The Different Rules for Prior Inconsistent Statements and Prior Consistent Statements

January 12, 2016
By Hanging Out with Carl Gunn


  • Prior inconsistent statements are always admissible to impeach a witness, so long as they’re in fact inconsistent.
  • Prior inconsistent statements are admissible for their truth only if given under oath at a trial, hearing, or other proceeding.
  • A prior statement may be “inconsistent” without being directly contradictory; the somewhat ambiguous test is whether a reasonable person could infer the two statements were produced by inconsistent beliefs.



The post last month on prior consistent statements (see “The New (Well, Sorta New) Rule on Prior Consistent Statements” in the December 2015 link at the right) brought to mind their opposite – prior inconsistent statements – and the different rules that govern them.  Prior inconsistent statements are treated differently in two respects – one that makes them more useable and one that makes them less useable.

The way in which prior inconsistent statements are more useable than prior consistent statements is that they are always admissible, so long as they’re inconsistent.  The Ninth Circuit has described it as a “basic rule of evidence” that “prior inconsistent statements may be used to impeach a witness.”  United States v. Monroe, 943 F.2d 1007, 1012 (9th Cir. 1991) (quoting United States v. McLaughlin, 663 F.2d 949, 952 (9th Cir. 1981) and United States v. Hale, 422 U.S. 171, 176 (1975)).  See also United States v. Bao, 189 F.3d 860, 865-66 (9th Cir. 1999) (same quote).  Further, “[a] prior inconsistent statement is admissible to raise the suggestion that if a witness makes inconsistent statements, then his entire testimony may not be credible; such an inference does not depend on whether either the prior statement or the subsequent in-court statement is true.”  Bao, 189 F.3d at 866 (emphasis added).  This contrasts with the rule governing prior consistent statements, which are admissible only when made prior to the motive to fabricate if offered to rebut a charge of fabrication or when they logically rehabilitate the witness in some other way.  (See last month’s post for more detail on this.)

A prior statement may be “inconsistent” without being directly contradictory, moreover.  A statement is “inconsistent” “if under any rational theory it might lead to any relevant conclusion different from any other relevant conclusion resulting from anything the witness said.”  Weinstein’s Federal Evidence § 613.04[1].  See also United States v. Morgan, 555 F.2d 238, 242 (9th Cir. 1977) (defining statement as “inconsistent” “whenever a reasonable man could infer on comparing the whole effect of the statements that they had been produced by inconsistent beliefs” (quoting earlier edition of Weinstein’s Federal Evidence)).  “Even the inability to answer a question may be inconsistent with a previous affirmative response to the same question.”  Weinstein’s Federal Evidence § 613.04[1].  Given the fuzziness of this test, trial judges “retain a high degree of flexibility in deciding the exact point at which a prior statement is sufficiently inconsistent with a witness’s trial testimony to permit its use in evidence.”  See, e.g., United States v. Tran, 568 F.3d 1156, 1162-63 (9th Cir. 2009) (not abuse of discretion to find “vague,” “reluctant,” and “evasive” in-court testimony inconsistent with prior statement); United States v. Adamson, 291 F.3d 606, 612 (9th Cir. 2002) (noting that “witnesses may be impeached ‘by their previous failure to state a fact in circumstances in which that fact naturally would have been asserted” (quoting Jenkins v. Anderson, 447 U.S. 231, 239 (1980))).

Prior inconsistent statements are less useable than prior consistent statements in another respect, however, and I think it’s one we often forget in the heat of battle (or trial).  While prior inconsistent statements are always admissible to impeach a witness’s credibility, they are admissible for the truth of the matter asserted only when “given under oath subject to penalty of perjury at a trial, hearing, or other proceeding.”  Fed. R. Evid. 801(d)(1)(A).  See, e.g., Morgan, 555 F.3d at 242 (grand jury testimony).  Exactly what constitutes a “proceeding” under this rule is subject to debate, compare United States v. Castro-Ayon, 537 F.2d 1055, 1058 (9th Cir. 1976) (applying rule to statement taken under oath by immigration agent) with United States v. Day, 789 F.2d 1217, 1222-23 (6th Cir. 1986) (questioning Castro-Ayon) and United States v. Livingston, 661 F.2d 239, 242-43 (D.C. Cir. 1981) (holding statement given to postal inspector under oath not admissible under Rule 801(d)(1)(A) and noting that “the Rule seems to contemplate situations in which an official verbatim record is routinely kept, whether stenographically or by electronic means, under legal authority” (quoting 4 D. Louisell and C. Mueller, Federal Evidence § 419, at 171 (1980))), but the requirement that the statement be given under oath subject to the penalty of perjury is a requirement which the type of prior statements most often used in criminal cases will usually fail to satisfy, see, e.g., United States v. Ragghianti, 560 F.2d 1376, 1380 (9th Cir. 1977) (statement to FBI agent not admissible as substantive evidence under Rule 801(d)(1)(A) because not given under oath subject to penalty of perjury).

In our role as practical trial lawyers, we may scoff at this distinction, on the theory that a lay jury is unlikely to appreciate the distinction.  The Ninth Circuit has acknowledged that the distinction is “subtle” and pointing it out to a jury “may be fruitless since conjecture takes over.” Ragghianti, 560 F.2d at 1381.  Still, the court has described it as “a crucial distinction,” id.,  and given it force in at least two ways.  First, the court has held, perhaps because of the fact that the distinction is subtle, “that proper implementation of the rule requires ‘an explicit admonition to the jury by the court at the time a prior inconsistent statement is admitted, and also an instruction at the close of trial, that the statement may be considered only as bearing on credibility.’”  Id. (quoting Bartley v. United States, 319 F.2d 717, 719 (D.C. Cir. 1963)).  Second, the fact that “the maximum legitimate effect of the impeaching testimony can never be more than the cancellation of the adverse answer,” Ragghianti, 560 F.2d at 1381 (quoting United States v. Cunningham, 446 F.2d 194, 192 (2d Cir. 1971)), gives rise to the prohibition on calling a witness for the sole purpose of impeaching the witness with a prior inconsistent statement, which was discussed in a prior post a year and a half ago (see “They Don’t Get to Impeach Something They Knew They Were Going to Get” in the June 2014 link at the right).  Then a third way to give force to the distinction between the use for impeachment and the use as substantive evidence is in arguing the sufficiency of the evidence and/or the weight of the evidence in a harmless error or plain error analysis.  The fact that the prior inconsistent statement has no substantive value means it can’t be considered in evaluating the evidence for these purposes.

So remember these distinctions.  They can be useful in appellate and post-trial litigation even if they’re not for practical arguments made to a jury.