In re Dependency Of A.C.

Opinion

NOTICE: SLIP OPINION

(not the court’s final written decision)

The opinion that begins on the next page is a slip opinion. Slip opinions are the written opinions that are originally filed by the court. A slip opinion is not necessarily the court’s final written decision. Slip opinions can be changed by subsequent court orders. For example, a court may issue an order making substantive changes to a slip opinion or publishing for precedential purposes a previously “unpublished” opinion. Additionally, nonsubstantive edits (for style, grammar, citation, format, punctuation, etc.) are made before the opinions that have precedential value are published in the official reports of court decisions: the Washington Reports 2d and the Washington Appellate Reports. An opinion in the official reports replaces the slip opinion as the official opinion of the court. The slip opinion that begins on the next page is for a published opinion, and it has since been revised for publication in the printed official reports. The official text of the court’s opinion is found in the advance sheets and the bound volumes of the official reports. Also, an electronic version (intended to mirror the language found in the official reports) of the revised opinion can be found, free of charge, at this website: https://www.lexisnexis.com/clients/wareports. For more information about precedential (published) opinions, nonprecedential (unpublished) opinions, slip opinions, and the official reports, see https://www.courts.wa.gov/opinions and the information that is linked there.

NOTICE: SLIP OPINION

(not the court’s final written decision)

The opinion that begins on the next page is a slip opinion. Slip opinions are the written opinions that are originally filed by the court. A slip opinion is not necessarily the court’s final written decision. Slip opinions can be changed by subsequent court orders. For example, a court may issue an order making substantive changes to a slip opinion or publishing for precedential purposes a previously “unpublished” opinion. Additionally, nonsubstantive edits (for style, grammar, citation, format, punctuation, etc.) are made before the opinions that have precedential value are published in the official reports of court decisions: the Washington Reports 2d and the Washington Appellate Reports. An opinion in the official reports replaces the slip opinion as the official opinion of the court. The slip opinion that begins on the next page is for a published opinion, and it has since been revised for publication in the printed official reports. The official text of the court’s opinion is found in the advance sheets and the bound volumes of the official reports. Also, an electronic version (intended to mirror the language found in the official reports) of the revised opinion can be found, free of charge, at this website: https://www.lexisnexis.com/clients/wareports. For more information about precedential (published) opinions, nonprecedential (unpublished) opinions, slip opinions, and the official reports, see https://www.courts.wa.gov/opinions and the information that is linked there.

IN THE SUPREME COURT OF THE STATE OF WASHINGTON

In the Matter of the Dependency of No. 100966-6

A.C., ORDER AMENDING OPINION

a minor child.

It is hereby ordered that the majority opinion of González, C.J., filed March 9, 2023, in the above entitled case is amended as indicated below. All references are to the slip opinion.

On page 1, line 4, after “judge, at a” delete “shelter care” and insert “dependency fact-

finding”.

On page 3, line 4, after “The court later” delete “found AC dependent” and insert “maintained out-of-home placement”.

On page 3, line 12, after “The court held” delete “another dependency” and insert “a dependency fact-finding”.

On page 6, line 10, after “dependent at a” insert “dependency fact-finding”.

On page 7, line 17, after “dependency” insert “fact-finding”.

In re Dependency of A.C., No. 100966-6 (order amending opinion)

On page 12, line 4, after “dependency” insert “fact-finding”.

DATED this 18th day of April, 2023.

___________________________________

Chief Justice

APPROVED:

______________________________ ______________________________

______________________________ ______________________________

______________________________ ______________________________

______________________________ ______________________________

FILE THIS OPINION WAS FILED FOR RECORD AT 8 A.M. ON

MARCH 9, 2023

IN CLERK’S OFFICE SUPREME COURT, STATE OF WASHINGTON MARCH 9, 2023 ERIN L. LENNON

SUPREME COURT CLERK

IN THE SUPREME COURT OF THE STATE OF WASHINGTON

)

In the Matter of the Dependency of ) No. 100966-6 )

A.C., ) En Banc )

a minor child. ) Filed: March 9, 2023 _______________________________)

GONZÁLEZ, C.J.— The State has the sobering emergency power to take a child away from their parents for the child’s own protection. Our statutes and constitutions constrain that power. Among those constraints is the State’s obligation to promptly prove to a judge, at a shelter care hearing conducted under the rules of evidence, that its exercise of power was justified. At that hearing, the child’s parents have the right to challenge the State’s exercise of power and the State’s evidence.

One type of evidence concerns us today: hearsay. “Hearsay” is an out-of-

court statement offered to prove the truth of the thing said. Hearsay is extraordinarily difficult to challenge because the person who made the statement is not in court, not under oath, and not subject to cross-examination. For those

In re Dependency of A.C., No. 100966-6 reasons, our rules of evidence generally do not allow hearsay to be offered for its truth.

But hearsay is often allowed for other limited purposes. For example, experts qualified to offer their opinions in court can explain how they reached their opinions. If using hearsay is acceptable in that expert’s field, the expert may rely on and testify about that hearsay to explain how they reached their opinions. Such hearsay must be used only for the reason it is offered: to explain the expert’s opinion, not—as happened here—as a shortcut to getting untestable evidence before the judge.

Here, the State concedes the trial judge erred by relying on the enormous amount of hearsay evidence offered by the State for its truth. The State, however, contends that error was harmless. Concluding otherwise, we reverse.

BACKGROUND

CC 1 and VC were driving through eastern Washington when CC went into premature labor. CC gave birth to AC in a nearby hospital. AC’s umbilical cord tested positive for cannabis. VC is AC’s father. Hospital staff noted that CC was disabled, that CC and VC were homeless, and that they had no baby supplies. The hospital reported its concerns to the State, and the State sent social worker

1 CC is referred to by the initials CP in the proceedings below.

3

In re Dependency of A.C., No. 100966-6 Michelle Woodward to investigate. Woodward contacted CC’s family from whom she heard reports of the couple’s domestic violence, criminal history, and drug use.

The State took custody of AC and temporarily placed him with a foster family. The court later found AC dependent at a contested shelter care hearing and ordered CC to participate in random drug testing and an evidence-based parenting program. The court also ordered the State to provide regular, supervised visitation. At about this time, a new social worker, Diana Barnes, was assigned to AC.

Over the next few months, CC participated in court-ordered services and supervised visitations. The court ordered no services for VC, but he joined CC on her visits. It appears there was often conflict between VC and hospital staff, visitation staff, and law enforcement during this time.

The court held another dependency hearing in January 2021 where Woodward, Barnes, and parenting therapist Logan Wright testified in support of AC’s dependency. Woodward and Barnes relied extensively on hearsay based largely on secondhand reports and statements rather than their own personal interactions or investigations. None of these reports were submitted into evidence, no records custodian authenticated them, and none of the out-of-court witnesses whose statements were recorded in those reports were called to testify. Barnes and Woodward also relied extensively on their conversations with various hospital

4

In re Dependency of A.C., No. 100966-6 staff, nurses, police, therapists, and other members of CC’s family who raised concerns about the couple. These various people were also not called to testify. Counsel for VC made two unsuccessful objections to the hearsay presented through the social workers. VC entered a standing hearsay objection.

The court also qualified Woodward as an expert on potential domestic violence. Woodward testified that in her expert opinion, CC and VC presented signs of domestic violence. Barnes was never explicitly qualified as an expert, but she was treated as such by the court. She was also allowed to give her expert opinion and relay the hearsay she relied on to form that opinion.

The court found AC dependent on several grounds, many of which relied heavily on the improperly admitted hearsay testimony.

The court found that VC had “ongoing mental health issues and aggression and/or violence.” Clerk’s Papers (CP) at 148. The court based this finding primarily on hearsay. The social workers testified that they heard VC was aggressive, refused to follow policies, and had been repeatedly arrested for disorderly behavior. From these encounters, the court concluded that “at every turn, there is an inability of the father to follow the basic policy/structure of what is required.” Id.

5

In re Dependency of A.C., No. 100966-6 The court also found that the parents’ past history with the criminal justice system and Child Protective Services supported dependency. This finding was also substantially based on hearsay. The hearsay referenced included CC’s past involvement with child protective services as told to Woodward by CC’s family. The court also specifically referenced a hearsay account of CC banging her head against the wall in frustration. The court also relied on nonhearsay, including VC’s criminal record.

