APPEL, Justice.

William Burkhalter challenged an unfavorable modification of his father's revocable trust that occurred just prior to his father's death. William claimed his brother, Steven Burkhalter, unduly influenced their father and tortiously interfered with the trust. William also alleged his father lacked the necessary testamentary intent when he made the modification. At trial, the district court directed a verdict for Steven on the tortious interference claim, and the jury returned a verdict for Steven on the testamentary capacity and undue influence claims. William appealed, claiming the district court gave the jury an erroneous instruction on undue influence that improperly heightened his burden of proof and prejudicially repeated the causation element. Steven cross-appealed, claiming the district court should have directed a verdict in his favor on all claims. On direct appeal, the court of appeals held the challenged instruction prejudiced William and remanded the case for a new trial on the undue influence claim. The court of appeals denied Steven's cross-appeal.

On further review, we conclude the district court's instruction accurately reflected the law of undue influence and did not unduly emphasize the causation element of the undue influence claim. As a result, we vacate the decision of the court of appeals and reinstate the jury verdict and the district court's judgment thereon.

I. Factual and Procedural Background.

Louis D. Burkhalter Jr. created the Louis D. Burkhalter Jr. Revocable Trust on January 14, 1980. Under the original terms of the trust, William was the primary beneficiary following the deaths of Louis and his wife, Margaret. Upon William's death, the trust directed the trustee to distribute the remaining assets to Louis's heirs. The trust designated the United States Bank, n/k/a U.S. Bank, N.A., of Cedar Rapids as trustee.

In 1995, Louis amended the trust to name William's wife, Cynthia, and his son, Matthew, as beneficiaries. Under the amended terms, Cynthia and Matthew would continue as beneficiaries following the deaths of Louis, Margaret, and William. Upon the deaths of Cynthia and Matthew, the remaining assets would be distributed to Louis's heirs.

In 2003, Louis amended the trust to allow the trustee, at its discretion, to distribute trust assets for the support of Louis's sister, Patricia. In 2004, Louis placed his home into the trust. At the time, William was living in the home with his family. Louis directed that William and his family have complete control of the residence as long as their decisions were reasonable and approved by the trustee.

By early July 2007, Louis, who was then ninety-eight years old, experienced a decline in health. He was moved from an independent-living wing at an assisted living community, where he had lived for several years, to a wing that provided twenty-four-hour care. Steven became aware of the change and, on July 9, traveled from his home in California to Iowa to visit his father. According to Steven, Louis and he had a conversation about the trust on July 11. During the conversation, Steven claimed, Louis indicated his desire to modify the trust to make Steven and William equal beneficiaries and instructed Steven to call the trustee to arrange the modification.

A few hours after the conversation with Louis, Steven met with a trust officer of the bank, who in turn called Louis's attorney. The trust officer, a longtime advisor of Louis, met with Louis later that day without Steven being present. Louis's attorney met with Louis the following day, July 12, again without Steven being present. The trust was modified July 13 to divide the trust assets equally between William and Steven.1 Louis died six days later.

On January 25, 2008, William filed a petition claiming that the modification was the result of undue influence, that Louis lacked the testamentary capacity to make the change, and that Steven intentionally and improperly interfered with the distribution plan of the trust. Trial was held from October 31, 2011, to November 4. Following the close of evidence, Steven moved for a directed verdict on all claims. The district court directed a verdict on the tortious interference claim, but refused to direct a verdict on the undue influence and lack of testamentary capacity claims.

The parties then proceeded to haggle on jury instructions. With regard to undue influence, William generally modeled his requested jury instructions on the Iowa Civil Jury Instructions of the Iowa State Bar Association. The Iowa Civil Jury Instructions provide one instruction setting forth the elements of undue influence and another defining undue influence. Iowa Civil Jury Instruction 2700.4 sets forth the elements of undue influence:

Iowa State Bar Ass'n, Iowa Civil Jury Instructions 2700.4 (2011). Iowa Civil Jury Instruction 2700.5 provides a definition of "undue influence":

Id. 2700.5.

The district court's proposed instruction merged these two model instructions. The district court's proposed Instruction No. 2 stated:

As is apparent, subparagraphs 1, 2, 3, and 5 of Instruction No. 2 stated the substance of the elements of undue influence contained in Iowa Civil Jury Instruction 2700.4. Subparagraph 4, however, contained language related to the definition of undue influence contained in Iowa Civil Jury Instruction No. 2700.5.

William objected to the court's proposed Instruction No. 2, but the district court declined to modify the instruction. The jury returned a verdict in favor of Steven on the remaining claims. William filed a motion for a new trial, which was denied, and this appeal followed.

We transferred the case to the court of appeals. The court of appeals reversed the district court and remanded for a new trial after concluding the district court's addition of subparagraph 4 prejudiced William. According to the court of appeals, the addition of subparagraph 4 "added another step the plaintiff had to prove, thereby rendering the instruction faulty either through repetition, or by giving undue emphasis to otherwise correct statements of law." Finally, the court of appeals noted it found sufficient evidence to deny Steven's directed verdict and send the undue influence issue to the jury.

