Amicus Curiae Brief — Selvage v. Collins

— ee ee) a ee —- Ee ———— ———— Oe a — Ve

Question Presented

At the time petitioner was tried, was there “cause” for not

raising a claim based upon arguments later accepted in Penry v.

Lynaugh, No. 87-6177 (June 26, 1989); and if not, would the

application of a procedural bar to the claim result in a “fun-

damental miscarriage of justice,” Smith v. Murray, 477 U. S.

478 (1986)?

TABLE OF CONTENTS

Table of Authorities Cited 2.0... ccc cece rene iv

DMG oni geccccetoesesceceenes 1

Summary of Factsand Case ... 0. ccc cee eee 2

PEE wccccc cee cee eee eee seees 4

I

Stability and clarity are desperately needed in the procedural

DT Suceeecetcesevecacesoveeses 5

oh ae

Only genuinely new rules are sufficiently novel to constitute

“couse” wader Reed v. Ross ........002ccceceee 8

Ill

New procedural rules ate almost never retroactive on habeas

DED oe eeceecscascecssees eesees 13

IV

Any claim which is “novel” under Reed is necessarily “new”

SEE Ke eee ees ceccereneapeeeres 17

A. The basic rule of preclusion ............... 17

B. The timing exception... . 2... 6.6 ee ee eee 20

C. Fundamental v. Nonfundamental Rules ....... 21

D. Procedural Default v. “Merits” ............. 23

PTE Bdeecudeetececcee-soewen 24

V

“Fundamental miscarriage of justice” must be based on

SPEED Cece ccc ccc esccceseces 25

Table of Authorities

Amadeo v. Zant, 108 S. Ct. 1771, 100 L. Ed. 2d 249 (1988) .6,8

Batson v. Kentucky, 476 U. S. 79, 90 L. Ed. 2d 69,

eS san 6 eked oe ews 6 aos 18

Bowen v. Kemp, 832 F.2d 546 (11th Cir. 1987) ........ .30

Brown v. Allen, 344 U. S. 443, 97 L. Ed. 2d 469,

I ce ek eae wk 6 O40 08 60 0 6° 5

California v. Brown, 479 U. S. 538, 93 L. Ed. 2d 934,

2. ban « Ske Ow ocd oie dis 6.66 « 16

Castille v. Peoples, 103 L. Ed. 2d 380, 109 S. Ct. 1056 (4989) . 7

Chambers v. Mississippi, 410 U. S. 284, 35 L. Ed. 2d 297,

in. 2a ea Leu ooh Gah 0610's 6.6 15

Coker v. Georgia, 433 U. S. 584, 53 L. Ed. 2d 982,

EE fae a ee Wie oa Wiele wesc oe ae

Desist v. United States, 394 U.S. 244, 22 L. Ed. 2d 248,

PURGE MEEEEEED 6 6 cone cew csc ccicsewces 15

Duckworth v. Eagan, 109 S.Ct. 2875,

ls TET ee ee 22

Dugger v. Adams, 109 S. Ct. 1211, 103 L. Ed. 2d 435 (1989) .26

Eddings v. Oklahoma, 455 U. S. 104, 71 L. Ed. 2d 1,

Te eee 3, 16, 21, 25

Engle v. Isaac, 456 U. S. 107, 71 L. Ed. 2d 783,

102 S. Ct. 1558 (1982)... ......20.. 8, 9, 10, 18, 19, 25

Enmund v. Florida, 458 U. S. 782, 73 L. Ed. 2d 1140,

102 S. Ct. 3368 (1982)... 2... eee 27, 29

Fay v. Noia, 372 U. S. 391, 9 L. Ed. 2d 837,

I cu dacvensccevcdes 5,6

Ford v. Wainwright, 477 U. S. 399, 91 L. Ed. 2d 335,

ee eee 15, 27

Francis v. Henderson, 425 U. S. 536, 48 L. Ed. 2d 149,

ee, cae b awed e's c's ee 6

Griffith v. Kentucky, 479 U. S. 314, 93 L. Ed. 2d 649,

I gs ot ck oes See bee eet 14, 15

Harris v. Reed, 109 S. Ct. 1038, 103 L. Ed. 2d 308 (1989) ... 7

Hilton v. Braunskill, 481 U. S. 770, 95 L. Ed. 2d 724,

ee SSE ae 23

In re Winship, 397 U. S. 358, 25 L. Ed. 2d 368,

aS 9

Jackson v. Virginia, 443 U.S. 307, 61 L.Ed.2d 560,

ne 30

Johnson v. New Jersey, 384 U. S. 719, 16 L. Ed. 2d 882,

UIE bG ers ck cee cdedeecccvcs 14

Johnson v. Zerbst, 304 U. S. 458, 82 L. Ed. 1461 (1938) . . . .23

Jurek v. Texas, 428 U. S. 262, 49 L. Ed. 2d 929,

I ci ccc clbccadeccceces 16

Linkletter v. Walker, 381 U. S. 618, 14 L. Ed. 2d 601,

ns a ceccebccces 14

Lockett v. Ohio, 438 U. S. 586, 57 L. Ed. 2d 973,

98 S. Ct. 2954 (1978) ... 2.0.00... 3, 14, 16, 20, 21, 29

Mackey v. United States, 401 U. S. 667, 28 L. Ed. 2d 404,

I ee Aas os 15

Martin v. Ohio, 480 U. S. 228, 94 L. Ed. 2d 267.