VC and CC disputed or denied nearly all of the hearsay allegations under oath at the hearing.

The Court of Appeals affirmed the dependency. In re Dependency of A.C., No. 37999-0-III, slip op. at 2 (Wash. Ct. App. Apr. 28, 2022) (unpublished) https://www.courts.wa.gov/opinions/pdf/379990_unp.pdf. That court held that the trial court erred by relying on hearsay evidence but held the error harmless because the remaining evidence sufficiently supported dependency. Id. at 22, 23, 27. The parents successfully sought our review. Ruling Granting Rev., In re Dependency of A.C., No. 100966-6, at 11 (Wash. July 20, 2022).

ANALYSIS

Our rules of evidence generally bar courts from relying on hearsay for its truth. ER 801(c), 802. All parties agree that such reliance is error and that the

6

In re Dependency of A.C., No. 100966-6 lower court committed that error by admitting hearsay evidence offered by the state for the truth of the matter asserted. But they disagree on whether that error was harmless and on the proper analytical approach used to determine harmlessness. We determine how to evaluate harmlessness de novo. See Erwin v. Cotter Health Ctrs., Inc., 161 Wn.2d 676, 687, 167 P.3d 1112 (2007) (citing Tapper v. Emp’t Sec. Dep’t, 122 Wn.2d 397, 402-03, 858 P.2d 494 (1993)).

For context, parents have a fundamental constitutional right to care for their child, but the State has the power to intervene to protect that child when necessary. See In re Dependency of Schermer, 161 Wn.2d 927, 941, 169 P.3d 452 (2007). A judge may declare a child dependent at a hearing and have that child taken into the State’s custody if there is “no parent . . . capable of adequately caring for the child, such that the child is in . . . danger of substantial damage to [their] psychological or physical development.” RCW 13.34.030(6)(c); Schermer, 161 Wn.2d. at 942;

Our rules of evidence apply at these hearings. RCW 13.34.110(1). Under those rules, parents “should not be deprived of their parental rights on hearsay.” In re Welfare of Ross, 45 Wn.2d 654, 655-56, 277 P.2d 335 (1954); see also ER 802. Hearsay is a statement made by an out-of-court declarant offered to prove the truth of the thing said. ER 801(c). It is inadmissible unless an exception applies. ER 802. One such exception allows an expert to share the hearsay facts supporting their expert opinion to explain how they reached that opinion. ER 703, 705; In re Det. of

7

In re Dependency of A.C., No. 100966-6 Marshall, 156 Wn.2d 150, 162-63, 125 P.3d 111 (2005); Grp. Health Coop. of Puget Sound, Inc. v. Dep’t of Revenue, 106 Wn.2d 391, 399-400, 722 P.2d 787 (1986). But a judge cannot rely on that hearsay as substantive evidence. Grp. Health, 106 Wn.2d at 399-400; State v. Wineberg, 74 Wn.2d 372, 384, 444 P.2d 787 (1968); Pierce County ex rel. Bellingham v. Duffy, 104 Wash. 426, 430, 176 P. 670 (1918).

Our Court of Appeals has properly applied these principles to background reports underlying social worker testimony. See, e.g., In re Welfare of J.M., 130 Wn. App. 912, 924-25, 125 P.3d 245 (2005) (“An expert’s use of the written reports of absent witnesses is not substantive evidence; they are admissible solely to show the grounds upon which the testifying expert’s opinion is based.”). For example, in X.T., an expert social worker based her testimony supporting a child’s dependency solely on written reports. In re Welfare of X.T., 174 Wn. App. 733, 735-37, 300 P.3d 824 (2013). Those reports were mostly hearsay. Id. at 735-37, 739. Based entirely on this testimony, the judge declared the child dependent. Id. at 735-37. The Court of Appeals reversed the dependency. Id. at 739. Given that the evidence rules—and constitutional due process protections—apply to dependency hearings, the trial court erred by relying on the unsworn out-of-court testimony in those reports as substantive evidence. See id. at 738.

8

In re Dependency of A.C., No. 100966-6 Similarly here, the State concedes the trial court erred by relying on the hearsay for its truth. The more difficult question is how to evaluate whether the error undermines the trial court’s dependency order.

Typically, we review a trial judge’s application of our evidence rules in two steps. We first review for whether the judge erred. See In re Welfare of M.R., 200 Wn.2d 363, 376, 518 P.3d 214 (2022). We then review for whether that error was harmless. See id. (citing State v. Bourgeois, 133 Wn.2d 389, 399, 403, 945 P.2d 1120 (1997)). The parties here agree that the trial court erred in using the admitted hearsay beyond its limited purpose. Thus our task is limited to determining whether that error was just harmless error. The parties disagree on the standard used to make that determination.

The parents ask us to apply the “materially affected” standard and reverse the dependency. The State asks us to apply the “substantial evidence” standard and affirm the dependency. But “‘[t]here is a striking difference between appellate review to determine whether an error affected a judgment and . . . appellate review to determine whether there is substantial evidence to support a judgment.’” Standen v. Whitley, 994 F.2d 1417, 1423 (9th Cir. 1993) (quoting ROGER J. TRAYNOR, THE RIDDLE OF HARMLESS ERROR 27 (1970)).

9

In re Dependency of A.C., No. 100966-6 Under the substantial evidence standard, an appellate court will affirm a trial court’s findings of fact if they are supported by substantial evidence. In re Dependency of M.P., 76 Wn. App. 87, 90, 882 P.2d 1180 (1994). “Substantial evidence exists if, when viewing the evidence in the light most favorable to the prevailing party, a rational trier of fact could find the fact more likely than not to be true,” or, in short, by a preponderance of the evidence. X.T., 174 Wn. App. at 737 (emphasis added) (citing M.P., 76 Wn. App. at 90-91). This standard is regularly applied to challenges to the sufficiency of the evidence. E.g., In re Welfare of Sego, 82 Wn.2d 736, 739, 740, 513 P.2d 831 (1973) (applying standard to parental rights termination cases); M.P., 76 Wn. App. at 90 (applying standard to dependency cases). It is also helpful when the trial court erred in admitting a piece of evidence, but that evidence is minor in light of all the properly admitted evidence. See State v. Gonzales Flores, 164 Wn.2d 1, 19, 186 P.3d 1038 (2008) (“Evidence that is merely cumulative of overwhelming untainted evidence is harmless.”); State v. Davis, 154 Wn.2d 291, 305, 111 P.3d 844 (2005) (“[T]he untainted evidence was overwhelming, and any error in admitting testimonial statements . . . was harmless beyond a reasonable doubt.”). In such cases, courts will affirm if the overwhelming untainted evidence would sustain the finding. Flores, 164 Wn.2d at 19; Davis, 154 Wn.2d at 305.

10

In re Dependency of A.C., No. 100966-6 In contrast, under the “materially affected” standard of review, “[a]n erroneous admission of evidence is ‘not prejudicial unless, within reasonable probabilities, the outcome of the trial would have been materially affected had the error not occurred.’” X.T., 174 Wn. App. at 739 (emphasis added) (internal quotation marks omitted) (quoting Bourgeois, 133 Wn.2d at 403). Under this standard, “[t]he improper admission of evidence constitutes harmless error if the evidence is of minor significance in reference to the overall, overwhelming evidence as a whole.” Bourgeois, 133 Wn.2d at 403 (citing Nghiem v. State, 73 Wn. App. 405, 413, 869 P.2d 1086 (1994)).

In Bourgeois, we applied the materially affected standard. 133 Wn.2d at 403.

In that case, the trial court erred in allowing three witnesses to testify about their fear of retribution for testifying at a murder trial. Id. at 411. But given the enormous amount of properly admitted evidence supporting the conviction and the slight amount of potential prejudice, we “[did] not find that within reasonable probabilities, the outcome of the trial would have been different had they not so testified.” Id. at 405. Accordingly, “[i]n light of the evidence as a whole, the error was harmless.” Id.

Following these cases, the materially affected test focuses on the prejudicial effect of a trial court’s error. The test asks specifically whether it was reasonably probable that absent the error, the outcome of the trial would have been different.

11

In re Dependency of A.C., No. 100966-6 See id. at 403-05; X.T., 174 Wn. App. at 739. Phrased differently, if a different outcome was reasonably probable without the error, then the error had a material effect and the judgment should be reversed.