Steven sought further review, which we granted.

II. Standard of Review.

Challenges to jury instructions are reviewed for correction of errors at law. Koenig v. Koenig, 766 N.W.2d 635, 637 (Iowa 2009). Error in giving a jury instruction "does not merit reversal unless it results in prejudice." Wells v. Enter. Rent-A-Car Midwest, 690 N.W.2d 33, 36 (Iowa 2004). When the challenged instruction is "`conflicting and confusing,'" error is presumed prejudicial and reversal is required. Koenig, 766 N.W.2d at 637 (quoting Waits v. United Fire & Cas. Co., 572 N.W.2d 565, 575 (Iowa 1997)). Reversal is also required when the instruction contains a material misstatement of law. Waits, 572 N.W.2d at 575.

III. Discussion.

A. Introduction. As noted by one modern treatise, the notion of undue influence upon a testator is "a slippery concept at best." Eunice L. Ross & Thomas J. Reed, Will Contests § 7:1 (2d ed.), available at http://www.westlaw.com (updated June 2013) [hereinafter Ross & Reed]. The treatise further notes that "[t]he courts have never been consistent in their definitions of the term, nor in the judicial tests used to evaluate proof of undue influence." Id. § 7:1.

The difficulty with the notion of undue influence in the context of a will is illustrated by the question of the appropriate standard of proof required to prevail on a claim. On the one hand, undue influence seems like a relative of fraud, which generally requires a heightened standard of proof. See, e.g., Joseph Traub Arenson, The Doctrine of Undue Influence in Anglo American Law 4 (1953) [hereinafter Arenson] (noting that fraud and undue influence are similar in that someone has been unfairly enriched at the expense of another, but that "[u]ndue influence is a more restricted and more indefinite concept"). On the other hand, undue influence can occur without a material misrepresentation or omission, see In re Estate of Raedel, 152 Vt. 478, 568 A.2d 331, 335 (1989), which makes it analogous to ordinary civil causes of action at law where a preponderance-of-the-evidence standard prevails.

This case requires us to consider the proper elements of an undue influence claim. In particular, the question arises whether an element of an undue influence claim is that the result is "clearly" the effect of undue influence. This "clearly" requirement sounds like clear-and-convincing-evidence-light, at least as to causation.

These important legal questions arise against a backdrop of some judicial discomfort with the claim. The occasional but potentially very sharp unfairness that will result if undue influence claims are abandoned altogether seems like too much for courts to bear. See Arenson at 68 (declaring the evolution of undue influence doctrine a fortunate jurisprudential gain that extends social justice for "the aged and infirm, the weak, the dependent, the subservient and the inexperienced"). Yet, opportunistic strike litigation where the testator is not available to defend what may very well have been a choice of free will is not very attractive either. See John H. Langbein, Will Contests, 103 Yale L.J. 2039, 2045 (1994) [hereinafter Langbein] (suggesting the will contest involving the Johnson & Johnson fortune revealed "deeply deficient institutions and procedures" in the United States that "would have been suppressed in short order" in any other Western country). Further, fact finders may be tempted not to focus on the free will of the testator, but rather on their own values or senses of propriety. Melanie B. Leslie, The Myth of Testamentary Freedom, 38 Ariz. L.Rev. 235, 245 (1996) [hereinafter Leslie] (arguing that even though the contestant bears the burden of establishing undue influence, "determinations often are more dependent on courts' normative views of the relationships between the testator, beneficiary and contestant than the actual presence or absence of factors often deemed indicative of undue influence").

So, courts have struggled with the concept of undue influence. Today, it is our turn.

B. Positions of the Parties. William focuses his fire on Instruction No. 2's portrayal of the elements of undue influence in two ways. First, William argues the addition of subparagraph 4 coupled with the "clearly" language of subparagraph 5 heightened his burden of proof to something higher than a preponderance of the evidence, which rendered the instruction conflicting, confusing, and prejudicial. According to William, the instruction allowed the jury to find the required elements of undue influence described in subparagraphs 1 through 4, but then conclude he could not prevail because subparagraph 5 required him to show the modifications to the trust "were clearly the result" of the circumstances described in the previous subparagraphs. He also points to Instruction No. 2's introductory paragraph, which instructed the jury to apply a preponderance-of-the-evidence standard. Second, William asserts Instruction No. 2 was unduly repetitive. William notes that subparagraphs 4 and 5 both address causation and that, if he proved subparagraph 4, he should have prevailed on his claim. Accordingly, William requests a new trial.

