Appeal Research Paper

Appellate Structures

Like the federal government, nearly forty states have two-tier appellate systems. Although the precise jurisdictional arrangements vary, the most typical pattern provides for one appeal as of right to an intermediate appellate court and for further review in the state’s highest court primarily on a discretionary basis—though review as of right in the highest court (often directly from the trial level, thus bypassing the intermediate appellate court) is typically afforded from imposition of a death sentence. In two-tier systems, the state’s highest court ordinarily concentrates on unifying and elaborating the law, and the intermediate appellate court, though also important in elaborating legal principles, focuses on error correction (Shapiro, p. 632). In the remaining states, appeals are heard directly by the state’s highest court. The overwhelming majority of appellate courts hear civil and criminal appeals alike—a scheme thought preferable because a specialized criminal court ‘‘is unlikely to attract the continuing attention, interest, and concern of the entire bar’’ (American Bar Association, ‘‘Commentary on Standard,’’ chap. 21, 1.2).

Appellate courts typically decide in multijudge panels, thus permitting several judges to review matters decided by a single trial judge. Traditionally, appeals have been decided after oral argument by published written opinion, but docket pressures have led many jurisdictions, in cases deemed routine, to abbreviate or eliminate oral argument and to affirm convictions by order or by unpublished opinions—practices that have generated considerable controversy (Stern § 2.2).

Appeals by The Defense

Nature of The Right

In McKane v. Durston, 153 U.S. 684, 687–88 (1894), the Supreme Court stated that a defendant has no federal constitutional right to an appeal. But in neither McKane nor in subsequent decisions that have reiterated that statement (e.g., Ross v. Moffitt, 417 U.S. 600, 606 (1974), and Griffin v. Illinois, 351 U.S. 12, 18 (1956)) did a state fail to provide any appellate review of criminal convictions. Doubts that the McKane dictum remains sound (e.g., Arkin; Meltzer) are unlikely to be resolved, for every state now provides some method of appeal from criminal convictions in serious criminal cases. (In Virginia, West Virginia, and New Hampshire, formally no appeal as of right exists, but the procedures that each state’s highest court follows in determining whether to grant discretionary review ensure substantive consideration of the appellant’s contentions; Arkin, pp. 513–514.)

Except for common provisions requiring review when a death sentence is imposed, appeals in criminal cases are elective. Most jurisdictions require the trial court at sentencing to notify the defendant of the right to appeal (e.g., Rule 32(c)(5) of the Federal Rules of Criminal Procedure).

Equal Protection and Due Process

Whether or not the U.S. Constitution confers a right to appeal, once state law confers such a right, a state may not, consistent with the Fourteenth Amendment, deny indigent defendants the right to meaningful appellate review. The Supreme Court first applied that principle in Griffin, holding that indigent defendants are entitled to a free trial transcript so that they would have ‘‘as adequate appellate review as defendants who have money enough to buy transcripts’’ (p. 19). The Court has read Griffin as ‘‘a flat prohibition against pricing indigent defendants out of as effective an appeal as would be available to others able to pay their own way’’—Mayer v. Chicago, 404 U.S. 189, 196–197 (1971)—and thus has invalidated a rule conditioning the right to appeal on payment of a filing fee (Burns v. Ohio, 360 U.S. 252, 258 (1959)).

The Griffin principle was extended in Douglas v. California, 372 U.S. 353, 357–58 (1963), where the Supreme Court ruled that on a defendant’s first appeal, granted as a matter of right, an indigent defendant is entitled to counsel to brief and argue the appeal. The right recognized in Douglas comprehends assistance of counsel that satisfies constitutional standards of effectiveness (Evitts v. Lucey, 469 U.S. 387 (1985)). But in Ross v. Moffitt, the Court declined to extend Douglas to require counsel for indigents who seek discretionary review before the state’s highest court. Focusing less on equal treatment of rich and poor appellants and more on ensuring adequate access to appellate review, the Court reasoned that a ‘‘meaningful appeal’’ at the second tier was possible without counsel, for a lawyer would already have briefed and argued the first appeal, ensuring that the ‘‘defendant’s claims of error are organized and presented in a lawyerlike fashion’’ (pp. 612, 615).