Here, the trial court erred when it admitted and relied on hearsay evidence introduced as background for an expert’s opinion, and we hold that the materially affected standard applies. Bourgeois, 133 Wn.2d at 403. See also Kotteakos v. United States, 328 U.S. 750, 765, 66 S. Ct. 1239, 90 L. Ed. 1557 (1946) (holding that the question for harmless error review is not whether enough evidence “support[ed] the result” but if “the error itself had substantial influence”); Standen, 994 F.2d at 1423 (holding that harmless error review requires considering “‘the probabilities of the effect of error on a reasonable trier of fact’” (quoting TRAYNOR, supra, at 27, 30)).

In applying this standard, we note that the burden of proof in dependency hearings is low in comparison to parental rights termination cases. 2 But

2 Compare In re Welfare of Key, 119 Wn.2d 600, 609, 836 P.2d 200 (1992) (holding that because dependency does not permanently deprive a parent of any rights, heightened constitutional protections are not required), and Schermer, 161 Wn.2d at 942 (noting that the preponderance of the evidence standard supports a dependency’s “important function of allowing state intervention in order to remedy family problems and provide needed services”), with Santosky v. Kramer, 455 U.S. 745, 758, 102 S. Ct. 1388, 71 L. Ed. 2d 599 (1982) (plurality opinion) (recognizing parents’s fundamental right to raise their child and holding that due process requires a standard of proof greater than preponderance of the evidence before that right’s termination), and Sego, 82 Wn.2d at 743-44 (holding that findings in termination cases must be supported by clear, cogent, and convincing evidence).

12

In re Dependency of A.C., No. 100966-6 dependency still implicates fundamental rights and must be approached with due solemnity. Schermer, 161 Wn.2d at 941. Our statutory scheme reflects the vital interests at stake and protects them in many ways, including by requiring that dependency hearings be conducted under the rules of evidence. RCW 13.34.110(1). As a natural consequence, parents “should not be deprived of their parental rights on hearsay.” Ross, 45 Wn.2d at 655-56; see, e.g., In re Ty.B., 878 A.2d 1255, 1266 (D.C. 2005) (holding that even if “an impartial trier of fact could reasonably have found” a child dependent on nonhearsay evidence, when “the inadmissible predominated over the admissible, the judgment cannot stand”).

Here, we are faced with an enormous amount of evidence considered beyond its properly limited purpose. While some of this hearsay was properly admitted for the narrow purpose of providing background for an expert opinion, it cannot serve as substantive evidence. ER 703, 705; Marshall, 156 Wn.2d at 162-63. Through that hearsay, the trial court heard accounts of domestic abuse, criminal warrants, heavy cannabis use, confrontations with medical staff, CC’s potential “developmental delays” and alleged physical disability, and—most damaging for VC—multiple encounters with the police. Very little of this information came directly from its source, and its introduction at trial allowed the parents to be examined on it without allowing them, in turn, to challenge and cross-examine those who made the statements originally.

13

In re Dependency of A.C., No. 100966-6 The parents were questioned about the hearsay on the stand, and their testimony was properly considered by the trial court. Credibility determinations are for the trier of fact. But here, the court, as the trier of fact, went beyond simply finding the parents’ testimony not credible. It repeatedly relied on the hearsay relayed by the social workers to explain their opinions for its truth. While the trial court was empowered to reject the parents’ version of events, it was not entitled to rely on hearsay—admitted only for the limited purpose of explaining an expert opinion—beyond its limited purpose. See Grp. Health, 106 Wn.2d at 399-400 (explaining that an expert’s “explanation is not proof of the facts . . . consider[ed]” (quoting Wineberg, 74 Wn.2d at 382)); cf. Duffy, 104 Wash. at 430 (ordering reversal where trial court relied on expert’s hearsay because hearsay “rests . . . on the veracity and qualifications of some other person or persons, not exposed to cross-examination in court nor speaking under the sanction of an oath”). The effect of the hearsay here is not just “slight[ly] prejudicial,” as it was in Bourgeois. 133 Wn.2d at 404. The taint of the improperly relied on hearsay—the court’s acceptance of that hearsay as truth—affected the court’s view of all the admissible evidence, which includes the parents’ admission to and dispute of the events described by the hearsay. In short, the trial court relied heavily on the hearsay for its findings and evaluations. The weight of that reliance, within reasonable probabilities, materially affected the outcome of the hearing.

14

In re Dependency of A.C., No. 100966-6 This is not a case where we can presume the judge, acting as the fact finder, disregarded the inadmissible hearsay. See generally State v. Gower, 179 Wn.2d 851, 856, 321 P.3d 1178 (2014) (holding inapplicable the presumption that judges do not rely on inadmissible evidence when a trial judge actually considers inadmissible evidence in their findings (citing State v. Read, 147 Wn.2d 238, 245- 46, 53 P.3d 26 (2002))). To the contrary, the trial court explicitly relied on the inadmissible hearsay and drew inferences from it. For example, the trial court found that “at every turn, there is an inability of the father to follow the basic policy/structure of what is required,” as a basis for dependency. CP at 148 (emphasis added). There was some admissible evidence that VC struggled to follow rules and expectations. But the majority of accounts where rule-following was required came from hearsay evidence relayed by hospital staff and therapists who did not testify. Nearly all of the trial court’s findings similarly and impermissibly rely on the hearsay for its truth.

We hold that within reasonable probabilities, the trial court’s impermissible reliance on hearsay prejudiced the parents and materially affected the outcome of the trial. We reverse the trial court’s dependency finding for AC as to both parents.3

3 VC also challenges the court-ordered chemical dependency and anger-management/domestic violence assessments. Given our disposition, we do not reach this issue.

15

In re Dependency of A.C., No. 100966-6 CONCLUSION

We reverse the Court of Appeals, reverse the dependency, and remand to the trial court for further proceedings consistent with our opinion.

____________________________

WE CONCUR:

_____________________________ ____________________________

_____________________________ ____________________________

_____________________________ ____________________________

_____________________________ ____________________________

In re Dependency of A.C.

No. 100966-6

MADSEN, J. (concurring/dissenting)—In this case, the dependency court improperly admitted hearsay. That the court erred is undisputed. The primary issue before this court is whether the error of improperly admitting hearsay evidence was harmless.

The parties propose different standards for analyzing harmless error: the State urges us to apply a substantial evidence test, while the parents would have us apply a materially affected standard. The majority chooses the materially affected standard— taken from a criminal case—and offers little explanation on the difference between the two, on how to apply the standard, or why the claimed errors satisfy the standard in this case. See majority at 11-13.

Although I join the majority’s conclusion that the rule for nonconstitutional harmless error is the materially affected standard, we must say more. First, I would hold that the harmless error standard for nonconstitutional error is the same in criminal and civil cases, and I would add that to “materially affect” the outcome of a proceeding means that a reviewing court must find that there is a reasonable probability that had the

No. 100966-6 Madsen, J., concurring/dissenting

error not occurred, the outcome would have been different in light of the remaining properly admitted evidence weighed under the prescribed burden of proof.

FACTS AND PROCEDURAL HISTORY V.C. and C.C. are the father and mother of A.C. Clerk’s Papers (CP) at 2. 1 After A.C. was born, the Department of Children, Youth, and Families investigated the family based on concerns about V.C. and C.C., eventually filing a dependency motion. The department assigned investigator Michelle Woodward and social worker Diana Barnes to work with the parents. At the contested fact-finding hearing, the dependency court admitted hearsay evidence about, among other things, V.C.’s interactions with law enforcement and also concerns about domestic violence. The court commissioner found that the department proved A.C. was dependent pursuant to RCW 13.34.030(6)(c). 2 The court’s written ruling states,

[By a] preponderance of the evidence standard, which means more likely than not[, the Court found:]

In this case, the court reviewed a lot of information and does have concerns of ongoing mental health issues and aggression and/or violence.

It appears to the court that no matter what type of situation the father is in, whether it be at the hospital, visitation, the bus plaza, etc., there is law enforcement involved. Ms. Wright [(a parenting skills instructor)] testified that the father’s engagement in the parenting program escalated to a point where the focus was no longer on the child or on caring for the child, but rather the focus was on the father’s escalation. It appears that at every turn, there is an inability of the father to follow the basic policy/structure of what is required.