Steven responds that Instruction No. 2 was a correct statement of law. He claims the requirement of subparagraph 5 that the action complained of must clearly be the result of undue influence is well supported by our caselaw. See, e.g., In re Estate of Dankbar, 430 N.W.2d 124, 128 (Iowa 1988); In re Estate of Davenport, 346 N.W.2d 530, 532 (Iowa 1984). He notes the rationale of the substantive standard is to prevent weak circumstantial evidence from supporting a finding of undue influence. See 1 Sheldon F. Kurtz, Kurtz on Iowa Estates § 4.45, at 175 (1981) [hereinafter Kurtz]. Steven also claims subparagraph 4 is a correct statement of law taken from the definition of undue influence contained in Iowa Civil Jury Instruction 2700.5 and supported by caselaw. See, e.g., In re Estate of Cory, 169 N.W.2d 837, 842 (Iowa 1969); In re Estate of Roberts, 258 Iowa 880, 888, 140 N.W.2d 725, 730 (1966). Under the circumstances, Steven sees no undue repetition of a claim or defense favorable to him. He argues it was not error to combine subparagraphs 4 and 5 because jury instructions must be read together and considered in their entirety. See Anderson v. Webster City Cmty. Sch. Dist., 620 N.W.2d 263, 265 (Iowa 2000); Leaf v. Goodyear Tire & Rubber Co., 590 N.W.2d 525, 536 (Iowa 1999).

Steven further claims nothing in Instruction No. 2 altered the standard of proof. Steven argues subparagraph 5 dealt solely with the preponderance-of-the-evidence standard and provided an accurate statement of law. He further notes, the preponderance-of-the-evidence standard was repeated in Instruction No. 2's first paragraph. Steven argues that because instructions must be read as a whole, if some part of Instruction No. 2 should not have been given, error was nonetheless cured because another part of the instruction properly advised the jury of the legal principles involved. See Moser v. Stallings, 387 N.W.2d 599, 605 (Iowa 1986). When the instructions are read as a whole, Steven argues, any error was not prejudicial.

Steven further responds by noting that there was nothing unduly prejudicial about subparagraphs 4 and 5. He notes that in order to be unduly prejudicial, an instruction must repeat a defense or claim favorable to one party in such a fashion as to be unfair. Steven asserts subparagraph 5 amounts to clarification of what must be shown to prevail on an undue influence claim.

In addition, Steven claims William's objection was not specific enough to alert the district court to the issues raised on appeal. Steven also cross-appeals, contending he was entitled to a directed verdict on the undue influence claim. Because of our disposition, we need not address these issues.

C. Iowa Caselaw on Undue Influence.

1. Standard of proof. The earliest Iowa undue influence cases embraced a preponderance-of-the-evidence standard. See Webber v. Sullivan, 58 Iowa 260, 267, 12 N.W. 319, 323 (1882) (approving undue influence instruction using a preponderance-of-the-evidence standard); McIntire v. McConn, 28 Iowa 480, 485-86 (1870) (noting the preponderance of evidence was against a finding of undue influence); see also Jamison v. Jamison, 113 Iowa 720, 723, 84 N.W. 705, 706 (1900) (rejecting "clear, satisfactory, and convincing" standard in law action against estate to enforce promises under will). In In re Townsend's Estate, 128 Iowa 621, 624, 105 N.W. 110, 110-11 (1905), however, the court applied a heightened standard of proof, holding that, at least where the testator was of sound mind and not an easily influenced individual, the required showing was one of clear and convincing evidence of undue influence.

After Townsend's Estate, however, we generally applied a preponderance-of-the-evidence standard for claims of undue influence. See, e.g., In re Behrend's Will, 233 Iowa 812, 817, 10 N.W.2d 651, 654-55 (1943); In re Busick's Will, 191 Iowa 524, 530, 182 N.W. 815, 818 (1921); Zinkula v. Zinkula, 171 Iowa 287, 299, 154 N.W. 158, 162 (1915); Ross v. Ross, 140 Iowa 51, 57, 117 N.W. 1105, 1107-08 (1908). Yet, there were at least two outliers. See Arndt v. Lapel, 214 Iowa 594, 606-07, 243 N.W. 605, 610 (1932) (applying a "clear, satisfactory, and convincing" standard to a claim that a grantee secured the transfer of a deed by undue influence); McNeer v. Beck, 205 Iowa 196, 198, 217 N.W. 825, 826 (1928) (same).

Finally, in In re Estate of Todd, 585 N.W.2d 273, 275-77 & n. 2 (Iowa 1998), we explicitly disavowed the heightened standard of proof for undue influence claims in will contests. We canvassed the law of other jurisdictions and recognized that some states applied a heightened standard in undue influence cases in probate, but noted the majority of states followed the preponderance standard. Id. at 275-76. We noted that challenges to testamentary dispositions on the grounds of undue influence in probate in Iowa were, by statute, triable as an action at law. Id. at 275. We cited the longstanding rule in Iowa that "`in law cases questions of fact are to be determined from the preponderance of the evidence.'" Id. at 276 (quoting Jamison, 113 Iowa at 723, 84 N.W. at 706). We reasoned that because will contests in Iowa were actions at law where the preponderance-of-the-evidence standard ordinarily applied, we would follow the majority rule. Id. We also expressly disavowed Townsend's Estate to the extent it stood for a different standard. Id. at 276 n. 2. We reaffirmed, however, that if a confidential relationship existed between the testator and the putative beneficiary, the burden shifted "to the recipient `to establish by clear and convincing proof that the advantage was procured without undue influence.'" Id. at 276 (quoting In re Estate of Herm, 284 N.W.2d 191, 200 (Iowa 1979)).