107 S. Ct. 1098 (1987)... 2... ee ee eee eee. 10

Matter of Spencer, 228 U. S. 652, 57 L.Ed. 1010 (1912) ... . 5

Miranda v. Arizona, 384 U. S. 436, 16 L. Ed. 2d 694,

404 oso © 6 WMG ul Awe 0:0 0's be 14

Murray v. Carrier, 477 U. S. 478, 91 L. Ed. 2d 397,

PE coca cavccedeccses< 6, 8, 23, 25

Penry v. Lynaugh, 109 S. Ct. 2934,

ee Passim

Quinones v. State, 592 S. W. 2d 933, cert. denied,

449 U.S. 893 (Tex. Crim. App. 1980) ............. 20

Reed v. Ross, 468 U.S. 1, 82 L. Ed. 2d 1,

CF EE ee eee Passim

Rock v. Arkansas, 483 U. S. 44, 96 L. Ed. 2d 37,

GE 15

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

rs hs hn a a eed & an 60.6 © « 14, 22

Selvage v. Lynaugh, 842 F. 2d 89 (Sth Cir. 1988) ........ 4

Selvage v. State, 680 S. W. 2d 17 (Tex. Crim. App. 1984) .. . 3

Smith v. Murray, 477 U. S. 527, 91 L. Ed. 2d 434,

PER EOE & ecw cee pees wees 8, 18, 25, 26

Stanford v. Kentucky, 109 S. Ct. 2969,

Ee ee ee eee 27

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

re as. he a ei ah Wee oes ole 22

Stovall v. Denno, 388 U. S. 293, 18 L. Ed. 2d 1199,

—. . + SE a ce, 14

Sumner v. Shuman, 483 U. S. 66, 96 L. Ed. 2d 56,

IE DS's > uc 0 6 ule Uru 6 alle lee Bie 16

Teague v. Lane, 109 S. Ct. 1060,

mF ee Passim

Thompson v. Oklahoma, 108 S. Ct. 2687,

TE Gn can uaweeecesccceces 28

Tison v. Arizona, 481 U. S. 137, 95 L. Ed. 2d 127,

Dy ciegucavcesesnevcecs 27, 29

United States v. Frady, 456 U. S. 152, 71 L. Ed. 2d 816,

EE ee eee ae eee 33

United States v. Johnson, 457 U. S. 537, 73 L. Ed. 2d 202,

ea eee 11

Wainwright v. Sykes, 433 U. S. 72, 53 L. Ed. 2d 594,

Ee ee eee 6

In re Wood, 140 U. S. 278, 35 L. Ed. 505(1891) ........ 5

Woodson v. North Carolina, 428 U. S. 280, 49 L. Ed. 2d 944,

nn oo: be one ee eas eh 60 0 6 6 6:8 16

United States Constitution

8 0 Ee 23

Treatises

1 W. Blackstone, Commentaries (1765) ............. 14

1 W. LaFave and A. Scott, Substantive Criminal Law,

eS re ae ee ee eee ee 10

Miscellaneous

ABA Criminal Justice Section, Report to the House of

PD ewes ecceetweseeeeeesecesneces 6

Friendly, Is Innocence Irrelevant? Collateral Attack on

Criminal Judgments, 38 U. Chi. L. Rev. 142 (1970) ..... 22

Lucas, Minority Report of ABA Task Force Co-Chairman

a i ia es

Robbins, Rationalizing Federal Habeas Corpus Review of

State Court Crimina! Convictions in Capital Cases (1989)

Scheidegger, Capital Punishment in 1987: The Puzzle Nears

Completion, 15 West. St. L. Rev. 95 (1987) .........

U.S. Bureau of Justice Statistics, Sourcebook or Criminal

Justice Statistics — 1988

me Cee Se 2 a. eS 2 2 ea ea Bee a

ae

IN THE

SUPREME COURT OF THE UNITED STATES

OcTOBER TERM, 1989

JOHN HENRY SELVAGE,

Petitioner,

vs.

JAMES LYNAUGH, Director,

Texas Department of Corrections,

Respondent.

On Writ of Certiorari to the United States

Court of Appeals for the Fifth Circuit

BRIEF AMICUS CURIAE OF THE

CRIMINAL JUSTICE LEGAL FOUNDATION

IN SUPPORT OF RESPONDENT

INTEREST OF AMICUS

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

profit California corporation organized to participate in litiga-

tion relating to the criminal justice system as it affects the

public interest.

1. CJLF has received written consent of the parties to file this brief.

CJLF seeks to bring the due process protection of the

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

ty to rapid, efficient and reliable determination of guilt and

swift execution of punishment.

The present case involves a collateral attack on a final

judgment based on a claim which defendant did not mention at

trial or on appeal and which is not a bedrock rule of fairness.

Such unecessary attack on the finality of criminal judgments

is contrary to the rights of victims and society which CJLF was

formed to advance.

SUMMARY OF FACTS AND CASE

On July 26, 1979, petitioner John Henry Selvage and his

accomplice Wilbur Kelly robbed a convenience store in Hous-

ton. One of the robbers, it is not known which, shot a custom-

er to death. Brief for Petitioner (“Pet. Brief”) 9-10.

Four days later, Selvage and Kelly robbed Ventura’s Jewel-

ry in Houston. One of them shot and killed Deputy Sheriff Al

Garza. Although off duty, Garza had prominently displayed

his badge prior to the shooting. It is not known which robber

fired the fatal shot. /d., at 7. After the shooting of Garza and

while they were fleeing the scene, both robbers shot at Ste-

phen Ventura, the owner of the store. /d., at 5-6. Selvage was

convicted of the murder of Deputy Garza.

In the penalty phase, the prosecution presented evidence

of the convenience store robbery and murder, of Selvage’s

prior record of drug possession and conspiracy to commit bank

robbery and of his attempted escape from jail. Jd., at 8-10.

The defense presented evidence of Selvage’s good upbringing,

of the apparently detrimental effect of his military experience

and of his potential for rehabilitation. /d., at 10-12.

The jury was instructed on the two relevant Texas special

issues, which focus first on deliberateness and expectation that

death will result and second on future dangerousness of the

defendant. J. A. 4. The jury answered both issues “yes,” ibid.,

and Selvage was accordingly sentenced to death. On direct

appeal to the state’s highest criminal court, the judgment was

affirmed. Selvage v. State, 680 S. W. 2d 17 (Tex. Crim. App.

1984).

Defense counsel did not accept the instruction in the bare

words of the statute as given. He specifically requested a sup-

plemental instruction to the effect that a non-triggerman was

completely exempt from capital punishment. J. A. 5-6. At no

time during trial Or direct appeal, however, did the defense

claim that the instructions were deficient for not allowing

mental condition or accomplice status to be considered as

mitigating factors. See J. A. 30-32; Selvage v. State, supra

(enumerating the grounds of error claimed). The trial was

conducted approximately two years after Lockett v. Ohio, 438

U. S. 586 (1978) and two years before Eddings v. Oklahoma,

455 U. S. 104 (1982). The appeal was decided two years after

Eddings.

In 1985, seven years after Lockett, Selvage filed a habeas

corpus petition in the federal district court. J. A. 70. This

petition made no claim that the jury had not been permitted to

consider mitigating factors. J. A. 46. The petition was denied,

and the denial was affirmed. J. A. 70.

On March 3, 1987, while federal proceedings were pending,

Selvage filed a state habeas petition, which was denied. J. A.

9. On March 22, 1988, ten years after Lockett but a year be-

fore Penry v. Lynaugh, 109 S. Ct. 2934, 106 L. Ed. 2d 256

(1989), Selvage filed a state habeas petition raising for the first

time the mitigation issue. The state trial court found the issue

procedurally barred, J. A. 34-35, and the Court of Criminal

4

Appeals adopted the trial court’s findings and conclusions.

J. A. 38.

On Selvage’s second federal petition, the state objected to the

mitigation question on grounds of both abuse of the writ, J. A.

45-50, and procedural bar, J. A. 51-52. The district court dis-

missed this claim but granted a stay on another claim. J. A. 63-

66. The Fifth Circuit Court of Appeals vacated the stay.

Although anticipating the holding of Penry, supra, that the

instructions might be found deficient, the court held that the

claim was procedurally barred. J. A. 81; Selvage v. Lynaugh,

842 F. 2d 89, 95 (Sth Cir. 1988).

SUMMARY OF ARGUMENT

Selvage contends that he had “cause” not to raise his claim

earlier, thus permitting consideration on federal habeas, be-

cause his claim was “novel” at that time within the meaning of

Reed v. Ros, 468 "|. S. 1 (1984). This contention canrot be

reconcileé with this Court’s decision in Penry v. Lynaugh, 109

S. Ct. 2936 106 L. Fu. 2d 256 (1989) thai the decision in Penry

did not annow ce a “new” rule within the meaning of Teague

v. Lane, 109 S. Ct. 1060, 103 L. Ed. 2d 334 (1989).

The degree of change from prior precedent which is re-

quired for “novelty” as “cause” under Reed is substantially

greater than the degree of change which invokes the nonretro-

activity bar of Teague. Therefore, absent one of three excep-

tions not present in this case, Reed is a dead letter after

Teague. Any claim which is sufficiently novel that counsel did

not even have “the tools to construct” it at the time of trial

must necessarily not be “dictated by precedent existing at the

time.”