1 Like the majority, I also refer to A.C.’s mother as C.C. Majority at 2 n.1. 2 RCW 13.34.030(6)(c) provides that a child is “dependent” if they have “no parent, guardian, or custodian capable of adequately caring for the child, such that the child is in circumstances which constitute a danger of substantial damage to the child’s psychological or physical development.”

No. 100966-6 Madsen, J., concurring/dissenting

Housing is also a concern to the court. There has been instability with the parents[’] housing situation since shelter care. The parents were on their way to Tacoma when the car broke down and the mother’s water broke at that time.

The past history is also concerning. The mother has prior CPS history where her child was not placed with her. The father reports having power of attorney over the mother. The court also considered testimony that the mother was banging her head on the wall in frustration. There has also been criminal history with the father that involves violence.

The needs of this child appear to be overshadowed by the parent’s actions. It was difficult for the court to follow the testimony of the father’s and whether or not he believes these services to be necessary and whether he will follow through with them.

The court is absolutely concerned about the safety of this three month old child. Here we have a child who is only three months and is unable to self-protect and speak for himself. The child was born with THC in his system. The court is concerned about mental health, aggression, chemical dependency, lack of parental experience, and inability to put the child’s needs before their own.

CP at 141 (Agreed Ord. of Dependency & Ord. of Disposition at 2). The court found that the parents were deficient due to unstable housing; past history; and V.C.’s mental health, aggressive and controlling behavior, and chemical dependency. As a result, the court ordered services for both parents. For C.C., the court required a neuropsychological and mental health evaluation, a domestic violence victim assessment, and evidence-based parenting classes. Verbatim Tr. of Proc. (VTP) at 172-73. For V.C., the court ordered assessments for anger management, domestic violence, and chemical dependency, as well as a neuropsychological evaluation and evidence-based parenting classes. VTP at 173- 74.

The Court of Appeals concluded that the dependency court erred when it considered some of the hearsay testimony from Woodward and Barnes for its truth. In re

No. 100966-6 Madsen, J., concurring/dissenting

Dependency of A.C., No. 37999-0-III, slip op. at 22 (Wash. Ct. App. Apr. 28, 2022) (unpublished), https://www.courts.wa.gov/opinions/pdf/379990_unp.pdf. The court identified two statements from the dependency order: no matter what type of situation V.C. is in, whether at the hospital, visitation, or the bus plaza, etc., law enforcement is involved and V.C.’s inability to follow basic and required policy and structure at every turn. Id. at 22-23. These statements were based on third party reports rather than firsthand experience, and constituted inadmissible hearsay.

Here, the parties agree with the Court of Appeals’ conclusion that those statements were hearsay. The parents additionally contend that the dependency court relied on other impermissible hearsay regarding housing instability and V.C.’s power of attorney, which they claim the Court of Appeals overlooked. The parents also argue that the hearsay evidence was not harmless error, that hearsay pervaded the record and the dependency court’s order, and reversal is the only remedy.

ANALYSIS

The majority agrees with the parents that the hearsay evidence about law enforcement and domestic violence was prejudicial, and it adds to that list testimony about V.C.’s criminal warrants, his heavy cannabis use, his confrontational behavior, and C.C.’s developmental and physical disabilities. Majority at 12. But the majority does not show its work. In fact, much of this testimony was not hearsay, as I explain below. Moreover, the majority’s list of hearsay statements lacks context, isolating pieces of evidence instead of scrutinizing the entire record to determine whether the claimed errors

No. 100966-6 Madsen, J., concurring/dissenting

were harmless. See State v. Thomas, 110 Wn.2d 859, 863, 757 P.2d 512 (1988) (quoting State v. Britton, 27 Wn.2d 336, 341, 178 P.2d 341 (1947)). In doing so, the majority shows only some possibility that the hearsay affected the result rather than a reasonable probability. See id. (“[I]t is not the fact that every event or omission in a trial might conceivably have some effect upon the verdict. Rather the inquiry is whether it has a material effect.”).

For the following reasons, I disagree that the hearsay errors were so prejudicial as to require reversal of the dependency order, and I disagree that a simple materially affected inquiry is the proper inquiry. Instead, we should explain what “materially affected” means and, in my view, the standard should include an evidentiary measure that focuses the analysis on the remaining admissible evidence.

A. Harmless Error

The harmless error doctrine and its application are far from straightforward, despite the majority’s treatment. Washington courts have struggled with harmless error (of the constitutional and nonconstitutional variety alike) for decades. See State v. Evans, 96 Wn.2d 1, 6, 663 P.2d 83 (1981) (Brachtenbach, C.J., concurring) (“The problem of harmless error has long plagued the courts.”); Dennis J. Sweeney, An Analysis of Harmless Error in Washington: A Principled Process, 31 GONZ. L. REV. 277, 286-96 (1996) (listing the numerous standards of review for harmless error). Different courts have used different rules to evaluate evidentiary harmless error, creating a tangled web of case law. In a seeming effort to settle some of this confusion, the majority selects a rule

No. 100966-6 Madsen, J., concurring/dissenting

for analyzing evidentiary harmless error: an error is not prejudicial unless, within reasonable probabilities, the outcome of the trial would have been materially affected had the error not occurred. State v. Bourgeois, 133 Wn.2d 389, 403, 945 P.2d 1120 (1997). When properly articulated, the rule is clear. What is unclear is why “materially affected” is the proper standard and what analysis a reviewing court should use to determine whether an error is harmless.

In my view, the majority offers little clarity for reviewing courts—which will be left questioning what the materially affected standard actually means and how to apply it. Instead of leaving it for future courts to divine, we should provide this guidance. I would clarify that an error is not prejudicial unless the appellant demonstrates a reasonable probability that the error materially affected the outcome based on the remaining properly admitted evidence when considered against the final legal question to be answered. Thus, the question here is whether there was a reasonable probability that the error materially affected the court’s conclusion that A.C. was dependent under RCW 13.34.030(6)(c) by a preponderance of the evidence, considering the remaining admissible evidence. I concur with the majority that the proper harmless error standard is the materially affected standard, but I disagree with its articulation and application of that standard.

Determining the Appropriate Standard for Nonconstitutional Harmless Error Our case law is largely consistent in articulating the standard for harmless nonconstitutional error. The error is not prejudicial unless, within reasonable

No. 100966-6 Madsen, J., concurring/dissenting

probabilities, the outcome of the trial would have been materially affected had the error not occurred. Bourgeois, 133 Wn.2d at 403 (citing State v. Tharp, 96 Wn.2d 591, 599, 637 P.2d 961 (1981); State v. Halstien, 122 Wn.2d 109, 127, 857 P.2d 270 (1993)); State v. Everybodytalksabout, 145 Wn.2d 456, 468-69, 39 P.3d 294 (2002).

The majority and the parties frame this case as a choice between two rules.

Majority at 8. This framing no doubt comes from the Court of Appeals’ decision below, which held that erroneous hearsay evidence was admitted and reviewed whether that error was harmless. Yet the court applied a standard of review for insufficient evidence. A.C., No. 37999-0-III, slip op. at 24 (citing In re Dependency of M.P., 76 Wn. App. 87, 90, 882 P.2d 1180 (1994)). The Court of Appeals’ reliance on M.P. for its harmless error analysis was misplaced. That case did not concern harmless error; it reviewed only an insufficiency of the evidence claim. See M.P., 76 Wn. App. at 90. By reviewing this case for substantial evidence, the Court of Appeals conflated the harmless error and insufficiency inquiries, an understandable misstep considering that under both standards, the court reviews the weight of evidence admitted at trial measured by the appropriate level of proof, and both V.C. and C.C. brought insufficient evidence claims. Many cases present both harmless error and insufficiency issues, and the similarities in standards could explain, at least in part, why some reviewing courts have conflated them.

