Wednesday, June 13, 2012

Indiana Supreme Court on Circumstantial and Direct Evidence

During the past 100 years and more, many judges and courts have taken a turn at trying to explain the difference between direct and circumstantial evidence. The Supreme Court of Indiana recently decided to do so as well. See Hampton v. State, 961 N.E.2d 480 (Ind. 2012). The Indiana Supreme Court was moved to stumble into this thicket because Indiana case law requires (as the Indiana Supreme Court saw it) a trial judge to instruct the jury that if the evidence against the criminal defendant consists of some kind of circumstantial evidence, the jury must find, not only that the defendant is guilty beyond a reasonable doubt, but also that the evidence is "so conclusive and sure as to exclude every reasonable theory of innocence." Id. at 482. In the eyes of the Indiana Supreme Court, the need for such a jury instruction made it necessary for the Court to discuss the difference between direct and circumstantial evidence. It did so. After discussing the (supposed) difference, the Court concluded that the "reasonable theory of innocence" instruction is necessary when "where the conduct of the defendant constituting the commission of a charged offense is proven exclusively by circumstantial evidence." Id. (emphasis added). But the court's attempt to explain the distinction between direct and circumstantial evidence was bound to be as confusing and as murky as it was -- because the court got off on the wrong foot -- because there is in fact no difference between direct and circumstantial evidence.

The Indiana court's opinion not only added to the considerable body of judicial and academic murk about the supposed difference between direct and circumstantial evidence. The opinion also endorsed, in an unfortunately-retrograde way, that circumstantial evidence is weaker than direct evidence and that, for that reason, circumstantial evidence requires special handling -- such as by the giving of the "reasonable theory of innocence" instruction. In my view, consistency requires that the instruction, which probably does no discernible harm, should be given in every criminal trial where the issue is factual guilt or innocence.

Here, for what it is worth (which is not much), is an extract from the Indiana Supreme Court's opinion (but the extract does have a useful survey of authorities):

1. Instructing Juries to Use Caution when Evidence is Circumstantial

A. Need for a Special InstructionThe importance of a "reasonable theory of innocence" instruction is deeply imbedded in Indiana jurisprudence. {PT: string of Indiana citations omitted} "[T]his Court has never departed from the conviction that the ['reasonable theory of innocence'] standard is a proper one to be employed at the trial court and a defendant is entitled to an instruction to that effect." Spears, 272 Ind. at 638, 401 N.E.2d at 335.

We note that a number of more recent American appellate decisions appear to place less emphasis on the need for similar instructions. See Irene Merker Rosenberg & Yale L. Rosenberg, "Perhaps What Ye Say Is Based Only on Conjecture"—Circumstantial Evidence, Then and Now, 31 Hous. L. Rev. 1371, 1400-01 nn.121-22 (1995) (noting that at least twenty-nine states have eliminated a specific jury instruction on circumstantial evidence).2  This trend sprang after the United States Supreme Court's decision in Holland v. United States, 348 U.S. 121, 75 S. Ct. 127, 99 L. Ed. 150, 1954-2 C.B. 215 (1954), in which the Court rejected a claim of reversible error for the refusal "to instruct [the jury] that where the Government's evidence is circumstantial it must be such as to exclude every reasonable hypothesis other than that of guilt." Id. at 139, 75 S. Ct. at 137, 99 L. Ed. at 166. Without extensive explanation, the Court concluded that "the better rule is that where the jury is properly instructed on the standards for reasonable doubt, such an additional instruction on circumstantial evidence is confusing and incorrect." Id. at 139-40, 75 S. Ct. at 137, 99 L. Ed. at 166. We understand Holland to hold that including an additional "reasonable theory of innocence" instruction is not required but not that it is constitutionally erroneous.