The “fundamental miscarriage of justice” exception, in

order to fulfill its function, must be based on an undeniable

injustice of the result and not on the procedure by which that

result was obtained. Where guilt is not in question, the penal-

ty-phase application of this exception shov'd be limited to

those grounds on which there is an overwhelming national

consensus against capital punishment.

I. Stability and clarity are desperately needed in the

procedural default area.

The procedural default rule has a long and uneven history

in this Court. Impatience with petitioners who omit claims at

trial and on appeal and then raise them on federal habeas

corpus can be seen as far back as In re Wood, 140 U. S. 278,

288-289 (1891) and Matter of Spencer, 228 U. S. 652 (1912).

“If defenses may be omitted at trials, rights of review omitted,

and yet availed of through habeas corpus, the whole course of

criminal justice will be deranged and, it may be, defeated.” /d.,

at 661.

Stare decisis has not fared well in this area, as this Court has

swung from one extreme to the other. In Brown v. Allen, 344

U. S. 443, 485-487 (1953), this Court held with unmistakable

clarity that the bar of procedural default was jurisdictional and

mandatory as a corollary to the then-recently-codified exhaus-

tion rule, admitting no exceptions other than incapacity or

interference by the state. See also, Fay v. Noia, 372 U. S. 391,

461-462 (1963) (Harlan, J., dissenting). Yet Brown was over-

ruled sub silento in Fay v. Noia, supra.

Fay swung to the opposite extreme, substantially abolishing

the procedural default rule by limiting it to cases of “deliberate

by-passing of state procedures.” /d., at 439. Yet as applied in

Noia’s case, this rule was no rule at all. Noia had intentionally

decided not to appeal. He said it was to avoid financial burden

to his family; his lawyer said it was to retain the benefit of his

6

lenient sentence of life. Jd., at 396-397, n. 3. Either way, it is

difficult to conceive of a bypass more deliberate.

In a series of decisions since Fay, this Court has applied a

“cause and prejudice” test to defaults occurring before, during

and after trial. Francis v. Henderson, 425 U. S. 536 (1976) (no

challenge to grand jury composition); Wainwright v. Sykes, 433

U. S. 72 (1977) (no contemporaneous objection to evidence at

trial); Murray v. Carrier, 477 U. S. 478 (1986) (omission from

appeal). Although not explicitly overruled, there is nothing

left of Fay’s deliberate bypass test. The cause and prejudice

test now applies across the board. Amadeo v. Zant, 108 S. Ct.

1771, 1776, 100 L. Ed. 2d 249, 259-260 (1988).

The cause and prejudice rule of Wainwright v. Sykes, suprc,

is thus the resultant vector of two vehemently opposing forces.

The law of procedural defaults has swung wildly to one ex-

treme and then to the other and finally settled on a moderate

course. The product of such a long and arduous battle should

command special respect lest we begin the whole cycle again.

The basic rule should not be so riddled with exceptions that

the exceptions swallow the rule.

The Sykes rule has not been without its detractors. A criti-

cism has been strongly voiced in recent months that the indef-

inite boundaries of the rule and its exceptions make the rule

more complicated than the merits of the petition, delaying and

complicating the habeas proceeding. Chief Judge Donald P.

Lay of the Eighth Circuit expressed this view in recent hear-

ings before the Senate Judiciary Committee.” The ABA

Criminal Justice Section has voiced a similar criticism. ABA

2. November 8, 1989. As of this date there does not appear to be a

published transcript of this testimony.

7

Criminal Justice Section, Report to the House of Delegates

24-26 (“ABA Report”).°

There are three possible solutions to the complication of

the rule: (1) abandon the rule and return to Fay v. Noia; (2)

abandon the exceptions and return to Brown v. Allen; or (3)

clarify the rule and limit the exceptions. The opponents of

capital punishment naturally call for restoration of Fay. See

Robbins, Rationalizing Federal Habeas Corpus Review of

State Court Criminal Convictions in Capital Cases 71-73

(1989) (background paper for ABA Task Force). Others favor

the third alternative. See Lucas, Minority Report of ABA

Task Force Co-Chairman 11 (1989) reprinted in ABA Report,

supra, A-2, A-12. Whatever may be the legislative outcome of

this debate, the compelling stare decisis considerations dis-

cussed above point clearly to the third option as the path of

judicial development.

The scope of the rule itself was significantly clarified last

term. If the claim was presented to the state court, it is only

barred if the state court makes a plain statement to that effect.

Harris v. Reed, 109 S. Ct. 1038, 103 L. Ed. 2d 308 (1989). If

not presented to the state courts, it is barred if state law clearly

bars it. See Teague v. Lane, 109 S. Ct. 1060, 1067-1069, 103 L.

Ed. 2d 334, 346-348 (1989). Habeas claims never presented to ~

the state courts are not exhausted, however, unless “it is clear

that [the] claims are now procedurally barred under [state]

law.” Castille v. Peoples, 109 S. Ct. 1056, 1060, 103 L. Ed. 2d

380, 387 (1989) (italics added). If a claim is not exhausted, the

petition should be dismissed on that ground. /d., at 1059, L.

Ed. 2d at 385. The state court proceeding which follows dis-

3. This report bears no title other than the above; it is dated February,

1990, even though it was announced in a press release dated November

27, 1989. It is available from the ABA Criminal Justice Section, Wash-

ington, D. C.

missal for nonexhaustion should produce the plain statement

required by Hams.

Other cases have done much to clarify what constitutes

“causé.” The quintessential “cause” is intentional conceal-

ment by state officials of the evidence underlying the claims.

See Amadeo v. Zant, 108 S. Ct. 1771, 1777, 100 L. Ed. 2d 249,

260 (1988). Ineffective assistance of counsel is also cause.

Murray v. Carrier, 477 U. S. 478, 488 (1986). However, the

failure of effective counsel to recognize an available factual or

legal basis for a claim is not cause. /Jd., at 486. Neither is an

inadvertent failure to raise the issue on appeal, id., at 488, nor

a deliberate tactical decision not to pursue a claim, Smith v.

Murray, 477 U. S. 527, 534 (1986).

There remain three principal areas of confusion. The first

is when a state procedural bar is not “adequate” despite a

plain statement of the state court. The second is the use of

novelty of the claim as cause. The third is the scope of the

“fundamental miscarriage of justice” exception as applied to

sentencing. This case involves the second and third issues,‘

and it can clear up much of the complexity and confusion.

II. Only genuinely new rules are sufficiently novel tc

constitute “cause” under Reed v. Ross.

The issue of novelty of the argument as “cause” under

Sykes is bracketed by the decisions in Engle v. Isaac, 456 U. §

107 (1982) and Reed v. Ross, 468 U.S. 1 (1984). Both cases

involved claims that jury instructions had unconstitutionally

shifted the burden of proof to the defendant.

4. The first issue is not mentioned in either the questions presented in the

petition for certiorari or this Court's restatement of question | in the

order granting certiorari.

9

In Engle, the defendants claimed self-defense. The juries

had been instructed, without objection, that the defense had

the burden of proof. The Ohio Supreme Court later decided,

as a matter of statutory interpretation, that the prosecution

had the burden of proof, 456 U. S., at 111, but upheld its state

procedural bar against defendants who had not objected at

trial, id., at 116. Isaac and the others contended on federal

habeas that regardless of the interpretation of the statute

“Ohio could not constitutionally shift the burden of proving

self-defense to them.” /d., at 121.