Because indigent criminal appellants, unlike most civil appellants, typically have everything to gain and nothing to lose by seeking review, Douglas gave rise to the troublesome question of the appropriate role for a court-appointed counsel who believes an appeal utterly without merit. In Anders v. California, 386 U.S. 738, 744 (1967), the Court held that a lawyer who, after a ‘‘conscientious examination’’ of the case, finds an appeal to be ‘‘wholly frivolous’’ should so advise the court and request permission to withdraw. The opinion in Anders added, however, that because the court, not counsel, must decide whether the appeal is frivolous, the lawyer’s request must include ‘‘a brief referring to anything in the record that might arguably support the appeal’’ (p. 744), a copy of which must be furnished to the defendant, who then may raise additional points with the court. Some have criticized Anders for diverting limited resources from meritorious cases, though others have noted that appellate reversals sometimes occur even after appointed counsel has filed an ‘‘Anders brief ’’ explaining the hopelessness of the case (Hermann, p. 709), and some states have further limited counsel’s latitude by prohibiting withdrawal altogether (Warner, pp. 643–651). More recently, however, in Smith v. Robbins, 120 S. Ct. 726 (2000), the Supreme Court relaxed the strictures governing counsel, ruling that the Anders procedures are not the only way to satisfy the Constitution. In Smith, the Court approved a state procedure under which counsel’s filing on appeal did not identify any arguable issues, but merely (1) summarized the case history; (2) attested that counsel had reviewed the record, consulted with his client, and supplied the client with a copy of the brief; and (3) requested that the court examine the record for arguable issues.

The Final-Order Requirement

In general, appeal may be taken only from a final judgment, which typically means after conviction and imposition of sentence. The final judgment rule, though not unique to criminal cases, has been followed there with particular stringency because ‘‘the delays and disruptions attendant upon intermediate appeal,’’ which the rule is designed to avoid, ‘‘are especially inimical to the effective and fair administration of the criminal law’’ (DiBella v. United States, 369 U.S. 121, 126 (1962)).

Most jurisdictions do, however, permit appeal from some set of orders not strictly final. In Cohen v. Beneficial Industrial Loan Corp., 337 U.S. 541 (1949), a civil case, the Supreme Court concluded that a pretrial ruling should be deemed ‘‘final’’ for purposes of appeal in the federal system if (1) the lower court has fully decided the question; (2) the decision was not merely a step toward final disposition of the merits of the case but instead resolved a collateral issue; and (3) the decision involved an important right that would be lost, probably irreparably, if review had to await final judgment.

The Supreme Court has found the collateral order rubric of Cohen applicable in only three criminal cases. In Stack v. Boyle, 342 U.S. 1, 4 (1951), the Court ruled that a defendant may immediately appeal a pretrial order setting bail. (Whether under a variant of the collateral order doctrine or specific statutory authorization—as federal law now provides, see 18 U.S.C. § 3145— both defense and prosecution are typically authorized to appeal bail decisions.) Under the Cohen rationale, appeal has also been permitted from a trial court’s denial of a motion to dismiss an indictment when the defendant claimed to be immune from prosecution under the double jeopardy clause (Abney v. United States, 431 U.S. 651, 662 (1977)) or the speech or debate clause (Helstoski v. Meanor, 442 U.S. 500, 508 (1979)); in both cases, the Supreme Court reasoned that the right at issue would be undermined by the mere occurrence of the trial. But the Court has not applied the Cohen doctrine expansively, refusing to permit an immediate appeal from a pretrial order that disqualified the defendant’s counsel (Flanagan v. United States, 465 U.S. 259, 270 (1984)) or that denied a defense motion presenting the claim that the prosecution was vindictive (United States v. Hollywood Motor Car Co., 458 U.S. 263, 270 (1982)) or that the defendant had been denied a speedy trial (United States v. MacDonald, 435 U.S. 850, 853 (1978)).

Many states follow Cohen’s collateral order doctrine or some similar approach that permits immediate appeal of some orders not strictly final. While a few states authorize interlocutory review more broadly—for example, where ‘‘appeal would be in the interest of justice’’ (Utah Code Crim. Proc. § 77–18a–1)—typically judicial authorization is required and is sparingly provided.