2. Elements of undue influence. While most Iowa cases reject a clear and convincing standard of proof generally for undue influence claims, Iowa law has employed a heightened substantive standard for finding undue influence. In In re Ankeny's Estate, 238 Iowa 754, 760, 28 N.W.2d 414, 417 (1947), we cited with favor a note in the Iowa Law Review as "set[ting] out clearly the rules which should govern" undue influence claims. The note stated:

Comments on Recent Cases, Wills — When Undue Influence Becomes a Jury Question, 30 Iowa L.Rev. 321, 322 (1945) (emphasis added) (quoting In re Leisch's Will, 221 Wis. 641, 267 N.W. 268, 271 (1936)). Thus, although most Iowa cases employed a preponderance-of-the-evidence standard as the burden of proof, the substantive requirements of the claim demanded proof that a person was "unquestionably" subject to influence and that the result clearly appeared to be the effect of undue influence.

The requirement that a testator must be "unquestionably" subject to undue influence was abandoned in Johnstone v. Johnstone, 190 N.W.2d 421, 426 (Iowa 1971), for a requirement that a testator be "susceptible to undue influence." (Internal quotation marks omitted.) The Johnstone court noted the former phrase had been imported from Leisch's Will, a 1936 Wisconsin Supreme Court case, but that the Wisconsin court did not repeat the qualification in later cases. Johnstone, 190 N.W.2d at 425. The Johnstone court reasoned the language, if literally applied, would make it almost impossible to prove an undue influence claim. Id.

While we have abandoned the requirement that a testator be "unquestionably" subject to undue influence, we have not departed from the requirement that the result in a will contest clearly show the effects of undue influence. Prior to Estate of Todd, our caselaw repeated the "clearly" requirement numerous times. See, e.g., In re Estate of Bayer, 574 N.W.2d 667, 671 (Iowa 1998); Estate of Dankbar, 430 N.W.2d at 128; Estate of Davenport, 346 N.W.2d at 532; Estate of Herm, 284 N.W.2d at 201; Frazier v. State Cent. Sav. Bank, 217 N.W.2d 238, 244 (Iowa 1974). In Estate of Todd, even though we rejected the higher standard of proof for will contests, we nonetheless continued to recognize as valid our caselaw that declared an element of undue influence was a result that was "clearly the effect of undue influence." 585 N.W.2d at 277 & n. 4. Following these cases, Iowa Civil Jury Instruction No. 2700.4 requires that the result be "clearly brought about by undue influence."

3. Summary of Iowa caselaw. In sum, while recognizing the split in caselaw across the country, we have held that in a will contest where no fiduciary relationship is involved, undue influence must be proved by a preponderance of the evidence. Id. at 275-77. Though we have rejected the clear, satisfactory, and convincing standard of proof, we have nonetheless held an element of undue influence is a "result clearly the effect of undue influence." Id. at 277 & n. 4; see also, e.g., Ankeny's Estate, 238 Iowa at 760, 28 N.W.2d at 417. We have not regarded rejection of a clear, satisfactory, and convincing standard of proof as inconsistent with this heightened "clearly" causation requirement in undue influence cases. See Estate of Todd, 585 N.W.2d at 277.

D. Caselaw on Undue Influence from Other States.

1. Standard of proof. As indicated in Estate of Todd, most states apply a preponderance-of-the-evidence standard to undue influence claims. See, e.g., Rose v. Dunn, 284 Ark. 42, 679 S.W.2d 180, 182 (1984); Lamborn v. Kirkpatrick, 97 Colo. 421, 50 P.2d 542, 544 (1935); In re Estate of West, 522 A.2d 1256, 1264 (Del.1987); Caranci v. Howard, 708 A.2d 1321, 1324 (R.I.1998); In re Estate of Duebendorfer, 721 N.W.2d 438, 446 (S.D.2006); In re Estate of Waters, 629 P.2d 470, 473 (Wyo. 1981); see also In re Estate of Cooch, 367 Mich. 445, 116 N.W.2d 740, 743 (1962) (requiring "evidence of probative force beyond mere suspicion"); Ross & Reed § 7:11 n. 3 (citing cases holding a will contestant must prove undue influence by a preponderance of the evidence). The preponderance-of-the-evidence standard is defended on the ground it is the same standard generally applied in civil actions. See In re Estate of Otto, 494 N.W.2d 169, 173 (N.D.1992) ("The governing standard for undue influence challenges, as with most civil matters, is a preponderance of the evidence."); see also Hack v. Janes, 878 So.2d 440, 444 (Fla.Dist.Ct.App.2004) ("Absent clear direction from the Legislature, we find no reason to deviate from that standard here.").