Isaac put forth precisely the same reasons for default which

Selvage proffers in the present case: novelty and futility. /d.,

at 130. The Court dealt with futility in no uncertain terms.

Futility “cannot alone constitute cause ....” I/d., at 130 (ital-

ics added). No matter how solid the wall of authority, the state

court must be presented with the opportunity to overrule that

authority before that argument can be the basis of a federal

claim that the prisoner is unconstitutionally detained.

While Isaac’s claita of novelty was also rejected, that rejec-

tion was not categorical. /d., at 13i. The test adopted was

whether the tools to construct the constitutional claim were

available to the defense. /d., at 133. The necessary tool had

been provided years before the trial by /n re Winship, 397 U. S.

358 (1970). Engle, at 131.

To evaluate this “tools” test, it is important to recognize

how broad a gap exists between Winship and Isaac's claim.

The holding of Winship is “that the Due Process Clause pro-

tects the accused against conviction except upon proof beyond

a reasonable doubt of every fact necessary to constitute the

crime with which he is charged.” Winship, 397 U. S., at 364.

isaac did not deny that he had knowingly caused serious physi-

cal harm to another, the facts constituting the crime. Engle, at

114, n. 12. His defense was that his actions were justi ied by

self-defense. /d., at 114. There is an important difference

10

between requiring the state to prove the offense and requiring

it to disprove a defense. Martin v. Ohio, 480 U. S. 228, 231-

234 (1987) explains this difference at length.

For an earlier case to have provided the tools to construct a

claim, then, the claim need not follow directly or inevitably

from that case. Reasonable people may differ as to whether

the case supports the claim. There may be contrary precedent

on point. See Engle, 456 U. S., at 133, n. 41.

The claim need not be so obvious that every competent

attorney raises it. If significant numbers of lawyers are making

the claim, even though many or most are not, then the tools

exist. In that event, there is no “cause” for not raising the

claim. /d., at 133-134.

In Reed v. Ross 468 U. S. 1 (1984), the jury in a murder case

was instructed that use of a gun raises a presumption of malice

shifting the burden of proof to the defendant. /d., at 6-7. The

argument against this instruction would seem on its face to be

considerably /ess novel than the argument advanced in Engle.

Malice is traditionally an element of murder to be proved by

the prosecution, while self-defense is the kind of affirmative

defense which many jurisdictions have required the defendant

to prove. See i W. LaFave and A. Scott, Substantive Criminal

Law § 1.8(c), 71-78 (1986). The Reed court distinguished

Engle by the fact that the trial in Reed had occurred before

Winship. Reed, 468 U. S., at 19-20.

Novelty will constitute cause “where a constitutional claim

is SO novel that its legal basis is not reasonably available to

counsel.” /d., at 16. The most common instance of a claim

without a pre-existing “reasonable basis” is where “this Court

has articulated a constitutional principle that had not been

previously recognized but which is held to have retroactive ap-

plication.” Id., at 17 (italics added). Reed then equates “not

previously recognized” with the “clear break with the past”

11

test. Ibid. This “clear break” language is taken from United

States v. Johnson, 457 U. S. 537 (1982), where the Court indi-

cated that “clear break” cases are “decisions whose nonretro-

activity is effectively preordained.” Id., at 553-54.

Reed then identifies three categories of clear breaks, still —

relying on Johnson. The simplest and most obvious is a deci-

sion of the Supreme Court overruling one of its own prece-

dents. 468 U.S., at 17. “Second, a decision may ‘overtur[n] a

longstanding and widespread practice to which this Court has

not spoken, but which a near-unanimous body of lower court

authority has expressly approved.’ ” IJbid., quoting Johnson

(italics added). These two categories should cause little diffi-

culty for reasons we will explain below.* ©

The principal problem lies in the third category, where a

“decision may ‘disapprovie] a practice this Court arguably has

sanctioned in prior cases.’ Jbid., quoting Johnson (italics

added). This is a very vaguely defined category. Given the

volume of this Court's opinions, the divergence of views

among its Members and the creativity of lawyers in extrapolat-

ing from opinions, a vast array of practices might be said to

have been arguably sanctioned.

The test of novelty applied to this category is even more

vague.

“Whether an attorney had a reasonable basis for press-

ing a claim challenging a practice that this Court has

arguably sanctioned depends on how direct this

Court's sanction of the prevailing practice had been,

how well entrenched the practice was in the relevant

jurisdiction at the time of defense counsel's failure to

challenge it, and how strong the available support is

S. See part IV A at p. 11, infra.

12

from sources opposing the prevailing practice.” Jd., at

17-18.

The litigation-breeding potential of this test is apparent. It

bears more than a passing resemblance to the “test” for retro-

activity which this Court has since abandoned as unworkable.

See Teague v. Lane, 109 S. Ct. 1060, 1070-1072, 103 L. Ed. 2d

334, 350-352 (criticism of the Linkletter standard).

The threads of retroactivity are woven throughout the fab- ~

ric of Reed v. Ross. They show clearly in all three of the opin-

ions filed in that case. The Reed majority noted that retroac-

tivity was a distinguishing characteristic of the primary catego-

ry of cases to which its rule applied. Reed, supra, 468 U. S.., at

17. The majority also lifted its “clear break” test directly out

of retroactivity iaw. bid.

Justice Powell rested his deciding vote on the state’s own

procedural default in not raising the retroactivity issue. /d., at

20. Had the state only raised the issue, Reed would have gone

the other way.

The dissent noted the paradoxical result of considering

“novelty” as “cause” without considering retroactivity:

“Consequently, we have the anomalous situation of a

jury verdict in a case tried properly by then-prevailing

constitutional standards being set aside because of

legal developments that occurred long after the North

Carolina conviction became final.

“Like the Court of Appeals, the Court proposes to

adopt ‘novelty’ as a possible form of ‘cause’ under

Wainwright v. Sykes to justify ignoring the State's

procedural default rule. But this equating of novelty

with cause pushes the Court into a conundrum which it

refuses to recognize. The more ‘novel’ a claimed con-

stitutional right, the more unlikely a violation of that

13

claimed right undercuts the fundamental fairness of

the trial.” /d., a 21-22 (Rehnquist, J., dissenting).

The correction to the anomaly and the solution to the

conundrum, we submit, is to always consider retroactivity with

any Reed claim. When this is done within the structure of

Teague v. Lane, 109 S. Ct. 1060, 103 L. Ed. 2d 334 (1989), the

difficulties anticipated by the Reed dissent disappear, and the

question of fundamental fairness will become central to the

decision.

III. New procedural rules are almost never retroactive

on habeas under Teague.

The essence of a habeas petitioner’s Reed claim is that the

law has changed in an unexpected way and that he should not

be “punished” for a lack of clairvoyance. But are the people

of a state not entitled to make the same claim? Should a

judgment that was perfectly legal when entered be set aside

because a court did not follow a procedure which it had no

obligation to follow at the time? The people’s side of the

unexpected change ledger is the issue of retroactivity. The

crux of this case is the inconsistency between defendant's claim

of novelty and the holding of Penry v. Lynaugh, 109 S. Ct.

2934, 106 L. Ed. 2d 256 (1989) on the retroactivity question.

