. Also, the federal

court did not properly defer to California’s determination that

this was the appropriate sentence for Andrade. Most impor-

tantly, the Ninth Circuit applied the wrong standard for

analyzing whether a state court decision reasonably applies this

Court’s precedents.

The problem with the Ninth Circuit’s approach to the

reasonable application prong begins with the standard it applied

to this test. There is no detailed test to determine when a state

court’s application of Supreme Court precedent is objectively

unreasonable. In the cases addressing this issue, this Court

simply analyzed the state court opinion and the relevant

Supreme Court precedents and concluded that the state decision

was either unreasonable, see Williams v. Taylor, 529 U.S. 362,

397 (2000); Penry v. Johnson, 532 U.S. 782, 803-804 (2001)

(Fifth Amendment claim), or reasonable. See id., at 795

(Eighth Amendment claim). Bell v. Cone, 535 U.S. __ (No.

01-400, May 28, 2002) (slip op., at 15-16). Other than inform-

23

ing the courts that unreasonableness must be determined

objectively. and is more than mere error, see Williams, supra,

at 410, this Court has given the lower federal courts little

guidance in this admittedly difficult area. See ibid. As a

consequence, the circuits have adopted their own tests for

determining the reasonableness of a state court’s decision. The

Ninth Circuit’s is far afield.

The Ninth Circuit standard for reasonableness under

§ 2254(d) is “clear error.” Under this standard, a court first

determines whether the state court decision was correct under

its own interpretation of Supreme Court precedent. If it

disagrees with the state court decision, it next determines

whether this “error” is “clear error.” “Clear error” is defined as

“where the Court of Appeals is left with a ‘definite and firm

conviction’ that an error has been committed.” Van Tran v.

Lindsey, 212 F. 3d 1143, 1153 (CA9 2000). If the error is

labeled “clear,” then the state court decision is deemed unrea-

sonable under § 2254(d). See id., at 1153-1154.

Focusing on the purported state court error starts the

analysis on the wrong foot. Congress did not equate the

unreasonable application standard with state court “error.” See

Williams, 529 U.S., at 410. Focusing on state court error backs

away from the deference accorded to state courts under

§ 2254(d). It does not consider that in cases of disagreement

with the lower federal courts, the state courts are very often

correct and the federal courts are the ones in error. See

Scheidegger, Habeas Corpus, Relitigation, and the Legislative

Power, 98 Colum. L. Rev. 888, 942 (1998). The Ninth

Circuit’s “approach . . . focuses on how sure the habeas court is

that the state court has committed error, not whether the state

court decision reveals an increment of wrongness beyond

error.” Francis S. v. Stone, 221 F. 3d 100, 110 (CA2 2000).

The clear error standard also deflects the habeas court from

addressing the AEDPA issue first. See Bell v. Jarvis, 236 F. 3d

149, 162, n. 9 (CA4 2000). This Court should provide further

24

guidance on how to determine when a state court decision is

unreasonable.

Courts use tests in other contexts to determine whether an

act is reasonable. The Fourth Amendment proscribes “unrea-

sonable searches and seizures... .” U.S. Const., Amdt. 4.

This Court guides efforts to assess whether a search or seizure

is reasonable through a host of tests and standards. The

reasonable suspicion standard helps courts and officers deter-

mine when a stop and frisk is reasonable. See Terry v. Ohio,

392 U.S. 1 (1968). Reasonableness can be detached from the

Fourth Amendment’s warrant requirement when the search is

deemed reasonable after balancing society’s interest in the

search against the individuals’ privacy interests. See Vernonia

School Dist. No. 47J v. Acton, 515 U. S. 646, 664-665 (1995).

Giving the federal courts more tools to determine when a state

court decision is reasonable is consistent with the letter and the

spirit of § 2254(d).

A proper standard for reasonableness must reflect the fact

that genuinely unreasonable state court decisions are rare. State

courts are quite willing and capable of enforcing the criminal

defendant's constitutional protections, see Sione v. Powell, 428

U. S. 465, 493, n. 35 (1976); the AEDPA is a recognition of

this fact. If federal habeas courts commonly find “unreason-

able” state decisions, then AEDPA’s deference to state convic-

tions evaporates.