There is a contrary minority view. Some states heighten the standard and require contestants to prove their claim by clear and convincing evidence. See, e.g., Estate of Langley, 586 A.2d 1270, 1271 (Me.1991); In re Davis' Will, 14 N.J. 166, 101 A.2d 521, 522 (1953); Chapman v. Varela, 146 N.M. 680, 213 P.3d 1109, 1114 (2009); In re Estate of Hamm, 67 Wis.2d 279, 227 N.W.2d 34, 35 (1975); see also Ross & Reed § 7:11 n. 4 (citing cases holding a will contestant must prove undue influence by clear and convincing evidence). In In re Estate of Bennett, 19 Kan.App.2d 154, 865 P.2d 1062, 1068 (1993), the court explained the rationale underlying the application of a clear-and-convincing-evidence standard to an undue influence claim is that a will is presumed valid and should be overcome only upon a substantial showing. The court also characterized undue influence as a species of fraud, which ordinarily requires a higher standard of proof. Id. at 1069.

A somewhat different rationale for a heightened standard was expressed in Taylor v. Commissioner of Mental Health & Mental Retardation, 481 A.2d 139, 141 (Me.1984), a case involving the question of the appropriate standard for release of a defendant found not guilty by reason of insanity. The court emphasized the purpose of a standard of proof is to "`instruct the factfinder concerning the degree of confidence our society thinks he should have in the correctness of factual conclusions for a particular type of adjudication.'" Id. at 150 (quoting In re Winship, 397 U.S. 358, 370, 90 S.Ct. 1068, 1076, 25 L.Ed.2d 368, 379 (1970) (Harlan, J., concurring)). Thus, the standard of proof results "from a reasoned balancing of all the interests, public and private, that are implicated in the particular factual determination." Id.

Hawaii employs a mixed approach. Generally, the standard of proof in an undue influence case is a preponderance of the evidence. In re Estate of Herbert, 90 Haw. 443, 979 P.2d 39, 51 (1999). If, however, the undue influence is shown by circumstantial evidence, "the indirect evidence must be of a clear and convincing character." Id. at 53 (citation and internal quotation marks omitted).

Finally, in as many as thirteen jurisdictions, there appears to be no caselaw articulating the standard of proof in undue influence cases involving wills. See Ross & Reed § 7:11 n. 5. In some jurisdictions, courts have adopted language other than a preponderance of the evidence or clear and convincing evidence to describe the necessary standard of proof. For instance, one appellate court has stated the evidence must be "of sufficient character, substance, and weight to furnish a firm foundation for a jury's verdict." Fischer v. Heckerman, 772 S.W.2d 642, 646 (Ky.Ct.App.1989). What such a vague phrase means is unclear, but such phrases are designed to weed out cases appellate courts regard as simply too speculative to support relief.

2. Elements of undue influence. Many states have tried to identify elements or factors supporting a claim of undue influence to set aside a will, with variations from state to state. See Ross & Reed § 7:2. Some states have approached the problem with a flexible factor test reminiscent of the equitable origins of undue influence rather than a more structured approach involving mandatory legal elements. See, e.g., Crittell v. Bingo, 36 P.3d 634, 639 (Alaska 2001) (listing various factors to consider); Mullin v. Brown, 210 Ariz. 545, 115 P.3d 139, 144 (Ct.App.2005) (listing eight factors); Bye v. Mattingly, 975 S.W.2d 451, 457 (Ky.1998) (establishing seven "badges" of undue influence); Ruestman v. Ruestman, 111 S.W.3d 464, 479 (Mo.Ct.App.2003) (identifying factors to be taken into account); In re Estate of Eggebrecht, 291 Mont. 174, 967 P.2d 388, 391 (1998) (indicating court should take into account all surrounding circumstances).

In those states that establish elements of undue influence, a majority does not contain a requirement that the will clearly be the result of undue influence and instead simply require that the will was the result of undue influence. See, e.g., Dinan v. Marchand, 279 Conn. 558, 903 A.2d 201, 204 n. 1 (2006); Gmeiner v. Yacte, 100 Idaho 1, 592 P.2d 57, 63 (1979); In re Estate of Bacon, 169 Misc.2d 858, 645 N.Y.S.2d 1016, 1019 (1996); In re Will of Campbell, 155 N.C. App. 441, 573 S.E.2d 550, 560 (2002); In re Estate of Dion, 623 N.W.2d 720, 728 (N.D.2001); West v. Henry, 173 Ohio St. 498, 184 N.E.2d 200, 202 (1962); Rothermel v. Duncan, 369 S.W.2d 917, 922 (Tex.1963).