Teague v. Lane, 109 S. Ct. 1060, 103 L. Ed. 2d 334 (1989)°

is the culmination cf this Court’s long and difficult struggle

with the question of how far back a new decision should reach

in undoing the work of judges, lawyers and juries who had duly

obeyed the rules in effect at the time. Whenever a rule is not

applied retroactively, the contention is made that the Court

6. Teague is a plurality opinion, but its rule was accepted by a majority in

Penry v. Lynaugh, 109 S. Ct. 2934, 2944, 106 L. Ed. 2d 256, 274 (1989).

14

has sanctioned a constitutionally flawed conviction or denied

someone his constitutional rights. See, e.g., Johnson v. New

Jersey, 384 U. S. 719, 736 (1966) (Black, J., dissenting). That

would be true if one subscribed to the Blackstonian belief that

the Court merely discovers pre-existing law and does not make

new law. See 1 W. Blackstone, Commentaries 69-70 (1765).

However attractive the Blackstone approach may be in theory,

it loses any connection with reality when applied to require-

ments as far removed from the text and history of the Consti-

tution as the Miranda warnings, the exclusionary rule or the

rule of Lockett v. Ohio, 438 U. S. 586 (1978).

There are, no doubt, people in prison today who were con-

victed before Miranda v. Arizona, 384 U. S. 436 (1966) based

largely on voluntary confessions obtained without the now-

required warnings. Yet their continued incarceration is not

the result of any denial of any constitutional right. The pris-

oner was legally convicted because the right to warnings simply

did not exist at the time; this Court created that right in Mi-

randa. There may be doubts about the legitimacy of creating

“constitutional” rights so far removed from either the text or

fundamental fairness, see Rose v. Lundy, 455 U. S. 509, 543-

544, n. 8 (1982) (Stevens, J., dissenting), but no realistic ap-

praisal can deny that that is what the Miranda Court actually

did.

With the recognition that some rules are genuinely “new”

came a need to decide how far back to apply these new rules.

A three-part test to be applied separately to ench new rule was

laid down in Linkleiter v. Walker, 381 U.S. 618 (1965) and

Stovall v. Denno, 388 U. S. 293, 297 {1967). This test was

heavily dependent on the value judgments of the judges apply-

ing it, however, and application of it was inconsistent. Sec

Teague, supra, 109 S. Ct., at 1072, 103 L. Ed. 2d, at 352.

In Griffith v. Kentucky, 479 U. S. 314 (1987) and Teague v.

Lane, supra, the Linkletter-Stovall approach was replaced by

15

the approach suggested by Justice Harlan in Desist v. United

States, 394 U. S. 244, 256-269 (1969) (dissent) and Mackey v.

United States, 401 U. S. 667, 675-702 (1971) (separate

opinion). All new rules now apply to all cases on direct review,

Griffith, 479 U. S., at 323, and with two rare exceptions do not

apply to cases which became final prior to the announcement

of the rute, Teague, 109 S. Ct., at 1075, 103 L. Ed. 2d, at 356.

The first step in a Teague analysis is to determine whether a

rule is “new.” “In general . . . a case announces a new rule

when it breaks new ground or imposes a new obligation on the

States or the Federal Government.” /d., at 1070, L. Ed. 2d, at

349. The examples given illuminate the meaning of this sen-

tence. Rock v. Arkansas, 483 U. S. 44 (1987) broke new

ground by extending the right to present a defense, established

in Chambers v. Mississippi, 410 U. S. 284 (1973), into the

previously uncharted territory of hypnotically refreshed testi-

mony. Ford v. Wainwright, .'77 U. S. 399 (1986) imposed a new

obligation on the states by constitutionalizing the long-

estabuished common-law prohibition of execution of the in-

sane. Neither claim was novel. Chambers and the common

law provided ample tools to construct the respective claims.

Yet both are “new rules” under Teague.

The Teague Court's alternative formulation of newness is

more emphatic. “To put it differently, a case announces a new

rule if the result was not dictated by precedent existing at the

time the defendant’s conviction became final.” 109 S. Ct., at

1070, 103 L. Ed. 2d, at 349 (italics in original). Because no

precedent existing at that time could be said to dictate that the

fair cross-section requirement extended to the petit jury, such

an extension would be a new rule. /bid.

Penry v. Lynaugh, 109 S. Ct. 2934, 106 L. Ed. 2d 256 (1989)

presented an unusual twist on the newness problem. The

difficulty in the question presented in Penry was not caused by

a dearth of authority but rather by an excess. A tension exist-

16

ed between two lines of cases decided by this Court, see Cali-

fornia v. Brown, 479 U. S. 538, 544 (1987) (O’Connor, J.,

concurring), and either side could construct a reasonable

argument using one of the two lines.

In Woodson v. North Carolina, 428 U. S. 280, 304 (1976), a

plurality held that capital sentencing must take into account

relevant facets of the defendant's character and record, includ-

ing “compassionate or mitigating factors stemming from the

diverse frailties of humankind.” In Lockett v. Ohio, 438 U. S.

586, 604 (1978), the plurality found a constitutional require-

ment that “the sentencer . . . not be precluded from consider-

ing as a mitigating factor, any aspect of a defendant's character

or record and any of the circumstances of the offense that the

defendant proffers as a basis for a sentence less than death.”’

Eddings v. Oklahoma, 455 U. S. 104, 110 (1982) raised the

Lockett plurality rule to majority status.

Taking these authorities alone, the Penry case would have

been a foregone conclusion. Asking the jurors to decide if the

murder was deliberate, if the defendant will be dangerous in

the future and if the defendant overreacted to any provocation

did noi permit consideration as mitigation of Penry’s abused

childhood, precisely the type of mitigation that Eddings had

heid must be considered. 455 U. S., at 113-115. The problem

was that the Texas system of ashig only these three questions

had been sustained against a facial attack in Jurek v. Texas, 428

U. S. 262 (1976). Four Justices of this Court believed that

Jurek and not Eddings dictated the result. Penry, 10° S. Ct., at

2965, 106 L. Ed. 2d, at 300 (Scalia, J., dissenting in part).

7. Ttaiics in original. A footnote at this point reserves the life prisoner

question subsequently decided in Sumner v. Shuman, 483 U.S. 66

(1987).

17

The “newness” issue in Penry and the merits of the decision

thus both turn on the interpretation of Jurek. Penry says that

Jurek did not give a blanket endorsement to the Texas manner

of instructing the jury without regard to whether those instruc-

tions permitted consideration of all mitigating evidence prof-

fered in a particular case. Rather, says Penry, Jurek was based on

an assurance, not fulfilled in Penry’s case, that all such evidence

would be considered. /d., at 2946-2947, L. Ed. 2d, at 278. If Penry

is new then it is also wrong on the merits.® If Penry is correct on

the merits, then its interpretation of Jurek must also be correct.

Penry therefore does not “gut” Teague, see id., at 2965, L. Ed. 2d,

at 301 (Scalia, J., dissenting in part), but rather leaves it un-

touched. A rule is new if it is not dictated by prior precedent.

Accepting the Penry majority’s interpretation of Jurek, Penry is

indeed dictated by Lockett-Eddings.

IV. Any claim which is “novel” under Reed is

necessarily “new” under Teague.

A. The Basic Rule of Preclusion.

When the definition of “new rules”

is juxtaposed with the definition of

“novel claims” we can see that “novel” “Cae

is a subset of “new.” That is, any claim

which is “novel” under Reed must

necessarily be “new” under Teague, at

least if newness and novelty are deter-

mined as of the same time. The con-

verse, however, is not necessarily true.