Bail Pending Appeal

Following conviction a defendant no longer enjoys the presumption of innocence, and thus the criteria governing release pending appeal are generally stricter than those applied pending trial. Some jurisdictions deny bail pending appeal for a category of more serious offenses; others make bail pending appeal, unlike bail pending trial, a matter of discretion rather than of right; still others make bail available only when the defendant demonstrates that the appeal raises a substantial question.

Appeals by The Prosecution

Prosecution appeals typically require specific statutory authorization, and virtually every jurisdiction authorizes appeals from at least some orders. Some jurisdictions, including the United States, essentially permit government appeals from all decisions dismissing charges, whether before or after trial, except when further prosecution would be barred by the double jeopardy clause (e.g.,18 U.S.C. § 3731)—a bar that applies after a jury verdict of not guilty or after any other judgment deemed to constitute an acquittal.

Many jurisdictions also authorize interlocutory appeals by the prosecution from specified orders—most commonly, pretrial decisions to suppress evidence (e.g., 18 U.S.C. § 3731)—in part because if an erroneous decision to suppress leads to an acquittal, double jeopardy principles will preclude a government appeal. Because of the obvious concern about delay and disruption, interlocutory appeals are rarely permitted once trial has commenced.

Extraordinary Writs

Either the defense or the prosecution may seek an extraordinary writ (e.g., mandamus or prohibition) from an appellate court to review decisions not otherwise appealable. Traditionally, such writs were available only in narrow circumstances—where, for example, the lower court failed to perform a ministerial duty or lacked jurisdiction. During the latter part of the twentieth century, many courts have relaxed the standards governing such writs (e.g., to comprehend a gross abuse of discretion or a serious legal error of general significance). Nonetheless, many jurisdictions express greater reluctance to issue such writs in criminal than in civil matters (e.g., Will v. United States, 389 U.S. 90 (1967)).

Mootness and Related Doctrines

No appeal will lie when post-trial events—for example, the death of the convict—render the case moot. Most jurisdictions have now departed from the traditional view that an appeal is moot whenever the sentence has been fully satisfied— that is, when the defendant has paid any fine and served the full period of any imprisonment or probation. A limited departure from the traditional position, in cases in which a fine has been paid, treats an appeal as alive if state law permits remittance of the fine upon overturning of the conviction.

A far broader and more common departure from the traditional view permits defendants to appeal, even where a sentence has been fully served, in order to avoid harmful collateral consequences of criminal convictions (e.g., possible enhanced punishment under recidivism statutes or testimonial impeachment should the convict testify in the future) (Sibron v. New York, 392 U.S. 40, 50–58 (1968)). When collateral consequences are presumed to exist, as they are in many jurisdictions, this doctrine approaches in practice, if not in theory, the view taken by a few jurisdictions that quite apart from collateral consequences, a conviction is never moot because the ‘‘stigma of guilt’’ remains even after the sentence has been satisfied (e.g., Jackson v. People, 376 P.2d 991, 994 (Colo. 1962)). However, an appeal that challenges the legality not of the conviction but only of a sentence that has been fully served is likely to be deemed moot, unless collateral consequences from the harsher sentence can be demonstrated (North Carolina v. Rice, 404 U.S. 244, 248 (1971)).

Concurrent Sentence Doctrine

Where a defendant had been sentenced to equal concurrent sentences on different counts, some appellate courts, after upholding the conviction on one count, will not consider challenges to the remaining counts. In Benton v. Maryland, 395 U.S. 784 (1969), the Supreme Court concluded that this so-called concurrent sentence doctrine could not, in light of Sibron, be justified on mootness grounds but stated that it ‘‘may have some continuing validity as a rule of judicial convenience’’ (p. 791). Following Benton, all but two of the federal circuits have embraced the concurrent sentence doctrine as a discretionary matter of judicial administration; only a few state courts have followed suit.

Scope of Appellate Review

In General

Appellate review, based as it is upon the written record assembled at the trial level, is often deferential, particularly with regard to discretionary trial management decisions, issues heavily intertwined with fact and testimonial credibility, and the trier of fact’s determination of guilt. Review of the trial court’s elaboration of legal standards is generally de novo, and many jurisdictions also review de novo the application of constitutional or other legal standards to the facts (e.g., Ornelas v. United States, 517 U.S. 690, 699 (1996)).