[T]he Constitution neither prohibits trial courts from defining reasonable doubt nor requires them to do so as a matter of course. Cf. Hopt v. Utah, 120 U.S. 430, 440-41, 7 S. Ct. 614, 618-20, 30 L. Ed. 708 (1887). Indeed, so long as the court instructs the jury on the necessity that the defendant's guilt be proved beyond a reasonable doubt, see Jackson v. Virginia, 443 U.S. 307, 320, n.14, 99 S. Ct. 2781, 2789, n.14, 61 L. Ed. 2d 560 (1979), the Constitution does not require that any particular form of words be used in advising the jury of the government's burden of proof. Cf. Taylor v. Kentucky, 436 U.S. 478, 485-86, 98 S. Ct. 1930, 1934-35, 56 L. Ed. 2d 468 (1978). Rather, "taken as a whole, the instructions [must] correctly conve[y] the concept of reasonable doubt to the jury." Holland v. United States, 348 U.S. 121, 140, 75 S. Ct. 127, 137, 99 L. Ed. 150, 1954-2 C.B. 215 (1954).
Victor v. Nebraska, 511 U.S. 1, 5, 114 S. Ct. 1239, 1243, 127 L. Ed. 2d 583, 590 (1994) (alteration in original).

Subsequent to Holland, opinions from several states began to conclude that a "reasonable theory of innocence" instruction incorrectly suggests that circumstantial evidence is inherently less reliable.3 See, e.g., State v. Humpherys, 134 Idaho 657, 661-62, 8 P.3d 652 (Idaho 2000); People v. Bryant, 113 Ill. 2d 497, 499 N.E.2d 413, 420-21, 101 Ill. Dec. 825 (Ill. 1986); State v. Wilkins, 215 Kan. 145, 523 P.2d 728, 737 (Kan. 1974); State v. Lewisohn, 379 A.2d 1192, 1210 (Me. 1977); State v. Smith, 92 N.M. 533, 591 P.2d 664, 671-72 (N.M. 1979); State v. Jenks, 61 Ohio St. 3d 259, 574 N.E.2d 492, 502 (Ohio 1991), superseded by state constitutional amendment on other grounds as recognized in State v. Smith, 80 Ohio St. 3d 89, 1997 Ohio 355, 684 N.E.2d 668 (Ohio 1997); Easlick v. State, 2004 OK CR 21, 90 P.3d 556, 559 n.3 (Okla. 2004); State v. Roddy, 401 A.2d 23, 34-35 (R.I. 1979); State v. Gosby, 85 Wn.2d 758, 539 P.2d 680, 684-86 (Wash. 1975). Still other states reversed course by simply citing the Holland decision without any cogent analysis. See State v. Grippon, 327 S.C. 79, 489 S.E.2d 462, 464-67 (S.C. 1997) (Toal, J., concurring) (quoting Rosenberg & Rosenberg, supra, at 1402) ("[T]he widescale abandonment of the special circumstantial evidence charge has resulted  [486]  mostly from a 'bandwagon effect' following Holland, rather than from a reasoned rejection of the longstanding rule.").

In contrast, however, numerous jurisdictions uphold the requirement of a "reasonable theory of innocence" instruction. For example, New York's high court explained:
[T]he rule [requiring the instruction] draws attention to the fact that proof by circumstantial evidence may require careful reasoning by the trier of facts. By highlighting this aspect, the rule hopefully forecloses a danger legitimately associated with circumstantial evidence—that the trier of facts may leap logical gaps in the proof offered and draw unwarranted conclusions based on probabilities of low degree.
People v. Ford, 66 N.Y.2d 428, 488 N.E.2d 458, 465, 497 N.Y.S.2d 637 (N.Y. 1985) (citation omitted). Other jurisdictions continue to instruct that juries should consider whether circumstantial evidence excludes every reasonable hypothesis, explanation, or theory of innocence. See, e.g., People v. Bacon, 50 Cal. 4th 1082, 116 Cal. Rptr. 3d 723, 240 P.3d 204, 225-26 (Cal. 2010); Davis v. State, 285 Ga. 176, 674 S.E.2d 879, 880-84 (Ga. 2009); State v. Percy, 822 So. 2d 823, 828 (La. Ct. App. 2002); People v. DeWitt, 173 Mich. App. 261, 433 N.W.2d 325, 329 (Mich. Ct. App. 1988); McInnis v. State, 61 So. 3d 872, 875-76 (Miss. 2011); State v. Steele, 211 N.W.2d 855, 867 (N.D. 1973); State v. Teague, 680 S.W.2d 785, 790 (Tenn. 1984).