Most claims will cease to be novel under Reed long before the

rules they propose cease to be new under Teague.

8. Defendant's interpretation of Jurek, Pet. Brief 57-58, is essentially the same

as that of the Penry dissent.

18

Returning to the three categories of potentially novel

claims outlined in Reed v. Ross, 468 U.S. 1, 17 (1984), the first

category is a square overruling by this Court of one of its own

precedents. While Reed correctly notes that a claim based on

the overruling case would be novel, ibid., there can also be no

doubt that such a claim would be new under Teague. See

Teague, 109 S. Ct., at 1067, 103 L. Ed. 2d, at 345-346 (reaffirm-

ing nonretroactivity on habeas of Batson v. Kentucky, 476 U. S.

79 (1986)).

The second category requires that the disapproved practice

had been both longstanding and widespread and had been

approved by a nearly unanimous body of lower court opinion.

Reed, at 17. Again, there can be no question that the disap-

proval of such a body of case law would be a new rule under

Teague.”

The present case fits in the third category: disapproval of a

practice arguably sanctioned by past Supreme Court cases.

Ibid. The prior sanction of the disapproved practice is certain-

ly arguable if the argument was accepted by four Justices of

this Court, see Penry v. Lynaugh, 109 S. Ct. 2934, 2965-2967,

106 L. Ed. 2d 256, 301-303 (1989) (Scalia, J., dissenting in

part). Yet arguable sanction alone is not enough for novelty.

Reed also considers directness of the sanction, entrenchment

of the practice in the relevant jurisdiction and strength of

available support for opposing the pri. -tice.

Comparing Reed with Engle v. Isaac, 456 U.S. 107 (1982),

it is apparent that the entrenchment factor cannot alone make

a claim novel. Engle held that even complete futility cannot

alone constitute cause, id., at 130, and this holding was reaf-)

firmed after Reed in Smith v. Murray, 477 U. s $27, 535

9. The present case does not fall into this category, as the Tex.)s practice is

not widespread. Defendant says it is unique. Pet. Brief 65.

19

(1986). Reed also considers the directness of this Court’s sanc-

tion of the practice, but if the sanction were truly direct the

case would fall into the first category (outright overrul.ag) and

not the third. :

The principal focus, then, must be on “how strong the

available support is from sources opposing the prevailing prac-

tice.” Reed, 468 U.S., at 17-18. This focus is consistent with

Engle and Smith, because the basic question is whether de-

fense counsel had the tools to construct the claim, Engle, 456

U. S., at 133, not whether the prosecution had the tools to tear

it down. Here the inconsistency of the defense’s position

becomes glaring.

If a rule is dictated by existing precedent of the Supreme

Court, then the available support from opposing sources is not

merely “strong”; it is by definition overpowering. A finding

that the rule is not new under Teague thus establishes the

strength factor of Reed to such a degree that it necessarily

overrides the other two factors. How can anyone say that a

result is dictated by precedents of the Supzeme Court and yet

maintain that counsel lacks the tools to construct that claim?

The argument to that effect really should be written by

Lewis Carroll, but defense counsel turns in a reasonable im-

personation. For evaluating whether a rule is new under

Teague, the argument goes, this Court should interpret its

prior opinions in the way it has in the past, but for the Reed

novelty evaluation the Court should jump through the look-

ing-glass into a reversed world where all the rules are differ-

ent. See Pet. Brief 76-79. Legal practitioners immersed in the

local system, we are told, are somehow excused from reading

and understanding this Court’s precedents. See id., at 79.

\

The argument just does not hold water. Whatever defi-

ciencies American legal education may have in other respects,

it does teach lawyers to do one task very well: to read, under-

20

stand and interpret appellate decisions. We all read the same

cases. Law school has made sure we kisow how. There is no

justification for the double standard defendant proposes.

Penry’s holding that its rule is not “new” for Teague pre-

cludes the contention that the claim is “novel” for Reed.

Conversely, a finding of novelty would require that Penry be

fully or partially overruled and that the rule established in that

case either overturned or at least not applied to any sentence

which became final before June 26, 1989. Either way, defend-

ant cannot claim the benefit of Penry.

B. The Timing Exception.

There is one exception worth noting to the general rule of

preclusion. Novelty for Reed and newness for Teague are not

determined as of the same time. Reed looks at the time of the

procedural default, 468 U. S., at 13, while Teague looks to the

date the conviction became final, 109 S. Ct., at 1075, 103 L.

Ed. 2d, at 356. It is remoiely possible that the law could take

such a radical turn in the interim that a claim which was una-

vailable at the time of he default had become dictated by

precedent by the date of finality.

This possibility need not detain the Court in the present

case, because no such radical turn in the law occurred between

the 1980 trial and the 1984 finality. The plurality opinion in

Lockett v. Ohio, 438 U. S. 586 (1978) provided the tools to

construct the claim in 1980, and Penry’s lawyer used them to

do exactly that. See Penry v. Lynaugh, 109 S. Ct. 2934, 2942,

106 L. Ed. 2d 256, 273 (1989); see also Quinones v. State, 592

S. W. 2d 933, 947 (Tex. Crim. App. 1980), cert. denied, 449

21

U. S. 893."°

Between trial and finality, the cnly important development

was the elevation of the Lockett plurality rule to majority sta-

tus in Eddings v. Oklahoma, 455 U. S. 104 (1982). The distinc-

tion between a plurality and a majority is important for deter-

mining when a rule-is dictated by existing prevedent,'" but it is

not important for deciding whether counsel had the tools to

construct the claim. No competent lawyer would abandon a

claim based on a decision of the Supreme Court merely be-

cause it was a plurality decision. In the death penalty area

especially, most of this Court’s jurisprudence has started out as

plurality decisions. See generally, Scheidegger, Capital Pun-

ishment in 1987: The Puzzle Nears Completion, 15 West. St.

L. Rev. 95, 107-110 (1987). Furthermore, Lockett contains a

solid majority on the need to consider Selvage’s principal miti-

gating factors of accomplice status and lack of intent to kill.

See Lockett, 438 U. S., at 615-616 (Blackmun, J., concurring in

the judgment). .

C. Fundamental v. Nonfundamental Rules.

If a rule must necessarily be new in order to be rovel, two

difficulties anticipated in Reed vanish. The Reed dissent noted

that “novel” claims were unlikely to affect the fundamental

fairness of the trial, 468 U. S., at 22 (Rehnquist, J., dissenting),

but a novel procedural claim can now apply on habeas only if it

10. Selvage’s trial and Penry’s took place in February and March,

respectively, of 1980. J. A. 8; 109 S. Ct., at 2941, 106 L. Ed. 2d, at 271.

Quinones was decided in January but did not become final until denial of

certiorari in October. The issue was very much alive at the time of

Selvage’s trial.

ll. Defe t assumes that Lockett and not Eddings is the cut-off date for

application of the Penry rule. Pet. Brief 76. Amicus disagrees, but this

case does not present that question.

22

qualifies for the second Teague exception; it must involve on

“absolute prerequisite to fundamental fairness” the absence of

which “seriously diminish[es] the likelihood of obtaining an

accurate conviction.” Teague v. Lane, 109 S. Ct. 1060, 1077,

103 L. Ed. 2d 334, 358-359 (1989). Substantive claims that an

act cannot be punished at all qualify for the other exemption.

Id., at 1073, L. Ed. 2d, at 353. Thus, the rovel but peripheral

claims anticipated by the dissent are excluded.