Conviction by Guilty Plea

Appeals from a conviction by guilty plea are typically limited to claims that the trial court lacked jurisdiction, that the procedure for entry of the plea was defective, that the sentence was illegal, and, in some states, that the charge failed to state an offense. Other objections are generally deemed to have been waived by entry of the plea. While convictions by plea far outnumber convictions after trial, appeals from conviction at trial far outnumber appeals from conviction by pleas (Davies, p. 558).

The federal government (see Fed. R. Crim. Proc. 11(a)(2)) and nearly half the states permit a defendant (typically only with government consent and court approval) to enter a conditional guilty plea, which reserves the right to appeal on a specified issue. If the appeal prevails, the defendant is then permitted to withdraw the earlier plea and to plead anew.

Review of Sentence

In the United States, unlike many other countries, appellate review of a sentence imposed under traditional indeterminate systems rarely extended beyond ensuring that the sentence did not exceed the statutorily authorized punishment or was not influenced by factors that could not constitutionally be considered (e.g., Dorszynski v. United States, 418 U.S. 424, 431–32 (1974)). Appellate review of sentences has become far more common and somewhat more robust, however, in the last quarter of the twentieth century, for several reasons: (1) a number of states have extended review of sentencing decisions to embrace claims of clear abuse or clear mistake; (2) a significant minority of states have adopted determinate sentencing systems, which typically authorize appeals contending that the sentence violated applicable rules; and (3) review of death sentences is now routine to ensure compliance with the complex state and federal rules governing capital punishment.

Prosecution appeals of sentences have most commonly been authorized in determinate sentencing systems (e.g., 18 U.S.C. § 3742(b)), and in general do not violate the double jeopardy clause. However, the prosecution may not appeal the sentencer’s refusal to impose a death sentence—a decision ordinarily treated as an ‘‘acquittal’’ of capital punishment (Bullington v.Missouri, 451 U.S. 430, 446 (1981)).

Issues not Properly Raised

Every jurisdiction prescribes rules of pretrial and trial practice governing when and in what fashion particular objections (e.g., to the adequacy of the charge, to the admission of evidence, or to jury instructions) must be made. To induce compliance with those rules and promote orderly judicial administration, appellate courts ordinarily will not consider objections that were not properly presented at the trial level and consequently not ruled upon by the trial court.

Limited exceptions to this rule are typically recognized. First, a procedural requirement that itself violates due process cannot bar appellate review (Reece v. Georgia, 350 U.S. 85 (1955)). In addition, most states permit a defendant to challenge the trial court’s jurisdiction for the first time on appeal; states divide more evenly on whether they permit appeal of issues not raised below but based on newly announced legal decisions.

The most important and virtually universal exception authorizes appellate courts to consider ‘‘plain error,’’ whether or not properly raised below. To qualify as plain error under the federal doctrine, an error must not only be clear and obvious, but must also be shown by the defendant to be prejudicial, in the sense of likely affecting the outcome of the case (United States v. Olano, 507 US. 725 (1993)). In determining whether to reach an issue raised for the first time on appeal, many states will consider, in addition to factors similar to those that govern under federal law, whether the legal issue is of general significance, whether it is constitutional in nature, and whether its consideration would promote judicial economy or the public interest.

In the end, plain error has proven hard to define and has a somewhat discretionary character. In practice, it is more likely to be found when defense counsel’s representation was questionable or when the evidence of guilt is relatively weak.

Harmless Error on Appeal

In General

A determination on appeal that there was error at trial does not always require reversal. Rather, because minor errors are common and ‘‘[a] defendant is entitled to a fair trial but not a perfect one’’ (Lutwak v. United States, 344 U.S. 604, 619 (1953)), every jurisdiction follows some variant of the rule that harmless errors may be disregarded.