We believe that discarding the "reasonable theory of innocence" jury instruction is unwise. HN1 While a criminal conviction may properly rest entirely upon circumstantial evidence, there is a qualitative difference between direct and circumstantial evidence with respect to the degree of reliability and certainty they provide as proof of guilt. Such a supplemental instruction is a safeguard urging jurors to carefully examine the inferences they draw from the evidence presented, thereby helping to assure that the jury's reasoning is sound. Additionally, it serves to "reiterat[e] the magnitude of the ['proof beyond a reasonable doubt'] standard to juries when the evidence before them is purely circumstantial." Nichols, 591 N.E.2d at 136. In this regard, the "reasonable theory of innocence" instruction informs the jury that if a reasonable theory of innocence can be made of the circumstantial evidence, then there exists a reasonable doubt, and the defendant is entitled to the benefit of that doubt.

Such a "reasonable theory of innocence" instruction, when appropriate, is not satisfied by the instruction on reasonable doubt. The State argues that our statement in Nichols—that the "reasonable theory of innocence" instruction is a way of restating "proof beyond a reasonable doubt"—renders the instruction "duplicitous."4 Appellee's Br. at 12. To the contrary, providing the jury with an additional cautionary instruction in evaluating circumstantial evidence not only supports but further enhances the concept of requiring proof beyond a reasonable doubt. It admonishes the jury to tread lightly where the evidentiary gap between logical certainty and guilt is more tenuous. For these reasons, we find it altogether appropriate that juries receive, where appropriate, a "reasonable theory of innocence" instruction in addition to the standard reasonable doubt instruction.

B. Language of the Instruction
This special advisement has traditionally been accomplished in Indiana by an instruction like the one rejected by the trial court in the present case. Language within Indiana Pattern Jury Instruction 12.01 provides: "Where proof of guilt is by circumstantial evidence only, it must be so conclusive in character and point so surely and unerringly to the guilt of the accused as to exclude every reasonable theory of innocence." Ind. Pattern Jury Instruction—Criminal 12.01 (emphasis added). Effective juror comprehension and application, however, are compromised by impediments in the instruction's language—especially the uncertainties embedded in the phrases "proof of guilt" and "by circumstantial evidence only."

First, the instruction's language—that it applies where proof of "guilt" is "circumstantial only"—is potentially confusing. "Proof of guilt" could require that any evidence which supports an individual element of the charged offense be circumstantial, or it could require that all the evidence related to all elements of a charge be circumstantial. Compare Spears, 272 Ind. at 636-40, 401 N.E.2d at 334-35 (finding sufficient absence of direct evidence to warrant "reasonable theory of innocence" instruction despite direct evidence that the defendant and the victim were fighting earlier in the evening and that the victim was later found unconscious in the hallway, and that he died from skull fractures likely caused when his head hit the floor), with Davenport v. State, 749 N.E.2d 1144, 1149-50 (Ind. 2001) (affirming refusal of "reasonable theory of innocence" instruction where victim died of strangulation and had thirty-nine injuries and witness observed defendant beating on the victim several hours earlier). In most criminal prosecutions, there is some direct evidence that the charged crime was committed by someone. Thus, if the phrase "by circumstantial evidence only" is construed to not require jurors to consider whether there exist reasonable alternative theories of innocence whenever the occurrence of a criminal offense is shown in part by direct evidence, then the instruction would rarely, if ever, be used. On the other hand, the mens rea element for a criminal offense is almost inevitably, absent a defendant's confession or admission, a matter of circumstantial proof. Thus, requiring jurors to consider the possible existence of an alternate reasonable theory of innocence whenever proof of the mens rea element is circumstantial would lead to use of the instruction in most all criminal cases. See Spears, 272 Ind. at 639-40, 401 N.E.2d at 335 (requiring the "reasonable theory of innocence" instruction when circumstantial evidence is the exclusive proof of the actus reus5 but not when used to  [488]  prove the mens rea: "To hold otherwise would require a circumstantial evidence instruction in every case involving a crime containing the element of intent. Unnecessary confusion would result from such a course.").