As in Williams, this Court can look to the circuit courts for

a standard to apply to § 2254(d). See Williams, 529 U. S.., at

405 (partially adopting Green v. French, 143 F. 3d 865 (CA4

1998)). In O'Brien v. Dubois, 145 F. 3d 16 (CA1 1998), the

First Circuit devised a useful standard for resolving reasonable

application claims. “[F]or the writ to issue, the state court

decision must be so offensive to existing precedent, so devoid

of record support, or so arbitrary, as to indicate that it is outside

the universe of plausible, credible outcomes.” /d., at 25.

Another formulation of this standard comes from the Seventh

Circuit. “The statutory ‘unreasonableness’ standard allows the

25

state court’s conclusion to stand if it is one of several equally

plausible outcomes . . .. Some decisions will be at such tension

with governing U. S. Supreme Court precedents, or so inade-

quately supported by the record or so arbitrary, that a writ must

issue.” Hall v. Washington, 106 F. 3d 742, 748-749 (CA7

1997). This standard is not foreign to this Court’s analysis. It

has relied on the term “plausible” in a similar context. See

Christianson v. Colt Industries Operating Corp.,486U .800,

819 (1988) (decision of coordinate court to transfer jurisdiction

upheld under law-of-the case if “plausible”). Scheidegger,

supra, 98 Colum. L. Rev., at 914-915 (discussing Christianson,

supra).

Instead of examining the state court decision for error,

habeas courts should see whether the decision is plausible. This

is more than exchanging adjectives. “Plausible” is defined as

“seemingly or apparently valid, likely or acceptable, credible

....” American Heritage Dictionary 1388 (3d ed. 1992). This

shifts the inquiry away from whether the state decision was

wrong to whether there are reasons that it could be right. Ifa

reason for the decision can be found, it is not unreasonable.

Federal courts applying §2254(d) should first look for the

positive in state court opinions before examining them for error.

Kibbe v. Dubois, 269 F. 3d 26 (CA1 2001) illustrates these

principles. Kibbe was convicted in a Massachusetts court for

arson of a building. /d., at 28. The defendant ran from the

police as they approached the crime scene. Once he was caught

and read his Miranda rights, the defendant did talk to the

police. See id., at 28-29. At trial he testified that he ran

because he was on parole, but he had not given this explanation

to the police. See id., at 29. Defense counsel commented in the

opening and closing arguments that Kibbe voluntarily told the

police what he was doing. See ibid. The prosecution argued at

closing that Kibbe’s failure to mention his reason for running

from the police damaged his credibility, and that guilt should be

inferred from flight. See ibid. The District Court found that

this violated Doyle v. Ohio, 426 U. S. 610, 617-619 (1976) and

the state court opinion was contrary to Doyle. See Kibbe,

supra, at 31. It also found that this was an unreasonable

application of Doyle, but declined to develop that holding. See

id., at 32, n. 4. As an alternative holding, the District Court

held that the state court had unreasonably applied the rule of

Anderson v. Charles, 447 U.S. 404 (1980) to conclude that the

defendant had waived his right to silence in its entirety, thus

distinguishing Doyle. See Kibbe, supra, at 31. The District

Court surveyed the treatment of similar cases in the federal

circuits to come to the conclusion that “I am firmly convinced

that error occurred and that the Kibbe decision is an unreason-

able outcome.” /d., at 33.

After finding that the state court decision was not contrary

to clearly established law, see id., at 35-36, the Circuit Court

turned to the unreasonable application standard. Applying the

First Circuit’s “plausible, credible outcomes” standard, the

Circuit Court noted that the case fell outside the Doyle and

Charles decisions. See id., at 37. The fact that the issue was

unresolved increases the chance “that there are other reason-

able, yet contradictory, interpretations of Supreme Court

precedent.” /bid. The state court's interpretation of Doyle and

Charles was one alternative to the District Court’s. See ibid.