There is no agreement, however, as to just how the harmless-error standard should be formulated or how demanding it should be. On the first question, the most common approaches are (1) an ‘‘outcome-impact’’ approach, which focuses on whether the error influenced the jury in reaching its verdict; and (2) the ‘‘correct result’’ approach, which focuses on the force of the evidence against the defendant, error aside. The latter approach has often been criticized, on the grounds that it transforms the appellate court into a trier of fact and that even guilty defendants have a right to a fair trial, but it has not disappeared from the case law (Edwards, pp. 1192– 1194). The decisions do not, however, always clearly apply a well-defined approach, often adverting to a range of factors in reaching a conclusion without specifying clearly the nature of the harmless error analysis. In any event, the verbal formulas may matter less than the attitude brought to the harmless error inquiry by the appellate judges. No doubt, the stronger the appellate court’s belief that the defendant was guilty, the more likely it is that the error will be found harmless; indeed, concern has been expressed that appellate courts too readily find serious errors to be harmless when convinced of the defendant’s guilt (Edwards, pp. 1191–1192).

States also vary in how strong a showing of harmlessness must be made. In the federal system, reversal is required if the appellate court determines that the error ‘‘had substantial and injurious effect or influence in determining the jury’s verdict’’ (Kotteakos v. United States, 328 U.S. 750, 776 (1946)). In the states, standards range from demanding that the defendant prove it more probable than not that an error affected the outcome (People v. Lukity, 596 N.W.2d 607, 612 (Mich. 1999)) to requiring that the prosecution establish that an error was harmless ‘‘beyond a reasonable doubt’’ (Commonwealth v. Story, 383 A.2d 155, 162 (Pa. 1978)).

Errors Involving Constitutional Rights

When the error is one of federal constitutional law, the question whether the error was harmless is governed by a federal standard. In Chapman v. California, 386 U.S. 18, 24 (1967), the Court, rejecting the view that federal constitutional error is never harmless, ruled that such error required reversal unless the prosecution could demonstrate that it ‘‘was harmless beyond a reasonable doubt.’’ The basis for imposing this requirement on state courts is uncertain: it is unclear why a state should not be free to adopt a harmless error standard on appeal that is less favorable to the defendant than the Chapman standard, when the Supreme Court has insisted that the state could eliminate appeals by the defendant altogether (Meltzer, p. 12).

While some have viewed the Chapman test as too strict, contending that an appellate court can rarely find the requisite degree of certainty that the error had no effect (Traynor, pp. 43–44), others have contended that by its nature the test may not be strict enough, for constitutional error (such as the admission of impermissible evidence) may have significantly shaped trial strategies in ways not apparent to the appellate court (Saltzburg, p. 990).

The Supreme Court has stressed that constitutional errors are presumptively subject to harmless error analysis (Rose v. Clark, 478 U.S. 570, 579 (1986))—even, for example, admission of a coerced confession (Arizona v. Fulminante, 499 U.S. 279, 285 (1991)), or the failure to instruct the jury of the need to find an essential element of the offense (Neder v. United States, 119 S. Ct. 1827, 1831 (1999)). The Court has, however, recognized a limited class of fundamental constitutional errors—including total deprivation of the right of counsel, denial of the right to selfrepresentation, trial before a biased judge, racial discrimination in selection of the grand jury, denial of the right to a public trial, and improper instructions on proof beyond a reasonable doubt—that are ‘‘so intrinsically harmful’’ and that so infect the entire trial process that they defy harmless error analysis and require automatic reversal (Neder v. United States, p. 1833).

Effect of Reversal on Appeal

A retrial is ordinarily permitted after reversal of a conviction, except where retrial itself is the harm (as would usually be true when reversal was based on a claim of immunity, double jeopardy, or denial of a speedy trial) or where reversal was for insufficient evidence. Double jeopardy principles do not forbid imposition of a stiffer sentence after reconviction, but to protect a defendant’s freedom to appeal, the Supreme Court ruled in North Carolina v. Pearce, 395 U.S. 711, 725 (1969), that due process requires that ‘‘vindictiveness against a defendant for having successfully attacked his first conviction must play no part in the sentence he receives after a new trial,’’ and, indeed, that even an appearance of vindictiveness must be avoided. Consequently, in Pearce the Court held that a judge may impose a more severe sentence upon a defendant after a retrial only if the reasons for doing so are made part of the record and are based upon ‘‘objective information concerning identifiable conduct on the part of the defendant occurring after the time of the original sentencing proceeding’’ (p. 726).