From our review of jurisdictions employing an instruction similar to our "reasonable theory of innocence" instruction, we find no consistent approach to resolving these issues. Mississippi requires an instruction requiring the "exclusion of all reasonable hypotheses consistent with innocence" whenever there is any "direct evidence" such as "an admission or confession by the defendant to a significant element of the offense, or eyewitness testimony to the gravamen of the offense charged." McInnis, 61 So. 3d at 876 (emphasis omitted) (quoting Kirkwood v. State, 52 So. 3d 1184, 1187 (Miss. 2011)) (internal quotation marks omitted). In Michigan, juries are instructed that "[i]f the direct and circumstantial evidence, taken together, is open to two reasonable constructions, one indicating guilt and the other innocence, it is your duty to accept the construction indicating innocence." DeWitt, 433 N.W.2d at 329 (emphasis added). But Tennessee calls for the instruction only when "the evidence is entirely circumstantial."6 State v. Knight, 969 S.W.2d 939, 941-42 (Tenn. Crim. App. 1997) (emphasis added); see also Teague, 680 S.W.2d at 790.

Notwithstanding the strong support in Indiana case law for the concept of a "reasonable theory of innocence" instruction, in actual fact, the refusal of such instruction has rarely been found to be error, often based on the presence of evidence liberally deemed to be "direct" rather than "circumstantial." See, e.g., Davenport, 749 N.E.2d at 1149-50 (finding direct evidence where son observed his mother being beaten by the defendant, sought to intervene but was struck by defendant and went to bed, and the next day the mother was found murdered by strangulation, which was not witnessed by the son); Clemens v. State, 610 N.E.2d 236, 243-44 (Ind. 1993) (finding direct evidence from the defendant's admission that he was present when the victim sustained his mortal injuries and the pathologist testified that those injuries must have been inflicted intentionally); Chapman v. State, 556 N.E.2d 927, 931 (Ind. 1990) (finding direct evidence from testimony of bank teller who identified defendant as person standing outside bank before the robbery, but who could not see the facial features to identify the hooded robber who perpetrated the robbery). While this Court in Spears, which reversed a murder conviction, evaluated the necessity of giving a "reasonable theory of innocence" instruction by considering whether there was any direct evidence showing the actus reus of a charged crime, rather than just direct evidence of any element of the crime, 272 Ind. at 636-40, 401 N.E.2d at 334-35, this distinction has not yet been applied in any of our ensuing cases. Other than Spears, we find no case from this Court since 1928 that reversed a conviction due to a trial court's refusal to give a "reasonable theory of innocence" instruction. Landess v. State, 200 Ind. 440, 164 N.E. 267 (1928).7
These issues are not clarified by the present pattern instruction, which fails to clearly inform jurors whether the requirement of heightened scrutiny of circumstantial evidence applies only when there is a complete absence of direct evidence on every element of an offense, or when there is an absence of direct evidence with respect to a significant element or crucial component of guilt, or when there is an absence of direct evidence proving any single element of the charged statutory offense, or otherwise.

Second, the instruction unnecessarily calls upon the jury to determine whether evidence of guilt is "circumstantial." Such an evaluation is already the province of the trial judge in deciding whether such instruction is required in light of the nature of the evidence presented. Distinguishing between direct and circumstantial evidence as proof of a particular fact is a legal determination appropriate for judicial evaluation. It may require intricate legal analysis.