On the other hand, the Reed majority was concerned that

refusal to recognize novelty as cause would result in defense

counsel iaundati;:g the trial court with “any and all remotely

plausible constitutional claims that could, some day, gain rec-

ognition.” Reed, 468 U. S., at 15-16. If a claim of error is so

fundamental as to qualify for the second Teague exception,

competent counsel will surely raise it to preserve it for appeal,

regardless of what rules may apply on habeas. If the claimed

error is not so fundamental, Teague has eliminated any incen-

tive to object merely to preserve the claim for habeas.

The interaction of the procedural default rule and the

Teague nonretroactivity rule thus form another step in the

evolution of habeas corpus. That evolution, in recent years,

has begun to move away from relitigation ad infinitum of the

detailed code of criminal procedure n~w deemed to be incor-

porated in the Due Process Clause, se. ly, Is Innocence

Irrelevant? Collateral Attack on Crimina, :dgments, 38 U.

Chi. L. Rev. 142, 155-156 (1970), and toward a focus on fun-

damental fairness and actual innocence. Teague removed

retroactivity for nonfundamental procedural claims. Stone v.

Powell, 428 U. S. 465 (1976) eliminated most of the exclusion-

ary rule claims. Some Members of the Court would go further.

See Rose v. Lundy, 455 U. S. 509, 547 (1982) (Stevens, J., dis-

senting) (eliminate nonfundamental claims altogether); Duck-

worth v. Eagan, 109 S. Ct. 2875, 2881, 106 L. Ed. 2d 166,179'

(1989) (O’Connor, J., concurring) (eliminate Miranda claims).

23

Murray v. Carrier, 477 U. S. 478. 496 (1986) recognized the

importance of actual innocence.

Recognition of the rule amicus proposes would be a signifi-

cant step along this path. Claims of fundamental unfairness

raising serious doubt of the guilt of the accused could go for-

ward under Reed regardless of procedural default, but periph-

eral claims dreamed up only after a defendant's main conten-

tions have failed would always be barred by either the proce-

dural default rule or the nonretroactivity rule.

D. Procedural Default v. “Merits.”

Because debate on the proce dural default rule usually

invokes calls to forget procedure and go to the “merits,” a few

words on that point may be in order. Although the procedural

default rule is based on the procedure at trial and on appeal, it

does go to the merits of the habeas petition. A habeas pro-

ceeding is not an appeal. United States v. Frady, 456 U.S. 152,

165 (1982). It is a civil lawsuit by a prisoner against officers of

the state, accusing them of violating the Constitution, laws or

treaties of the United States. 28 U. S. C. § 2254(a); Hilton v.

Braunskill, 481 U.S. 770, 776 (1987). The state is no longer

accusing the petitioner; that proceeding is ove; and final. The

petitioner is accusing the state. That is why he has the burden

of procf. See Johnson v. 7#/>st, 304 U. S. 458, 468-469 (1938).

The procedural defaw't rule is not a forfeiture or penalty

imposed on the accusing petitioner; it is an affirmative defense

afforded to the accused state. Virtually all federal habeas

claims today - ~2 ultimately based on the requirement that a

state not “deprive any person of life, liberty, or property, with-

out due process of law.” U.S. Const. amend. XIV, § 1. This

clause, along with all that is deemed incorporated in it, does

not require state trial judges to scrutinize their procedures sua

sponte for every creative chailenge that might some day find

Ze

favor with this Court. The state may properly place the bur-

den of raising most objections on the defendant.

If the defendant has been provided with an effective attor-

ney, if the trial is conducted in accordance with the bedrock

elements of procedural fairness which are central to an accu-

rate determination of guilt,'? and if the trial judge and appel-

late court rule correctly on all objections made by the

defense,'* then the officials of the state have dore their job

correctly. That being the case, the fundamental element of

the habeas petitioner's cause of action is missing. He has re-

ceived the process due. His custody is not in violation of the

Constitution of the United States. The state court's “error” in

not sustaining an objection which the defendant never made

does not alter that conclusion. A ruling on the procedural bar

is thus a ruling on the merits of the habeas petition.

E. Conclusion.

A claim ceases tc Se novel under Reed v. Ross, 468 U.S. 1

(1984), at the very latest, when it ceases to be new under

Teague v. Lane, 109 S. Ct. 1060, 103 L. Ed. 2d 334 (1989). In

most cases it will have ceased to be novel long before. Absent

a radical shift in the law between the date of default and the

date of finality, a holding that a rule is not new under Teague

absolutely precludes any contention that a claim under that

rulc is novel under Reed. A rule which is new under Teague,

and therefore potentially novel under Reed, is retroactive on

habeas corpus only in two rare circumstances nt present in

12 If not, the second Teague exception applies. See 109 S. Ci, at 1076, 103

L. Ed. 2d, at 357. Furthermore, failure to object to a fundamentally

unfaw procedure in any modern inal would be ineffective assisiance.

13. A “correct” ruling is oe made in accordance with the precedents

established by this Court as of the date of finality.

25

this case. Teague, 109 S. Ct., at 1075-1077, 103 L. Ed. 2d, at

356-358.

Penry v. Lynaugh, 109 S. Ct. 2934, 106 L. Ed. 2d 256 (1989)

therefore cannot be the basis for overturning Selvage’s sen-

tence. He failed to raise the objection at trial, and his claimed

excuse for not objecting is irreconcilable with the holding of

Penry that the rule in that case applies to cases which became

final after Eddings v. Oklahoma, 455 U. S. 104 (1982) but

before Penry.

Vv. “Fundamental miscarriage of justice” must be based

on substance, not procedure.

Where the verdict itself is correct, ie., the defendant is

actually guilty, there is nothing unfair about a procedural bar

blocking a claim which is itself procedural. In Engle v. Isaac,

456 U. S. 107, 135 (1982), this Court expressed its confidence

that “victims of a fundamental miscarriage of justice will meet

the cause-and-prejudice standard.” In Murray v. Carrier, 477

U. S. 478, 496 (1986), the Court recognized that there might

be extraordinary cases “where a constitutional violation has

probably resulted in the conviction of one who is actually

innocent,” and recognized actual innocence as a separate

exception to the procedural bar rule, independent of the cause

and prejudice test.

The difficulty comes in applying this exception to capital

sentencing where guilt is not in issue. In Smith v. Murray, 477

U. S. 527 (1986), the purported error had to do with the de-

fendant’s admission of prior misconduct during a psychiatric

interview arranged by his own attorney. /d., at 529-530.

Acknowledging the difficulty of translating actual innocence

into the penalty phase, the Court conciuded that the admission

of the evidence “did not serve to pervert the jury's delibera-

tions concerning the ultimate question whether in fact peti-

26

tioner constituted a continuing threat to society.” /Jd., at 538

(‘alics in original).

The intended emphasis in this statement appears to be on

the jury’s reaching an acceptable result based on all of the

information available, even though some of that information

may have been illegally obtained. The reference to perverting

the jury’s deliberations, however, might be construed to open

up any claim of procedural error in instructing the jury, select-

ing the jury or arguing to the jury.

The latter interpretation was adopted by the dissent in

Dugger v. Adams, 109 S. Ct. 1211, 103 L. Ed. 2d 435 (1989).

Adams was sentenced to death for molesting and mutdering an

eight-year-old girl. Jd., at 1213, n. 2, L. Ed. 2d, at 441, n. 2.