There are, however, other Midwestern states that adopt the same approach as Iowa. In a long line of cases, the South Dakota Supreme Court, citing established precedent, found undue influence in a will contest required "a result clearly showing the effects of undue influence." See, e.g., Stockwell v. Stockwell, 790 N.W.2d 52, 64 (S.D.2010); In re Estate of Pringle, 751 N.W.2d 277, 291 (S.D.2008); Estate of Duebendorfer, 721 N.W.2d at 446; In re Estate of Schnell, 683 N.W.2d 415, 421 (S.D.2004); In re Estate of Holan, 621 N.W.2d 588, 592 (S.D.2001); In re Estate of Dokken, 604 N.W.2d 487, 495 (S.D.2000); In re Estate of Elliott, 537 N.W.2d 660, 663 (S.D.1995); In re Estate of Smith, 520 N.W.2d 80, 85 n. 2 (S.D.1994); In re Estate of Till, 458 N.W.2d 521, 525 (S.D.1990); In re Estate of Zech, 285 N.W.2d 236, 240 (S.D.1979); In re Estate of Landeen, 264 N.W.2d 521, 523 (S.D.1978); In re Estate of Anders, 88 S.D. 631, 226 N.W.2d 170, 174 (1975); In re Estate of Metz, 78 S.D. 212,100 N.W.2d 393, 394 (1960).

The Supreme Court of Nebraska has taken a similar approach. See, e.g., In re Estate of Hedke, 278 Neb. 727, 775 N.W.2d 13, 28 (2009); In re Estate of Novak, 235 Neb. 939, 458 N.W.2d 221, 224 (1990); In re Estate of Peterson, 232 Neb. 105, 439 N.W.2d 516, 520 (1989); In re Estate of Villwok, 226 Neb. 693, 413 N.W.2d 921, 924 (1987); In re Estate of Price, 223 Neb. 12, 388 N.W.2d 72, 79 (1986); In re Estate of Gloe, 191 Neb. 395, 215 N.W.2d 98, 101 (1974); In re Estate of Gorthy, 169 Neb. 769, 100 N.W.2d 857, 864 (1960); In re Bainbridge's Estate, 151 Neb. 142, 36 N.W.2d 625, 626 (1949); In re Bowman's Estate, 143 Neb. 440, 9 N.W.2d 801, 806 (1943). Notably, both Nebraska and South Dakota, like Iowa, employ a preponderance-of-the-evidence standard to undue influence claims in will contests when no fiduciary relationship is present. See, e.g., Estate of Price, 388 N.W.2d at 76-77; Estate of Holan, 621 N.W.2d at 591.

Older Wisconsin cases also require a result clearly appearing to be the effect of the supposed influence. See, e.g., In re Estate of Komarr, 46 Wis.2d 230, 175 N.W.2d 473, 477-78 (1970); In re Will & Estate of Freitag, 9 Wis.2d 315, 101 N.W.2d 108, 109 (1960); In re Beyer's Estate, 262 Wis. 441, 55 N.W.2d 401, 403 (1952); In re Feeley's Estate, 253 Wis. 204, 33 N.W.2d 139, 142 (1948); In re King's Will, 251 Wis. 269, 29 N.W.2d 69, 72 (1947); In re Faulks' Will, 246 Wis. 319, 17 N.W.2d 423, 440 (1945); In re Scherrer's Estate, 242 Wis. 211, 7 N.W.2d 848, 853 (1943); In re Raasch's Will, 230 Wis. 548, 284 N.W. 571, 575 (1939); Leisch's Will, 267 N.W. at 271; In re Schaefer's Estate, 207 Wis. 404, 241 N.W. 382, 385 (1932). Later cases, however, seem to abandon the requirement, see, e.g., In re Estate of Dejmal, 95 Wis.2d 141, 289 N.W.2d 813, 819 (1980) (requiring "a result caused by, or the effect of such undue influence"), perhaps because Wisconsin has adopted a clear-and-convincing-evidence standard for undue influence claims, see, e.g., In re Estate of Sensenbrenner, 89 Wis.2d 677, 278 N.W.2d 887, 890 (1979) (requiring the presumption of a valid will to be overcome by clear, satisfactory, and convincing evidence of undue influence).

E. Secondary Authority on Undue Influence.

1. Burden of proof. The Restatement (Third) of Property provides that the burden of establishing undue influence rests on the party contesting the validity of the donative transfer. Restatement (Third) of Property: Wills and Other Donative Transfers § 8.3 cmt. b, at 144 (2003) [hereinafter Restatement (Third)]. The comment does not, however, address the appropriate standard of proof for that burden. Similarly, the Uniform Probate Code provides that contestants of a will have the burden of establishing undue influence, but does not indicate the appropriate standard. Unif. Probate Code § 3-407 (amended 2010), 8(II) U.L.A. 95 (2013). A leading treatise declares that "[i]t is usually said" that the burden of proof is on the party alleging the undue influence by a preponderance of the evidence. 3 William J. Bowe & Douglas H. Parker, Page on the Law of Wills § 29.79, at 696 (2004). Another treatise simply declares the burden of proving undue influence is on the contestant. William M. McGovern, Sheldon F. Kurtz, & David M. English, Principles of Wills, Trusts & Estates § 7.3, at 359 (2d ed.2012) [hereinafter Principles of Wills, Trusts & Estates].