Direct evidence is "[e]vidence that is based on personal knowledge or observation and that, if true, proves a fact without inference or presumption." Black's Law Dictionary 636 (9th ed. 2009). Conversely, circumstantial evidence is "[e]vidence based on inference and not on personal knowledge or observation." Id. Indiana case law has expressed it thusly: "Direct evidence means evidence that directly proves a fact, without an inference, and which in itself, if true, conclusively establishes that fact. Circumstantial evidence means evidence that proves a fact from which an inference of the existence of another fact may be drawn." Gambill, 675 N.E.2d at 675. And to explain the meaning of "inference," the jury in Gambill was further instructed: "An inference is a deduction of fact that may logically and reasonably be drawn from another fact or group of facts." Id. But every evidentiary assertion at trial is "direct evidence" of something. That is, it is a fact offered as true and if accepted by the jury as such, conclusively establishes that fact.

In the grouping of circumstantial evidence, difficulty has arisen from not keeping in mind that most circumstantial evidentiary facts must ultimately in turn become themselves a proposition and be proved by "direct" evidence and also from confining the latter term to assertions of some main fact in issue. For example, the finding of a bloody knife upon the accused after a secret killing is a circumstance from which an important inference may be drawn; yet this fact of the finding must be proved by some person's assertion.
IA John Henry Wigmore, Evidence § 25, at 955 (Tillers rev. 1983) (footnote omitted). Thus, as another example, footprints  [490]  or fingerprints that place an accused at the scene of a crime may be direct evidence of the accused's presence at some point in time but only circumstantial proof that the accused committed the charged offense.8

C. Reformulating the Use and Language of the Instruction

To preserve our historic recognition that juries in criminal cases should be reminded to use particular caution when considering whether to find guilt based solely on crucial circumstantial evidence, we conclude that a special instruction is appropriate, but we reformulate the manner of use and language of the instruction. First, we find it inappropriate to include language burdening the jury with the task of deciding whether to apply the reasonable theory of innocence standard. Whether an instruction is supported by the evidence is a matter for the trial court to determine, and it need not be reevaluated by the jury. Second, because Indiana jurisprudence recognizes the importance of such an instruction in certain cases involving circumstantial evidence but our case law reveals a reluctance to find reversible error for failure to give the instruction if there is substantial direct evidence of guilt, we elect to apply the approach taken in Spears and direct that the "reasonable theory of innocence" instruction is appropriate only where the trial court finds that the evidence showing that the conduct of the defendant constituting the commission of a charged offense, the actus reus, is proven exclusively by circumstantial evidence. As discussed above, to deny the availability of a "reasonable theory of innocence" instruction whenever there is any direct evidence of the fact that a criminal offense has occurred, however, could render the instruction unlikely ever to be used, but requiring the instruction whenever there is no direct evidence of any single element would compel its use in almost all criminal  cases because mens rea is often shown only by circumstantial evidence.

We thus hold that, when the trial court determines that the defendant's conduct required for the commission of a charged offense, the actus reus, is established exclusively by circumstantial evidence, the jury should be instructed as follows: In determining whether the guilt of the accused is proven beyond a reasonable doubt, you should require that the proof be so conclusive and sure as to exclude every reasonable theory of innocence.


The underlying flaw in the Indiana Supreme Court's attempted solution is that it did not see that the evidence in every criminal case (and, indeed, in every type of case) is "exclusively circumstantial" because in no instance are any of the facts at issue in judicial proceedings established by "direct evidence," i.e., without the intervention of inferences.

  • It is true that testimonial evidence has characteristics that distinguish it from from non-testimonial evidence. But testimonial evidence, like non-testimonial evidence, requires the use of inferences if testimonial evidence is to ever do its job, which is to serve as proof (to some degree) of some fact.


The dynamic evidence page
Evidence marshaling software MarshalPlan
Post a Comment