The trial court instructed the jury that their recommendation

was only advisory. /d., at 1213, L. Ed. 2d, at 440. The dissent

claimed that this “alleged error is thus global in scope: it nec-

essarily pervades the entire sentencing process.” /d., at 1224,

L. Ed. 2d, at 453 (Blackmun, J., dissenting).

The Adams majority, however, rejected this process-based

approach to fundamental miscarriage of justice.

“Demonstrating that an error is by its nature the kind

of error that might have affected the accuracy of a

death sentence is far from demonstrating that an indi-

vidual defendant probably is ‘actually innocent’ of the

sentence he or she received. The approach taken by

the dissent would turn the case in which an error re-

sults in a fundamental miscarriage of justice, the ‘ex-

traordinary case,’ Carnier, supra, at 496, into an all too

ordinary one.” /d., slip op. at 11, n. 6, S. Ct., at 1218,

L. Ed. 2d, at 446.

This passage tells us what fundamental miscarriage of jus-

tice is not, but regrettably does not tell us what it is. The

answer, amicus submits, is to look solely at the substance of

27

the result, disregard all procedure and resolve all discretionary

questions in favor of the verdict. By doing so, we can expose

judicial interference based on disagreement with the sentencer

or the legislature.

For guilt phase error the test of fundamental miscarriage of

justice is actual innocence. Why? Because everyone agrees

that throwing an innocent man in prison is an injustice. There

are no questions of policy here. There is no room for discre-

tion. The purpose of our judicial system is to convict and

punish the guilty. No rational person believes we should con-

vict the innocent.

An injustice is fundamental, justifying an exception to the

procedural bar rule, when there is near unanimity among the

American people that people in the petitioner's situation

should not receive the punishment. On the other hand, a

sentence is not a miscarriage of justice at all, much less a fun-

damental one, if the sentence is a legal one for the crime

committed and if a reasonable sentencer might well impose

that sentence for that crime.

The perimeter of authorized capital semtencing has been

staked out by this Court with great care and with explicit ref-

erence to the beliefs of the American people. See Stanford v.

Kentucky, 109 S. Ct. 2969, 2975-2976, 106 L. Ed. 2d 306, 318-

319 (1989). Death is not a permissible sentence for rape,

Coker v. Georgia, 433 U. S. 584, 592 (1977) (plurality), and

presumably not for other nonfatal crimes against individuals.

Crimes against sovereignty, such as treason and espionage,

have yet to be considered. Death is also not permissible in the

case of a nontriggerman without either intent to kill, Enmund

v. Florida, 458 U. S. 782, 797 (1982), or major participation

plus reckless disregard for life, Tison v. Arizona, 481 U. S$. 137,

157-158 (1987). Several groups are immune from capital pun-

ishment: the insane, Ford v. Wainwright, 477 U. S. 399 (1986);

persons under 16, at least without a legislatively-set minimum

age, Thompson v. Oklahoma, 108 S. Ct. 2687, 101 L. Ed. 2d

702 (1988); and probably the severely or profoundly retarded,

see Penry v. Lynaugh, 109 S. Ct. 2934, 2954, 106 L. Ed. 2d 256,

287 (1989).

This extensive set of restrictions, combined with the state

Statutory system, defines which murderers are death-eligible.

If the defendant is in fact a murderer, if his crime does in fact

fit the state’s definition of a capital offense, and if neither he

nor his crime in fact comes within any of the excluded classes

above, then death is an authorized sentence which our society

has determined to be not cruel and not grossly disproportion-

ate. This detailed and exhaustive body of law, built up so paiti-

fully over so many years, has identified and excluded those

cases where a death sentence would be a fundamental miscar-

riage of justice. The remainder of cases are those where rea-

sonable people may differ as to whether death is the appropri-

ate punishment.‘ If a particularly hard-nosed sentencer

could legitimately impose the sentence following an absolutely

perfect trial and sentencing hearing, then that sentence is not

a fundamental miscarriage of jwetice.

The present case illustrates the point. Giving Selvage the

benefit of the doubt, we will assume that his accomplice, Kelly,

was the “triggerman” in both killings. See Pet. Brief 30-31, n.

16. By Selvage’s own admission, then, he robbed the conven-

ience store, and in the course of that robbery Kelly shot and

killed an innocent customer. A mere four days later, with the

actual knowledge that Kelly is a “wild man” who shoots people

14. The overwheiming majority of the American people favor capital pun-

ishmeyit for murder, including an absolute majority in every demographic

group suryeyed. U.S. Bureau of Justice Statistics, Sourcebook of

Criminal Justice Statistics — 1988 at 223. Twenty-nine percent favor it

in all murder cases, and eleven percent oppose it in all cases. /d., at 230.

29

in the course of his robberies, Selvage committed another

robbery with Kelly. Both of them were armed, and Selvage

personally fired his gun at a person. Pet. Brief at 6.

Although Selvage’s accomplice status is relevant, his degree

of participation and resulting culpability are far greater than

Earl Enmund’s, see Enmund v. Florida, 458 U. S. 782, 784-785

(1982) or Sandra Lockett’s, see Lockett v. Ohio, 438 U. S. 586,

590-591 (1978). Selvage’s personal act of shooting at Mr.

Ventura makes his participation and his reckless disregard for-

life greater than that of the Tison brothers in Tison v. Arizona,

481 U.S. 137, 139-141 (1987).

In addition to his accomplice status and his mens rea (ex-

treme recklessness rather than intent), defendant relies heavily

on his psychiatric testimony. Pet. Brief 21-27. If Selvage were

genuinely insane then his execution would be a fundamental

miscarriage of justice. “Mental illness short of that,” as de-

fendant characterizes it, Pet. Brief 108, is a very different

matter. The sentence would still be well within the discretion

of the sentencer, and a great many people would give little or

no mitigating weight to the psychiatric testimony. The trial

judge seems to have been particularly unimpressed with the

psychiatrist. See J. A. 33.

In summary, then, defendant's contention that he does not

deserve his sentence is merely an opinion with which a great

many reasonable people would disagree. Many of them would

consider any lesser sentence to be a miscarriage of justice. In

light of the expansion of the excluded classes in Thompson and

Penry, supra, and the very substantial support for capital pun-

ishment for all murders, see note 14, supra, this will be true

any time the crime and the defendant do not fall into any of

Ps

30

the excluded classes described above.!*

A habeas petitioner is “actually innocent” of a death sen-

tence if, considering all available facts, the sentence was not

legally within the discretion of the sentencer to impose. Sel-

vage’s sentence was well within the sentencer’s discretion, and

therefore it is not a fundamental miscarriage of justice.

Conclusion

The decision of the Court of Appeals for the Fifth Circuit

should be affirmed.

Dated: December, 1989

Respectfully submitted,

KENT S. SCHEIDEGGER

Attomey for Amicus Curiae

Criminal Justice Legal Foundation

15. At one time, counsel for amicus proposed a “no rational person could

differ” standard. Cf. Jackson v. Virginia, 443 U. S. 307, 313 (1979).

After observing the misapplication of the harmless error rule in capital

cases, see, e.g. , Bowen v. Kemp, 832 F. 2d 546, 553 (11th Cir. 1987) (Fay,

J., dissenting), id., at 554 (Edmondson, J., dissenting), it appears that

such an indefinite rule offers too much temptation for those inclined to

second-guess the sentencer.