Even when the standards for insufficient evidence and harmless error are separated, questions remain. The cases setting out the materially affected standard above and those cited by the majority are exclusively criminal. See majority at 10-11. The case

No. 100966-6 Madsen, J., concurring/dissenting

before us is not—it is a civil dependency action. See In re Dependency of Grove, 127 Wn.2d 221, 226, 897 P.2d 1252 (1995) (acknowledging that a dependency action is civil). To state the obvious, civil and criminal proceedings diverge in numerous ways such as structure, burdens of proof, and penalties. E.g., In re Det. of Reyes, 184 Wn.2d 340, 347-48, 358 P.3d 394 (2015) (noting that certain rules of criminal law are not extended to sexually violent predator proceedings because they are civil in nature and distinct from criminal convictions and punishment is not an objective); Jones v. Sisters of Providence in Wash., 93 Wn. App. 727, 734, 970 P.2d 371 (1999) (stating the standard for waiving a jury trial differs between civil and criminal proceeding attributed to a court’s duty to safeguard a criminal defendant’s constitutional right to a jury trial); Helvering v. Mitchell, 303 U.S. 391, 397, 58 S. Ct. 630, 82 L. Ed. 917 (1938) (“The difference in degree of the burden of proof in criminal and civil cases precludes application of the doctrine of res judicata.”); 14A DOUGLAS J. ENDE, WASHINGTON PRACTICE, CIVIL PROCEDURE § 35:50 n.7, at 632-33 (3d ed. 2018) (recognizing that the burden of proof in a criminal case is higher than in civil cases and that the cases in which they are the same are relatively rare). It is far from certain, then, that the standard to be met for harmless error is the same for criminal and civil proceedings. 3

3 Nor is the substantial evidence standard of review the same in criminal and civil proceedings. Cf. majority at 9. In a criminal case, a conviction is affirmed if the appellate court, viewing the evidence in a light most favorable to the State, is satisfied that sufficient evidence justifies a rational trier of fact to find guilt beyond a reasonable doubt. 5 ELIZABETH A. TURNER, WASHINGTON PRACTICE: EVIDENCE LAW AND PRACTICE § 301.7, at 210 (6th ed. 2016); see, e.g., State v. Homan, 181 Wn.2d 102, 105, 330 P.3d 182 (2014). The criminal test is “somewhat more rigorous than” that in a civil case where courts review whether substantial evidence exists

No. 100966-6 Madsen, J., concurring/dissenting

The Court of Appeals has used the materially affected test in civil cases. E.g., Lutz Tile, Inc. v. Krech, 136 Wn. App. 899, 905, 151 P.3d 219 (2007) (Reversal is required “if it is reasonable to conclude that the trial outcome would have been materially affected had the error not occurred.” (emphasis added)); Brundridge v. Fluor Fed. Servs., Inc., 164 Wn.2d 432, 446, 191 P.3d 879 (2008) (same). 4 At other times, reviewing courts have omitted the word “materially” and simply stated that an evidentiary error is “harmless unless it affects the outcome of the case.” Mut. of Enumclaw Ins. Co. v. Gregg Roofing, Inc., 178 Wn. App. 702, 729, 315 P.3d 1143 (2013); Brown v. Spokane County Fire Prot. Dist. No. 1, 100 Wn.2d 188, 196, 668 P.2d 571 (1983).

As these cases illustrate, Washington courts have accepted but not decided that the same nonconstitutional harmless error standard applies in both criminal and civil cases. This case presents an opportunity to resolve the issue definitively. Therefore, I would hold that for harmless nonconstitutional error, in both criminal and civil cases, the standard is materially affected. Bourgeois, 133 Wn.2d at 403; Brundridge, 164 Wn.2d at 446.

to support a finding or verdict. 5 TURNER, supra; see, e.g., Mitchell v. Wash. State Inst. of Pub. Pol’y, 153 Wn. App. 803, 814, 225 P.3d (2009). This distinction does not influence either the majority’s or my harmless error analysis but should serve as a caution when importing possibly distinct standards between civil and criminal cases. 4 Despite citing Bourgeois, Brundridge slightly altered the standard when it omitted “materially affected” and instead stated that an “error is harmless unless it was reasonably probable that it changed the outcome of the trial.” 164 Wn.2d at 452. It may be that the wording (“materially affected the outcome” and “changed the outcome”) is a distinction without a difference.

No. 100966-6 Madsen, J., concurring/dissenting

While the materially affected standard has been applied in both civil and criminal contexts, our case law has been less than precise on how to apply the standard. The Bourgeois court explained, “In assessing whether the error was harmless, we must measure the admissible evidence . . . against the prejudice, if any, caused by the inadmissible testimony.” 133 Wn.2d at 403. In Tharp, on which Bourgeois relied, the court stated that “we focus on the evidence that remains after excluding the [inadmissible evidence].” 96 Wn.2d at 599; see also State v. Myers, 49 Wn. App. 243, 249-50, 742 P.2d 180 (1987) (“To determine the probable outcome, the focus must shift to the evidence which remains after the prior acts of misconduct have been excluded.”). Bourgeois measured the admissible evidence against the erroneous evidence to determine whether the outcome was affected, but Tharp examines the remaining evidence to determine whether the verdict would have been the same.

Bourgeois is internally inconsistent. It does little to clarify (and a lot to confuse)

how a reviewing court is to analyze harmless error. Are we to measure all the evidence, admissible and inadmissible, for prejudice according to Bourgeois? Or are we to apply Tharp, the authority Bourgeois relies on, and the many other cases that look only at the admissible evidence to decide whether the outcome would have been the same. That leads us back to the beginning. Bourgeois does not answer the question of how to determine prejudicial error, despite the majority’s treatment of the case.

Moreover, without further explanation the test itself creates confusion. What does “materially affected” mean? Is any influence on the verdict sufficient? Our past

No. 100966-6 Madsen, J., concurring/dissenting

harmless error decisions do not supply a definition for “materially.” When words are undefined, courts may look to the dictionary to determine the words’ common meaning. Allstate Ins. Co. v. Peasley, 131 Wn.2d 420, 424, 932 P.2d 1244 (1997). Black’s Law Dictionary defines “material” to mean “[o]f such a nature that knowledge of the item would affect a person’s decision-making; significant; essential.” BLACK’S LAW DICTIONARY 1170 (11th ed. 2019). To materially affect an outcome, then, means to significantly affect it or affect it in an essential way.

Adding to the confusion, the majority characterizes the standard as the error is prejudicial if it is reasonably probable the error materially affected the outcome, focusing on the erroneously admitted evidence. See majority at 10-12. But, under the case law, the standard is the opposite: an error is not prejudicial unless it is reasonably probable that the outcome would have been materially affected had the error not occurred. This means that the court considers whether it is reasonably probable the outcome would have been different had the hearsay not been admitted; that is, the probable outcome in light of the properly admitted evidence. The question is not the reasonably probable effect of the error (admission of hearsay) on the court’s decision but the reasonably probable outcome in the absence of the error (consideration of only the admissible evidence). This articulation focuses the court on the remaining admissible evidence, not on the erroneously admitted evidence.

Although we are now armed with a more nuanced definition of “material,” we also need to resolve how to apply the materially affected standard to the facts of this and

No. 100966-6 Madsen, J., concurring/dissenting

future cases. The majority does not help. It accepts Bourgeois, 5 then simply lists the hearsay evidence contained in the social workers’ testimony to conclude it materially affected the outcome of the dependency hearing. Id. at 12-13.

Bourgeois is similarly unhelpful. Bourgeois’ internal inconsistency limits its use and creates uncertainty, as does the majority’s incomplete effort in applying the law to the facts of the present case. Harmless error is already a doctrine rife with complexities. As the court of last resort in this state, we should be untangling the law, not proceeding as if the knots do not exist.

Instead, I would follow Tharp and clarify what materially affected means as an evidentiary matter. Under Tharp, the question is whether there is a reasonable probability that the court would have concluded by a preponderance of the evidence that A.C. was not dependent under RCW 13.34.030(6)(c) considering the remaining

5 In selecting the materially affected standard as articulated in Bourgeois, the majority invokes the fundamental rights of parents to raise their children and the “vital interests at stake.” Majority at 12 (citing In re Dependency of Schermer, 161 Wn.2d 927, 941, 169 P.3d 452 (2007)). Accordingly, the majority concludes, parents should not be deprived of their rights based on hearsay. Id. But the materially affected standard asks whether the court would likely have made the same decision based on the admissible evidence. Moreover, the parents here will not be deprived of their rights. Dependency performs the important function of allowing State intervention to remedy family problems and provide services. Schermer, 161 Wn.2d at 942. It does not permanently deprive parents of their fundamental rights, unlike a termination action. Id. at 943 (citing In re Dependency of A.W., 53 Wn. App. 22, 30, 765 P.2d 307 (1988)). And, equally important to a parent’s right is the safety of the child. Id. at 941. The State has an interest in protecting the physical, mental, and emotional health of children. Id. When that health is put at serious risk by parental deficiencies, the State has the “‘parens patriae right and responsibility to intervene to protect the child.’” Id. at 941-42 (quoting In re Welfare of Sumey, 94 Wn.2d 757, 762, 621 P.2d 108 (1980)). I am unsure how the stated parental considerations support the materially affected standard when we have said that on balance, “‘the rights and safety of the child . . . shall be the paramount concern.’” Id. at 942 (alteration in original) (quoting RCW 13.34.020).