2. Elements of undue influence. We begin our review of secondary authority of elements of undue influence by considering the Restatement (Third). Section 8.3 of the Restatement (Third) considers claims of undue influence, duress, or fraud. See Restatement (Third) § 8.3, at 143-44. Section 8.3 generally provides that a donative transfer can be found to have been procured by undue influence where it is shown that "such influence ... overcame the donor's free will and caused the donor to make a donative transfer that the donor would not otherwise have made." Id. § 8.3, at 143.

Comment e addresses cases where, as is often true, there is no direct evidence of undue influence and the claim must be proven through circumstantial evidence. See id. § 8.3 cmt. e, at 145. Comment e recites traditional elements of undue influence, including a causation element that "there was a result appearing to be the effect of the undue influence." Id. The term "clearly" does not appear in the comment.

However, there is authority applying an approach similar to that in Ankeny's Estate. For instance, the annotation appearing at 79 Am.Jur.2d Wills § 359, at 576-77 (2013) states the elements of undue influence include:

(Emphasis added.) Similarly, 95 C.J.S. Wills § 375, at 335 (2011) states the fourth element of undue influence is that "the result — reflected in the will — was clearly the effect of undue influence." (Emphasis added.) One treatise also declares a contestant must prove that "the will reflects a disposition clearly appearing to be the product of the undue influence." Principles of Wills, Trusts, and Estates § 7.3, at 354-55 (emphasis added).

3. The case for abandonment. While conventional legal authorities debate the nuances, critics of undue influence in wills have recently been energized. At least one has called for abandonment of the doctrine altogether. See generally Carla Spivack, Why The Testamentary Doctrine of Undue Influence Should Be Abolished, 58 U. Kan. L.Rev. 245 (2010) [hereinafter Spivack]. While no one in this litigation is calling for such a radical step, the underlying critique of the doctrine may shed some light on the less draconian choices presented in this case.

The case for abandonment notes, as a general matter, that courts have failed to provide any standard of clarity, fairness, or predictability that a legal doctrine should provide. E.g., id. at 246. According to one commentator, the undue influence doctrine "ostensibly safeguards testamentary freedom" yet determinations "often are more dependent on courts' normative views of the relationships between the testator, beneficiary and contestant than by the actual presence or absence of factors often deemed indicative of undue influence." Leslie, 38 Ariz. L.Rev. at 244-45. In short, because of its spongy character, it has been argued the law of undue influence may undermine testamentary freedom in order to promote social goals thought to be desirable. Id. at 236-37; see also Spivack at 248, 298.

F. Resolution of Issues in this Case.

1. Burden of proof to be employed. We first consider the substantive question of whether the court's instruction that causation must be clearly proved is consistent with Iowa law. Pursuant to Iowa Rule of Civil Procedure 1.924, the district court must "instruct the jury as to the law applicable to all material issues in the case." The instructions "`must convey the applicable law in such a way that the jury has a clear understanding of the issues it must decide.'" Boyle v. Alum-Line, Inc., 710 N.W.2d 741, 748-49 (Iowa 2006) (quoting Thompson v. City of Des Moines, 564 N.W.2d 839, 846 (Iowa 1997)). William challenges the "clearly" requirement in Instruction No. 2 as inconsistent with the established Iowa law that a preponderance of the evidence is the appropriate standard to prove undue influence.

Certainly William is correct in asserting that we have repeatedly held the standard of proof in undue influence cases is a preponderance of the evidence. See, e.g., Estate of Todd, 585 N.W.2d at 275-76. However, we have also repeatedly held the fourth element of an undue influence claim, causation, must be clearly proved. See, e.g., id. at 277 n. 4; Ankeny's Estate, 238 Iowa at 760, 28 N.W.2d at 417.

The question arises whether these two concepts can coexist. In Estate of Todd, we found no inconsistency in generally applying a preponderance-of-the-evidence standard to the elements of undue influence, but at the same time requiring that causation be clearly established. 585 N.W.2d at 277 & n. 4. The cases since Estate of Todd have applied our established undue influence law. The courts of Nebraska and South Dakota have adopted a similar approach. See, e.g., Estate of Price, 388 N.W.2d at 76-77; Estate of Holan, 621 N.W.2d at 591-92. There is at least some support for the approach in secondary commentary. In light of Estate of Todd, the authorities in South Dakota and Nebraska, and the supportive authorities, we conclude a requirement that causation be clearly established is not inconsistent with the preponderance-of-the-evidence standard generally applying to the other elements of undue influence.

