The best children’s games include a set of rules that are familiar, simple, harmonious with each other, and applicable all the time to anyone who plays. And so it is in the affairs of adults. In the criminal justice theater, our rules must have general applicability and must not contradict each other. While a child sees the injustice of another suddenly declaring Jacks “wild” in a game of cards, we adults may be slow to appreciate inherent unfairness in the uneven application of criminal justice rules. In other cases the unfairness may be too obvious to ignore yet protected by a hypocritical compartmentalization. This Article is the CLB’s effort to focus some attention on a long-festering hypocrisy of criminal justice appellate review.
The topic here is the comparison between the standards of review, in criminal appeals only, applied to assertions of insufficiency of the evidence (to support a conviction) and “harmless error” in the admission or exclusion of evidence.
Let’s see first how the Indiana Supreme Court describes the standard of review for sufficiency-of-evidence challenges to a criminal conviction:
“When reviewing a sufficiency of the evidence claim, we neither reweigh the evidence nor assess the credibility of witnesses. [citations omitted]. We only consider the evidence most favorable to the jury’s verdict, along with all reasonable inferences to be drawn therefrom, and will affirm a conviction if the probative evidence and reasonable inferences drawn from the evidence could have led the jury to find a defendant guilty beyond a reasonable doubt. [citations omitted].” Stephenson v. State, 742 N.E.2d 463, 496 (Ind. 2001).¹
Reasons to defer to such a degree to the finder of fact (whether judge or jury) include the finder’s opportunity to “observe the demeanor” of a witness so as to form an opinion of the relative verity of testimony. The central point of this deferential standard of review is that an appellate court should not weigh or reweigh evidence or judge witness credibility.
What about Federal courts? Here is a similar expression of deferential review in a criminal appeal to the Seventh Circuit:
“We review the evidence at trial in the light most favorable to the government and ‘will overturn a conviction based on insufficient evidence only if the record is devoid of evidence from which a reasonable jury could find guilt beyond a reasonable doubt.’ [citation omitted].” U.S. v. Hampton, 585 F.3d 1033, 1040 (7th Cir. 2009).
Now let’s compare the (deferential) standard of review for sufficiency appeals to the standard for assessing when trial court error is to be deemed “harmless.” Here is the Indiana state court standard for “harmless error” in admission or exclusion of evidence in a criminal trial:
“To determine whether error in the introduction of evidence warrants reversal, reviewing courts in Indiana are generally required to assess the probable impact of the evidence upon the jury, [emphasis added] [citation omitted], but various formulations have articulated the evaluation standard to be employed. [citation omitted] (whether there exists a ‘substantial likelihood that the questioned evidence contributed to the conviction’); [citation omitted] (whether erroneous evidence ‘is minor and unlikely to weigh appreciably against the defendant’); [citation omitted] (whether the resulting prejudice is ‘so slight that we can justly say that it did not affect the jury verdict’); [citation omitted] (‘whether the error itself had substantial influence’).”
“Despite variety in the language of our decisions, the undergirding concept remains consistent: an error will be found harmless if its probable impact on the jury, in light of all of the evidence in the case, is sufficiently minor so as not to affect the substantial rights of the parties.” Fleener v. State, 656 N.E.2d 1140, 1142 (Ind. 1995).
The central point of this protocol for review of error for harmlessness is that an appellate court is encouraged to weigh the evidence and judge credibility of the witnesses below when, in so doing, they can deprive a criminal appellant of a new trial free of the error which was appealed. That weighing of evidence which is forbidden for the purpose of aiding a criminal appellant will be utilized without second thought to transform otherwise reversible error into “harmless error,” or nothing of legal consequence.
What about the Federal standard for “harmless error” in the introduction of (criminal) trial evidence?
“An error is harmless if the untainted incriminating evidence is overwhelming.” U.S. v. Loughry, 650 F.3d 965, 975 (7th Cir. 2011).
Who decides that the untainted evidence is “overwhelming?” The answer is that appellate judges decide by weighing the evidence and judging credibility to aid the prosecution when they would never allow themselves such liberty to aid a criminal appellant.
Ask an appellate judge or justice whether he weighs the evidence or assesses witness credibility in criminal appeals challenging the sufficiency of the evidence below. The appellate judge or justice will most certainly reply with a “no.” Then ask whether he weighs the evidence or assesses witness credibility in determining whether (criminal) trial court error in the introduction of evidence is harmless. He will likely deny the inconsistency with a claim that weighing the evidence for “harmless error” evaluation is somehow different.
Here is a difference. By refusing to weigh evidence or assess witness credibility in criminal sufficiency appeals an appellate panel is deferring to the jury and its verdict. By weighing the evidence and assessing witness credibility in “harmless error” determinations (in the context of evidentiary error) the appellate panel is deferring to what it imagines the jury would have decided on evidence different from that upon which it made a finding of guilt. There should be no judicial imperative to honor on appeal one’s assessment of what a jury might have done had the case before them included more evidence from the defense or less evidence from the prosecution.
No one (or nearly no one) welcomes the prospect of countless criminal convictions being reversed and remanded for a new trial following a trivial error in the trial court pertaining to the receipt of evidence. But would that be the result of scrapping or revising the “harmless error” doctrine? A trial court’s evidentiary rulings are already protected by a deferential “abuse of discretion” standard of appellate review. Pavlovich v. State, 6 N.E.3d 969, 975 (Ind.Ct.App. 2014), trans. denied. Accordingly, the current hypocritical “harmless error” analysis is not applicable without the precedent of a trial court’s abuse of discretion, which would be a rarer circumstance than it is if only prosecutorial zeal were governed by prosecutorial caution and otherwise restrained by competent and impartial trial court judges. Allowing prosecutors and trial court judges to commit evidentiary error and then to escape the consequences of such through the current hypocritical “harmless error” doctrine only encourages the continued commission of error. With the removal of “harmless error” impunity prosecutors would learn to be more thoughtful and judges more careful in evidentiary rulings that might constitute an abuse of discretion. Consider also that evidentiary error in the trial court requires (except in the rare instance of fundamental error) a contemporaneous defense objection that is consistent with the error asserted on appeal.
The remedy to the described hypocrisy of weighing evidence in the COA and SCOTSI does not require a change in the standard of reviewing sufficiency of the evidence to sustain an errorless conviction. Relief from the hypocrisy does require first a recognition that two sets of rules are employed and second a scrapping or revision of the blatant inconsistency of the appellate weighing of evidence (from a cold, lifeless transcript) to benefit the State but never to benefit a defendant whose trial was tainted by an abuse of discretion.
¹The only known variation of this standard seems to be the (rarely applied) “incredible dubiosity” rule. Tillman v. State, 642 N.E.2d 221, 223 (Ind. 1994).