No. 100966-6 Madsen, J., concurring/dissenting

admissible evidence. This articulation combines the harmless error standard with a practical means of measuring how to satisfy it. The inquiry is not whether the error had an effect on the outcome but what would the reasonably probable outcome be in absence of the error? In this way, the harmless error standard asks the same question as a sufficiency of the evidence claim after removing the hearsay from consideration.

To determine whether erroneously admitting hearsay resulted in prejudice that materially affected the outcome, a court must examine the entire record. Bourgeois, 133 Wn.2d at 403; Britton, 27 Wn.2d at 341 (it is a reviewing court’s “duty . . . to scrutinize the entire record in each particular case, and determine whether or not the error was harmless or prejudicial”). A dependency determination requires a showing of parental deficiency, not parental unfitness. In re Dependency of Schermer, 161 Wn.2d 927, 943, 169 P.3d 452 (2007). This does not require proof of actual harm but a danger of harm. Id. at 951. Dependency courts have broad discretion in evaluating the risk of harm. Id.

Before thoroughly reviewing the record under harmless error, I consider first the parents’ additional hearsay challenges.

B. Additional Hearsay Challenges “Hearsay” is a statement made outside of court offered in evidence to prove the truth of the matter asserted. ER 801(c). Hearsay evidence is not admissible unless it falls under a recognized exception to the hearsay rule in which case its reliability is presumed. ER 802; State v. Athan, 160 Wn.2d 354, 383, 158 P.3d 27 (2007). “The hearsay prohibition serves to prevent the jury from hearing statements without giving the

No. 100966-6 Madsen, J., concurring/dissenting

opposing party a chance to challenge the declarants’ assertions.” Brundridge, 164 Wn.2d at 451-52.

1. Past History

The parents argue that the dependency court’s findings on their past history rely on hearsay. Specifically, C.C. contests Woodward’s testimony that C.C. was developmentally delayed, had banged her head against the wall in frustration, was not ambulatory, and had prior Child Protective Services (CPS) history in which C.C. lost parental rights to another child. The dependency order found that C.C. had hit her head against the wall and noted the past involvement with CPS.

C.C. is correct, in part. Woodward’s testimony about C.C. hitting her head is based on third party reports and constitutes hearsay. It was error for the dependency court to adopt it as a finding. Woodward’s testimony about C.C.’s ability to ambulate was based on hospital reports and was also hearsay, even though Woodward stated that she saw C.C. able to walk around and was not concerned about it. VTP at 33. And, C.C. told the court that she had physical difficulty getting up and down due to her weight.

Regarding C.C.’s first child, Woodward relayed information she obtained after speaking with C.C.’s extended family, who were not called to testify. However, C.C. confirmed Woodward’s testimony about her history with CPS. This statement was made by another but adopted by C.C. ER 801(d)(2)(ii). Adoptive admissions are considered a statement made by the party they are being offered against, even though a third party spoke them. Id. Therefore, it is not hearsay.

No. 100966-6 Madsen, J., concurring/dissenting

2. V.C.’s Power of Attorney and Other Comments about C.C.

Woodward testified to V.C.’s comments on caring for C.C. According to C.C., these statements are hearsay, they are unrelated to her parenting abilities, and no admissible documentary evidence of power of attorney was offered. Thus, C.C. argues, the Court of Appeals’ reliance on it was erroneous and prejudicial.

The record shows that V.C. told Woodward he has power of attorney over C.C.; he made all the decisions for the family as head of the household; and, because C.C. could not care for herself, he was her only caregiver. C.C. testified that when she was at the visitation center, she was unable to use certain equipment because her weight prevented her from “get[ting] up and down.” VTP at 132.

V.C.’s statements are not hearsay under ER 801(d)(2)(i), as they relate to V.C. A statement is not hearsay if it is offered against a party and is a statement that the party has adopted or believed to be true. ER 801(d)(2)(ii). Testimony that C.C. could not care for herself would generally be inadmissible against C.C. because admissions by one party are admissible only against that party and are not admissible as substantive evidence against another party. 5B KARL B. TEGLAND & ELIZABETH A. TURNER, WASHINGTON PRACTICE: EVIDENCE LAW AND PRACTICE § 801.34, at 43 (6th ed. Supp. 2022). Yet C.C. testified that at one point she could not get up and down, indicating a physical challenge. C.C. did not testify that V.C. was her only caregiver or that she was completely unable to care for herself, but she did testify that she had difficulty completing a task to care for her

16

No. 100966-6 Madsen, J., concurring/dissenting

child. Thus, C.C. is correct in part. V.C.’s statements that he was C.C.’s only caregiver could not be used as substantial evidence against her.

3. Unstable Housing

The court below also found that housing instability contributed to the finding of dependency. The written order notes that since A.C.’s shelter care hearing, the parents had not found stable housing and that they had been driving to Tacoma when A.C. was born. CP at 141. C.C. challenges this finding as hearsay.

Woodward testified that the parents’ living situation was unstable and that she thought they had been living in their car prior to A.C.’s birth. VTP at 21, 25. She stated that V.C. had told her “several times . . . that they actually prefer to live in cars; they think that’s a better way to live.” VTP at 21. Woodward discussed housing resources with the parents but later learned that because of V.C.’s registration as a sexual offender, many resources were unavailable to them. VTP at 22-23, 35. On cross examination, Woodward admitted she was not aware of the parents’ current housing arrangement. VTP at 28.

Barnes also testified about the parents’ housing instability. She noted that the parents had lived in their car, were currently staying in a hotel, but she was unsure about the parents’ long-term housing. VTP at 44-45. Barnes relayed that V.C had told her they “would rather live in their car to save money.” VTP at 45. Asked whether Barnes was aware of whether the parents had resources to afford housing, she stated they received Social Security income and temporary assistance funds. Id.

17

No. 100966-6 Madsen, J., concurring/dissenting

V.C. also testified. He disagreed that he and C.C. had lived in their car or preferred to do so. VTP at 79-80. Instead, he stated they would stay a couple of nights in the car but then stay at a friend’s house. VTP at 80. V.C. reported that when the weather was warmer, they would stay in their car a “few nights.” Id. V.C. and C.C. planned to reside long-term in Spokane, and they could afford the rent. VTP at 113. V.C. told the court he was open to the department visiting their current hotel room to determine whether it was suitable for A.C. VTP at 96. C.C.’s testimony echoed V.C.’s, saying they planned to stay in Spokane and she had Social Security income as financial support. VTP at 122. C.C. further testified that they were staying in a hotel at the time of the dependency hearing and had stayed in three different hotels since the shelter care hearing. Id.

Here, C.C. asserts that the unstable housing finding was based on hearsay and speculation, but she does not meaningfully argue this point. She offers no case law or other authorities to demonstrate why any of the above statements should not have been admitted. Rather, C.C.’s argument is that there is insufficient evidence to support the finding that the parents’ housing was unstable. Insufficiency is a separate inquiry from hearsay.

In any event, the testimony from Woodward and Barnes about what V.C. had told them is not hearsay. A statement is not hearsay if the statement is offered against a party and is the party’s own statement. ER 801(d)(2)(i). Admissions of a party-opponent may be admitted as substantive evidence. Saldivar v. Momah, 145 Wn. App. 365, 400, 186

18

No. 100966-6 Madsen, J., concurring/dissenting

P.3d 1117 (2008). Woodward and Barnes both stated that V.C. had said the parents preferred to live in their car, which V.C. disputed in his own testimony. It was not error for the dependency court to rely on this testimony from Woodward and Barnes.

C. Applying the Harmless Error Standard Having considered the parents’ additional hearsay claims, the next question is whether the outcome probably would have been different considering the admissible evidence had those errors not occurred. Essential to this inquiry is a thorough examination of the record, as our harmless error precedent requires. See Britton, 27 Wn.2d at 341.