The principle that Pearce announced has not been expansively applied. In Colten v. Kentucky, 407 U.S. 104 (1972), the Court held the principle inapplicable to sentences imposed after a defendant, convicted at trial, had exercised a statutory right to be tried de novo by a higher-level trial court. The Supreme Court stressed ‘‘that the court which conducted Colten’s trial and imposed the final sentence was not the court with whose work Colten was sufficiently dissatisfied to seek a different result on appeal; and it is not the court that is asked to do over what it thought it had already done correctly’’ (pp. 116–117). Similarly, in Chaffin v. Stynchcombe, 412 U.S. 17, 35 (1973), the Court found little potential for vindictiveness, and hence no constitutional defect, when the jury at the retrial, not knowing what the sentence had been at the original trial, imposed a stiffer sentence. And in Alabama v. Smith, 490 U.S. 794, 801–02 (1989), the Court declined to apply the Pearce presumption of vindictiveness when a judge had imposed a longer sentence after trial than he had after an earlier guilty plea that was later vacated. The Court ruled that because the factors that counsel leniency after a guilty plea were no longer present and because a trial gives the judge a fuller understanding of the circumstances than does a plea colloquy, there was no basis to presume that a stiffer sentence was motivated by vindictiveness.

Bibliography:

1. American Bar Association. Criminal Appeal. In Standards for Criminal Justice, 2d ed. Vol. 4, chap. 21. Boston: Little, Brown, 1980.
2. Arkin, Marc‘‘Rethinking the Constitutional Right to Appeal.’’ UCLA Law Review 39 (February 1992): 503–580.
3. Davies, Thomas ‘‘Affirmed: A Study of Criminal Appeals and Decision-Making Norms in a California Court of Appeal.’’ American Bar Foundation Research Journal no. 3 (1982): 543– 648.
4. Edwards, Harry ‘‘To Err Is Human, but Not Always Harmless: When Should Legal Error Be Tolerated?’’ New York University Law Review 70, no. 6 (1995): 1167–1213.
5. Flango, Carol, and Rottman, David B. Appellate Court Procedures. Williamsburg, Va.: National Center for State Courts, 1997.
6. Hermann, Robert. ‘‘Frivolous Criminal Appeals.’’ New York University Law Review 47, no. 4 (1972): 701.
7. LaFave, Wayne, and Israel, Jerold H. Criminal Procedure, vol. 3, chap. 26, pp. 171–281. St. Paul, Minn.: West Publishing, 1984, 1991 supp.
8. Meador, Daniel. Criminal Appeals: English Practices and American Reforms. Charlottesville: University Press of Virginia, 1973.
9. Meltzer, Daniel ‘‘Harmless Error and Constitutional Remedies.’’ University of Chicago Law Review 61, no. 1 (1994): 1–39.
10. Orfield, Lester Criminal Appeals in America. Boston: Little, Brown, 1939.
11. Rossman, David. ‘‘‘Were There No Appeal’: The History of Review in American Criminal Courts.’’ Journal of Criminal Law & Criminology 81, no. 3 (1990): 518–566.
12. Saltzburg, Stephen ‘‘The Harm of Harmless Error.’’ Virginia Law Review 59 (1973): 988.
13. Shapiro, Martin. ‘‘Appeal.’’ Law & Society Review 14, spring (1980): 629–661.
14. Stacy, Tom, and Dayton, KiM. ‘‘Rethinking Harmless Constitutional Error.’’ Columbia Law Review 88, no. 1 (1988): 79–143.
15. Stern, Robert Appellate Practice in the United States, 2d ed. Washington, D.C: Bureau of National Affairs, 1989.
16. Stith, Kate. ‘‘The Risk of Legal Error in Criminal Cases: Some Consequences of the Asymmetry in the Right to Appeal.’’ University of Chicago Law Review 57, no. 1 (1990): 1–61.
17. Traynor, Roger. The Riddle of Harmless Error. Columbus: Ohio State University Press, 1970.
18. Warner, Martha ‘‘Anders in the 50 States: Some Appellants’ Equal Protection Is More Equal than Others.’’ Florida State University Law Review 23, no. 3 (1996): 625–667.