The First Circuit found that the state court’s interpretation of

Charles was credible, that “a court could plausibly argue that

this holding should be extended to cover the facts of Kibbe’s

case.” /d., at 38. This conclusion was supported by similar

results in federal appellate decisions. See id., at 38-39.

However, it was not necessary for these precedents to point out

the correct path of the law. “Furthermore, given its limited

scope of review under the AEDPA, a federal habeas court ought

not to provide the definitive answer to this open question.” /d.,

at 39. The state court decision did not have to be right, it only

had to “rest[{ |] upon a plausible interpretation of Supreme Court

precedent... .” /bid.

This approach addresses Andrade’s claim much more

accurately and economically than the Ninth Circuit’s. Because

27

the state of the relevant Eighth Amendment precedents is so

confused, contradictory but still reasonable alternative interpre-

tations of this Court’s cases are likely. The California court’s

opinion is a plausible interpretation of what authority can be

drawn from Harmelin, Solem v. Helm, 463 U.S. 277 (1983).

Rummel v. Estelle, 445 U.S. 263 (1980), and Hutto v. Davis,

454 U. S. 370 (1982).

This case is outside the facts of all four decisions. Harme-

lin and Hutto involved drug crimes, not recidivist schemes.

Solem, in addition to being diminished by Harmelin, involved

a harsher sentence. While Andrade’s sentence is long, it is still

less than the sentence of life without the possibility of parole in

Solem. More importantly, Andrade’s 40-year minimum is the

result of two convictions with consecutive sentences, not the

single conviction at issue in Harmelin, Solem, Hutto, and

Rummel. “\t would scarcely be competent for a person to assail

the constitutionality of the statute prescribing a punishment for

burglary on the ground that he had committed so many burglar-

ies that, if the punishment for each were inflicted on him, he

might be kept in prison for life.” State v. Four Jugs of Intoxi-

cating Liquors, 2 Atl. 586, 593 (Vt. 1886). Finally, Andrade’s

prior residential burglaries were more serious than Helm’s

third-degree burglaries. Residential burglary is more than a

simple property crime. The federal sentencing guidelines

require markedly enhanced sentences for offenders who have

been convicted in federal courts for “a crime of violence” or

“controlled substance offense” and have two or more prior

convictions of such offenses. See U. S. Sentencing Guidelines

§4B1.1(1). Among the listed violent crimes is residential

burglary. See id., at §4B1.2. As one federal judge has noted,

“these subsections establish that an offense is a ‘crime of

violence’ if injury occurs, is threatened, is attempted, or is

likely.” United States v. Rutherford, 54 F. 3d 370, 378 (CA7

1995) (Easterbrook, J., concurring). Since Solem has been

diminished, and Andrade’s criminal history is more severe than

Helm’s, it was plausible for the state court to distinguish Solem.

28

The California Court of Appeal’s duty was to apply in a

reasonable manner the principles of deference and objectivity

from Justice Kennedy’s opinion in Harmelin. The decision to

not impose its own sentencing philosophy in place of that of the

electorate and the sentencing judge’ is consistent with the

deference principle. Given the unsettled state of the law in this

area, Andrade’s more serious criminal history compared to any

of the four defendants in the relevant Supreme Court authori-

ties, and the fact that he was convicted of two separate felonies,

the California appellate court’s weighing of culpability and

severity of punishment is plausible.

IV. If this Court subsequently invalidates part of

California’s three strikes scheme, the defendant

can obtain relief in the California courts

even if his claim is rejected in the present case.

The AEDPA should foreclose any holding on the merits of

Andrade’s Eighth Amendment claim. The Eighth Amendment

status of California’s three strikes law will be addressed on the

merits in the separate case of Ewing v. California, No. 01-6978.

For the reasons CJLF intends to brief in that case, the state

should prevail on the straight merits. However, we will also

address here the contingency of a change in the law wrought by

Ewing.