1. Parents’ History

The dependency court found parental deficiency based in part on the parents’

history. In re Dependency of Brown, 149 Wn.2d 836, 841-42, 72 P.3d 757 (2003) (noting a parent’s past history is a factor that may be weighed in evaluating the current risk to a child). For C.C., the court’s written findings adopted some hearsay evidence, but it did not materially affect the outcome. V.C. identified no inadmissible hearsay, thus, there was no error.

Regarding C.C.’s history, Woodward testified, and the dependency court found, that C.C. frustratedly banged her head against the wall. CP at 141. C.C. also contends that the statements about her potential developmental delay and ability to ambulate affected the dependency order. Had this been the only evidence to conclude C.C. was deficient, the error would probably have affected the outcome.

19

No. 100966-6 Madsen, J., concurring/dissenting

But the court relied on other admissible evidence of C.C.’s history that supported the court’s finding of parental deficiency. The written ruling found that C.C. had been involved with CPS for a different child, who was not placed with her and was adopted by a family member. C.C.’s own testimony confirmed this: C.C. relinquished her rights to her first child before her incarceration. VTP at 136. Due to that incarceration, C.C had not parented a child for some time. Further, the department was unable to fully assess C.C.’s current parenting skills due to V.C.’s constant intervention and control over their conversations. VTP at 51. Indeed, neither parent’s skills could be properly assessed because of V.C.’s behavior, and Barnes concluded that the behavior would present an ongoing challenge given the parents’ desire to raise A.C. together. Id.

As to C.C.’s potential delays and ability to ambulate, the dependency court did not reference any concern that C.C. was vulnerable or had developmental delays in its written findings. Woodward mentioned the concerns in relation to C.C.’s ability to care for herself, which C.C. called into question when she testified that she had physical challenges due to her weight. The hospital’s report that C.C. was not ambulatory was also hearsay, but Woodward testified that she personally observed C.C. walking and did not share the hospital’s concern.

For V.C., the dependency court was troubled by his “power of attorney.” The record is replete with statements from V.C. that power of attorney provides him significant if not total control over C.C. and the family. Woodward testified that V.C. reminded her several times that he had power of attorney and that as head of the

20

No. 100966-6 Madsen, J., concurring/dissenting

household, he made all of the decisions. VTP at 15-16. V.C. told the investigator that power of attorney means he has full control over C.C. and everything she does. VTP at 19. When the investigator attempted to talk to C.C. alone about domestic violence, V.C. would not allow it because he “‘ha[s] power of attorney over [C.C.]’” and he had to speak for her. VTP at 20. Both parents confirmed that V.C. has power of attorney, though they disputed that he made all the family decisions. VTP at 85, 125.

The dependency court properly relied on this evidence. V.C. made damaging statements to the department’s representatives, who later testified to them. See ER 801(d)(2)(ii). V.C.’s statements justified the court’s concern about his role as decision- maker for C.C. and about his disruptive behavior that prevented the department from assessing the skills of either parent and hindered V.C.’s ability to adequately care for his child.

Finally, the dependency court found V.C.’s criminal history concerning. V.C.

reported to Woodward that he was convicted of four felonies: possessing cannabis, the sexual battery of a minor, aggravated battery of a corrections officer, and failure to register as a sex offender. None of V.C.’s criminal history included domestic violence. Barnes told the dependency court that she was concerned about V.C.’s ability to parent due to his 17 years of incarceration, which interfered with his experience parenting his other, now-adult children. VTP at 52. While Woodward included hearsay from law enforcement when testifying that V.C. had an active warrant from Idaho, Woodward testified that V.C. himself confirmed this information to her. VTP at 18.

21

No. 100966-6 Madsen, J., concurring/dissenting

In short, V.C.’s criminal history included violent offenses and his own statements provided damaging evidence about his relationship with C.C. Both parents testified about their lack of experience caring for their other children due to incarceration. The court found this past history—particularly V.C.’s domineering view of power of attorney—put A.C. in danger of substantial damage to his psychological or physical development. See RCW 13.34.030(6)(c).

2. V.C.’s Behavior and Domestic Violence The parents also argue that the hearsay evidence about domestic violence and law enforcement interactions materially affected the dependency court’s ruling. I disagree.

Woodward told the dependency court about the hospital staff’s domestic violence concerns, including an incident in which V.C. was physically removed from C.C.’s hospital room after grabbing hold of her, refusing to leave, and eventually damaging equipment. VTP at 14. Woodward also relayed information from C.C.’s family members about two acts of physical violence: V.C. allegedly dragged C.C. by her arms while she was pregnant and at one point assaulted her, which C.C. later said was her fault. VTP at 20-21. Both parents testified, disputing any violence in the relationship and stating that V.C. did not control C.C. VTP at 100-01, 125, 133.

The portions of Woodward’s testimony based on reports from hospital staff and C.C.’s family do not constitute substantive evidence but were properly admitted under ER 703 and 705. The Court of Appeals reasoned that when V.C.’s attorney objected to hearsay in the hearing, the State did not cite those evidentiary rules, but relied on their

22

No. 100966-6 Madsen, J., concurring/dissenting

substance. A.C., No. 37999-0-III, slip op. at 19. ER 703 states that the facts on which an expert bases their opinion need not be admissible if they are of a type reasonably relied on by experts in the field in forming opinions. ER 705 provides that an expert may testify to an opinion and give the reasons they relied on in doing so without disclosing the underlying facts unless the judge requires otherwise or they are questioned about those facts in cross examination. Nevertheless, these rules were not designed to allow witnesses to recall inadmissible evidence—ER 705 is not a means of shoehorning hearsay into evidence unless it is necessary to help the fact finder understand the expert’s opinion. Id. (quoting In re Det. of Marshall, 156 Wn.2d 150, 162, 125 P.3d 111 (2005)).

Had the dependency court’s findings on domestic violence and V.C.’s concerning behavior been based solely on the preceding hearsay, the parents would be correct that the error materially affected the outcome, that is, there was insufficient remaining evidence. But the record shows that other, admissible evidence was presented supporting both findings.

Woodward was qualified as an expert on domestic violence. She explained that it was common for there to be hidden confrontation with the victim in abusive relationships. VTP at 39-40. Specific to the parents’ relationship here, Woodward stated that V.C. told her how he controlled the family, making all the decisions, including how and where the money is spent. VTP at 40. Woodward was never allowed to speak with C.C. alone, and C.C. agreed that she never spoke with the department representatives without V.C. present. VTP at 14-15, 20, 126.

23

No. 100966-6 Madsen, J., concurring/dissenting

Excising the hearsay testimony from the hospital and C.C.’s family, substantial evidence exists in the form of Woodward’s testimony, which was based on her firsthand experience with the parents or statements from the parents themselves.

V.C. often demonstrated controlling and confrontational behavior. Woodward testified to threats V.C. made against her in online videos. VTP at 22. 6 In those videos, Woodward stated, V.C. talked about having a gun and was not afraid to use it. Id. Woodward reported that in her original contact with the parents, V.C. started the conversation with conflict: saying he had an attorney but then insisting on talking to Woodward and informing her that he would record their phone call. VTP at 14. When Woodward told V.C. she did not consent to the recording, he got “somewhat confrontational.” Id. Woodward offered to end the call, and at that point, V.C. decided against recording and they had a “fairly calm” conversation. VTP at 14, 15. Woodward also stated that she was unable to talk to C.C. directly because V.C. “pretty much controlled the conversation throughout. Mom was there, she did answer . . . some questions, but Dad was very forthcoming in reminding me several times that he had— power of attorney and that he was the head” of the “household.” VTP at 15-16. V.C. had “no problem verbalizing that he is the person in control in the relationship.” VTP at 40.

The testimony from the evidence-based parenting instructor, Logan Wright, illustrated V.C.’s aggressive behavior. In their first parenting session, V.C. became

6 V.C. does not challenge this testimony as hearsay on appeal. Assuming it comes under the standing hearsay objection made at trial, V.C.’s statements in the videos are admissible as a statement of a party-opponent. ER 801(d)(2).

24

No. 100966-6 Madsen, J., concurring/dissenting

frustrated when discussing the department’s involvement in A.C.’s case, but V.C. could be redirected. VTP at 70. At the second session, V.C.’s behavior escalated again, and Wright was unable to redirect him. VTP at 70-71. Importantly, Wright testified that V.C.’s behavior prevented him and C.C. from engaging in the session and prevented V.C. specifically from caring for A.C. VTP at 70-73. As a result, Wright asked V.C. to disengage from the sessions. VTP at 72. The dependency court relied on Wright’s testimony in its order, emphasizing that V.C.’s behavior escalated such that the focus was on him rather than on the child’s needs. See CP at 141 (“The needs of this child appear to be overshadowed by the parent’s actions.”).