The question remains, however, whether we should abandon the "clearly" requirement in causation. We decline to do so. A heightened causation element in undue influence cases makes sense. In cases involving challenges to wills based upon undue influence, the central issue is whether the acts of the testator were a product of free will or coercion. The testator, however, is not available to testify and, as a result, a speculative element is necessarily introduced into the claim. As colorfully noted in the commentary, will contests necessarily apply a "worst evidence" rule. Langbein, 103 Yale L.J. at 2044.

Further, it is not always easy to distinguish ordinary permissible influences on a testator from improper coercion. The injection of the word "clearly" into the fourth element of undue influence is designed to add a measure of protection to the free will of a testator, filter out claims that are unduly speculative, and to prevent the doctrine from expanding beyond its limited scope. All of the other elements of undue influence might be present — susceptibility, opportunity, and disposition — and, still, the will provisions might be the result of the testator's free will.

The heightened causation requirement of "clearly" ensures the other factors really mattered to the end result. As we noted in Estate of Davenport, weak circumstantial evidence that at most raises the possibility of influence is not sufficient. 346 N.W.2d at 532. We quoted at length from a leading Iowa probate treatise, which stated:

Id. (quoting 1 Kurtz § 4.45, at 175). The "clearly" requirement in our law helps prevent our undue influence doctrine from being over inclusive, as it will tend to deter juries from finding undue influence merely because they disagree with the terms of a will.

We recognize that many other jurisdictions reach this result by applying a clear and convincing evidentiary standard with respect to all elements of undue influence. We rejected that approach in Estate of Todd, 585 N.W.2d at 277. Neither party asks us to revisit the issue. In any event, we think it permissible to heighten the plaintiff's burden through the substance of an elements-based instruction. Our caselaw recognizes that will contests, by statute, are actions at law, which means jury trials are available and the applicable standard of review is a preponderance of the evidence. Yet, by judicially establishing a somewhat higher causation standard, we increase the likelihood that specious undue influence claims are weeded out.

We do not think the instruction is so confusing as to require reversal. The instruction required that in order to succeed, William had to prove his undue influence claim by a preponderance of the evidence. One of the elements — causation — requires clear proof. We think a reasonable jury would understand that it must find by a preponderance of the evidence that the changes were clearly the result of undue influence.

2. Repetition. We now turn to the issue of whether the instruction was unduly repetitive. Instructions may not give undue prominence to certain evidence involved in the case. E.g., State v. Marsh, 392 N.W.2d 132, 133 (Iowa 1986); see also 75A Am.Jur.2d Trial § 981, at 618-19 (2007). Further, instructions may not repeat even correct statements of law to the point of undue emphasis. E.g., Andrews v. Struble, 178 N.W.2d 391, 400 (Iowa 1970). In sum, the bottom line is that the instructions must not "give undue prominence to any particular aspect of a case," Vachon v. Broadlawns, 490 N.W.2d 820, 822 (Iowa 1992), which may include a "particular theory, defense, stipulation, burden of proof, or piece of evidence," Olson v. Prosoco, Inc., 522 N.W.2d 284, 287 (Iowa 1994). William asserts that by instructing the jury regarding the causation element of undue influence in both subparagraphs 4 and 5, the instruction was flawed as unduly emphasizing certain evidence in the case.

We have on a number of occasions found instructions that unduly emphasized certain evidence were flawed and required reversal. For example, in Clarke v. Hubbell, 249 Iowa 306, 315-16 86 N.W.2d 905, 910-11 (1957), we found that the district court, by repeating time and again the principle that a city was not an insurer for all injuries that occurred on its premises, committed reversible error. By its frequent repetition of a limiting concept in the law, the court gave undue emphasis to a potential defense in the case.

In Andrews, however, we found instructions were not repetitious where one instruction informed the jury that if it found the negligence of the defendant and other persons proximately caused the plaintiff's injury, the concurring negligence would not affect the plaintiff's right to recover solely from the defendant, and a second instruction informed the jury that if it found the defendant's negligence was the sole proximate cause of the injury, the plaintiff was entitled to recover. 178 N.W.2d at 400. We noted that while the latter statement may have been repetitious, the statement "was of some aid in clarifying the requirement of proximate cause." Id.

This case is more like Andrews than Clarke. Subparagraphs 4 and 5 did not contain the type of pointless repetition found in Clarke. Subparagraph 4 provided instructions related to the definition of undue influence not contained in subparagraph 5, and subparagraph 5 provided an additional legal requirement not contained in subparagraph 4 — namely, that the evidence must show that changes in the trust document were "clearly the result of the foregoing circumstances." While the instructions overlap to some degree, that is often the case in jury instructions that build upon concepts of law. Further, a single repetition coupled with a clarification of the law does not amount to error. Id.

IV. Conclusion.

For the above reasons, we vacate the decision of the court of appeals and affirm the judgment of the district court. Further, because of our resolution of this case, we need not address Steven's cross-appeal.

DECISION OF COURT OF APPEALS VACATED; DISTRICT COURT JUDGMENT AFFIRMED.