If this Court decides that the statute is unconstitutional as

applied in Ewing, the rejection of Andrade’s federal habeas

claim will not prevent him from obtaining reiief in the Califor-

nia courts. California allows successive habeas petitions for

claims “based on a change in the law which is retroactively

3. Three strikes sentences are not truly mandatory. The sentencing judge

has discretion to “strike” prior convictions in the interests of justice.

See Cal. Penal Code § 1385; People v. Superior Court (Romero), \3

Cal. 4th 497, 529-530, 917 P. 2d 628, 647 (1996); People v. Williams,

17 Cal. 4th 148, 161, 948 P. 2d 429, 437 (1998) (describing discretion).

——————— “-

29

applicable to final judgments.” People v. Clark, 5 Cal. 4th 750,

775, 855 P. 2d 729, 745 (1993). Given the unsettled state of the

Eighth Amendment in this area, any decision favoring Ewing

would be a new rule. A decision in Ewing’s favor would be

retroactive on federal habeas under the first prong of Teague v.

Lane, 489 U. S. 288 (1989) as it would prohibit a type of

punishment for a class of defendant because of their status or

offense. See Penry v. Lynaugh, 492 U.S. 302, 329-330 (1989).

California’s rule on retroactivity is even more generous. A

habeas petition may raise an issue previously rejected on appeal

“when there has been a change in law affecting the petitioner.”

In re Harris, 5 Cal. 4th 813, 841, 855 P. 2d 391, 407 (1993). If

this Court decides in Ewing’s favor, proper use of the AEDPA

in the present case will not keep Andrade from getting any

deserved relief in the California courts.

Federal habeas relief may not be granted if state remedies

are not exhausted. See 28 U. S. C. §2254(b). Exhaustion is

determined in the present tense. The fact that an unexhausted

remedy existed in the past does not preclude exhaustion if that

remedy is no longer available. Coleman v. Thompson, 501

U.S. 722, 732 (1991). Conversely, the fact that the state courts

did not offer a remedy at one point does not satisfy exhaustion

if they offer one now.

In Gusik v. Schilder, 340 U. S. 128 (1950), Gusik was

convicted of murder in a court-martial, and attacked the

conviction through a habeas petition in federal District Court.

See id., at 129. The Court of Appeals reversed the District

Court’s grant of relief, holding that Gusik first had to exhaust

a newly created administrative remedy. See id., at 129-130.

Therefore it dismissed the petition without prejudice to file a

new petition after exhausting the administrative remedy. See

id., at 130. This Court held that instead of dismissing the

petition, the Court of Appeals should have “held the case

pending resort to the new remedy... .” /d., at 133. If Gusik

obtains administrative relief then the petition is dismissed, if hc

is unsuccessful it is reinstated, avoiding the cost of relitigation

30

See id., at 133-134. While it disposed of the case differently,

this Court still agreed with the main point of the Court of

Appeal that the new remedy must be exhausted before being

addressed in a federal habeas petition.

If this Court rules in favor of Ewing, Andrade’s case should

be treated similarly. The federal courts should either dismiss or

withhold judgment on his claim until it is exhausted in the

California courts. The dismissal rule is more strict now than it

was at the time of Gusik. See Rose v. Lundy, 455 U.S. 509,

522 (1982). A remand petition following exhaustion is not

“successive.” Slack v. McDaniel, 529 U. S. 473, 478 (2000).

The statute of limitations “clock” resets upon announcement of

the retroactive new rule. 28 U.S. C. §2244(d)(1\(c). Califor-

nia’s liberal approach to retroactivity on habeas corpus guaran-

tees that Andrade would have his day in state court, and federal

review would remain available in the highly unlikely event that

California's courts fail to reasonably apply whatever new rule

may be created in Ewing.

CONCLUSION

The decision of the Ninth Circuit Court of Appeals should

be reversed.

June, 2002

Respectfully submitted,

CHARLES L. HOBSON

Attorney for Amicus Curiae

Criminal Justice Legal Foundation

7 iy