In light of the substantial evidence of V.C.’s controlling and aggressive behavior, the hearsay testimony about his interactions with law enforcement did not have a material effect on the outcome of the dependency hearing. V.C.’s encounters with the police were certainly damaging, but V.C. provided equally damaging evidence on his own.

The dependency court was also concerned about V.C.’s mental health. According to Woodward, V.C. had told her he suffered from posttraumatic stress disorder (PTSD) and anxiety, for which he was not receiving treatment. VTP at 19. V.C. also told Woodward that he was a narcissist and would not work with the department because he did not believe he needed services. VTP at 23, 26. When asked by the State about his mental health diagnoses, V.C. stated he was not diagnosed with PTSD 7 and was not

7 It is unclear whether V.C. indeed suffers from PTSD. V.C. testified that a provider did not diagnose him with the condition, but the briefing on appeal contradicts that testimony. C.C. explains V.C.’s agitation during the second session with Logan Wright as “likely triggered by

25

No. 100966-6 Madsen, J., concurring/dissenting

bipolar; he never said he was a narcissist, but he admitted had “trouble acclimating to certain parameters.” VTP at 83. V.C. explained that he voluntarily sought out mental health treatment because people informed him that he is “argumentative” and “hostile.” VTP at 84. Tellingly, V.C. stated that he needed mental health treatment but also does not believe he needs it. Id.

The record supports the dependency court’s finding about V.C.’s mental health, and V.C. offers no persuasive argument that the finding was based on inadmissible hearsay or that any other error affected it.

3. Housing Instability

The unstable nature of the parents’ housing was also a basis for parental deficiency. The parents challenge this finding. They contend that their financial circumstances and even homelessness cannot deprive them of their fundamental right to care for their child. The parents note that they had sufficient financial and community resources to find housing and that they had been living in a motel during the dependency hearing, which the department had not inspected.

The parents are correct that a finding of dependency cannot be based solely on the economic circumstances of the family. See In re Welfare of Warren, 40 Wn.2d 342, 345, 243 P.2d 632 (1952) (“[p]overty of a parent does not of itself make the children dependent”). This is not a case, however, where dependency rested primarily on the

PTSD and feeling his rights were being violated by CPS.” (C.C.’s) Suppl. Br. at 5. V.C. also agrees he suffered from PTSD. See (V.C.’s) Suppl. Br. at 9 (“V.C. suffered from PTSD.”).

26

No. 100966-6 Madsen, J., concurring/dissenting

parents’ lack of adequate housing. The dependency court was concerned with the parents’ ability to care for A.C. on multiple grounds, including potential domestic violence, V.C.’s domineering and aggressive behavior that overshadowed his ability to care for A.C., the parents’ lack of current parental experience, and V.C.’s cannabis use. Housing instability was only one ground, among many.

And, the dependency court’s housing instability finding is amply supported. As previously discussed, Woodward and Barnes testified that V.C. had told them that the parents preferred to live in their car either because they thought it was a better way to live or in order to save money. V.C. also testified, disputing that he and C.C. had ever lived in their car and claimed instead they merely stayed in the car. There was also testimony that the department provided housing resources, but due to V.C.’s sex offender status those resources were unavailable to the family. The dependency court heard the conflicting testimony about the parents living in their car, as well as testimony from C.C. that they had stayed in three different motels since the shelter care hearing and from V.C. that they had financial resources to afford their current living situation. The court weighed that testimony, determined the credibility of the witnesses, and concluded that housing instability remained a concern to A.C. We do not reweigh evidence or reassess witness credibility on appeal. In re Welfare of X.T., 174 Wn. App. 733, 737, 300 P.3d 824 (2013).

In summary, after excluding the inadmissible evidence discussed above, the remaining admissible evidence meets the preponderance standard to find that A.C. was

27

No. 100966-6 Madsen, J., concurring/dissenting

dependent—there was no parent capable of adequately caring for the child, such that the child is in circumstances that constitute a danger of substantial damage to the child’s psychological or physical development. RCW 13.34.030(6)(c). Therefore, the appellants have failed to show that but for error, the outcome would have been different.

The parents’ history includes significant periods of incarceration, hindering their ability to raise their other children, and shows a lack of recent parental experience. C.C. had already relinquished her rights to her first child. V.C. had been convicted of multiple felonies, some of which were violent offenses. V.C.’s limitless view of power of attorney over C.C. caused significant concern and prevented the department from assessing the abilities of either parent.

Considering the substantial admissible evidence of V.C.’s controlling and aggressive behavior, the hearsay evidence about V.C.’s interactions with law enforcement and domestic violence reports from hospital staff and C.C.’s family members would not have changed the outcome had they not been admitted. Woodward explained that V.C. was confrontational with her from their first interaction, he had made threats about Woodward online, and V.C. told her that power of attorney meant he controlled everything C.C. did, including speaking for her. This was the basis of Woodward’s opinion that domestic violence existed in the relationship. Moreover, Wright testified that V.C.’s behavior escalated to the point that V.C. could not be redirected, interfering with C.C.’s engagement and precluding V.C. from parenting his child.

28

No. 100966-6 Madsen, J., concurring/dissenting

V.C.’s mental health also presented concerns. Though V.C. disagreed, Woodward and Barnes testified that V.C. had told them he was a narcissist, suffering from anxiety and PTSD for which he was not currently receiving treatment. V.C. was voluntarily engaging mental health services, but he stated puzzlingly that while he needed the services he also believed he did not need them. For C.C., the dependency court’s written order did not mention any developmental issue; Woodward stated that she had observed C.C. walking around and had no concerns about C.C.’s ability to ambulate; yet C.C. herself testified that she could not physically get up and down at the visitation center, indicating a physical challenge. The court’s finding that C.C. banged her head against the wall was based on inadmissible hearsay, but the court relied on other admissible instances of concern to find parental deficiency, such as lack of recent parental experience due to past CPS involvement and incarceration, as well as domestic violence in her relationship with V.C.

As to housing instability, the parents’ statements to the department witnesses indicated a history of using their car as temporary shelter. C.C. testified that she and V.C. had stayed at multiple hotels in the Spokane area since the shelter care hearing. Aside from the local bishop, who said he would assist the family at a distance, V.C. and C.C. had no meaningful community support outside of each other, which was made more difficult by V.C.’s status as a registered sex offender. While the parents testified that they planned to stay in Spokane long-term and had financial resources, the dependency

29

No. 100966-6 Madsen, J., concurring/dissenting

court weighed the evidence presented and found the department’s witnesses as more credible.

The remaining admissible evidence supports the dependency court’s findings about the parents’ history, V.C.’s controlling and aggressive behavior, and unstable housing. Accordingly, the appellants cannot show that but for the erroneously admitted evidence the outcome would have been different. Based on these considerations, the department proved by a preponderance of the evidence that A.C. was dependent. RCW 13.34.030(6)(c). There is no reasonable probability that had the erroneously admitted hearsay not been admitted that the outcome would have been different.

CONCLUSION

The majority provides half an answer to the question presented in this case. I agree that materially affected is the standard for nonconstitutional harmless error. I would hold that the same standard applies in criminal as well as civil cases. But more must be said. When articulated properly, the materially affected standard focuses the harmless error analysis on whether, without the error, in this case admitting hearsay, there was a reasonable probability that the outcome would have been different. This articulation properly focuses on the remaining, admissible evidence. Further, I would also add an evidentiary measure to the standard, in this case, whether, had the hearsay not been admitted, there is a reasonable probability that the court would not have found by a preponderance of the admissible evidence that A.C. was a dependent child. It is important that future courts have a workable method for determining whether an error has

30

No. 100966-6 Madsen, J., concurring/dissenting

a material effect after a thorough evaluation of the evidence. For the present case, I have reviewed the entire record, as is the duty of an appellate court, and I would hold that the error here did not materially affect the outcome—the department proved parental deficiency justifying a finding of dependency under RCW 13.34.030(6)(c). Accordingly, I respectfully concur in part and dissent in part.

_____________________________________

_____________________________________