[No. 60026-1. En Banc. September 2, 1993.]
HESS v. NORTH PACIFIC INS. CO.
TIMOTHY K. HESS, ET AL, Respondents, v. NORTH PACIFIC INSURANCE COMPANY, Petitioner.
 Insurance – Construction of Policy – Ambiguity – Considered as a Whole. In determining whether an insurance policy is ambiguous, a court reads related provisions together and in context.
 Insurance – Construction of Policy – Question of Law or Fact – In General. The interpretation of an insurance policy is a question of law for the court.
 Insurance – Construction of Policy – Policy as a Whole – Meaning to All Provisions – In General. All the provisions of an insurance policy are construed together so as to give force and effect to each clause.
 Insurance – Construction of Policy – Ambiguity – What Constitutes – Complex Interrelationships. The complexity of an insurance policy or the necessity to interrelate policy provisions does not, by itself, render the policy ambiguous.
 Insurance – Claim for Loss – Amount of Loss – Replacement Cost – Rebuilding – Necessity. A replacement cost insurance policy provision limiting the insurer's payment of proceeds to the actual cash value of the damage unless actual repair or replacement is complete is not ambiguous and limits coverage to actual cash value if the insured chooses not to repair or replace the structure.
GUY, J., did not participate in the disposition of this case.
Nature of Action: After receiving insurance proceeds equal to the actual cash value of a cabin destroyed by fire, the owners claimed that they were entitled to the full replacement cost of the cabin even though they did not intend to replace it.
Superior Court: The Superior Court for Spokane County, No. 90-2-00883-0, William J. Grant, J., on April 24, 1991, entered a summary judgment in favor of the owners.
Court of Appeals: The court at 67 Wn. App. 783 affirmed the judgment, holding that the policy was ambiguous and had to be interpreted in favor of the owners.
Supreme Court: Holding that the policy unambiguously limits coverage to actual cash value if the insured chooses not to replace the structure, the court reverses the decision of the Court of Appeals and the judgment and grants judgment in favor of the insurer.
Helsell, Fetterman, Martin, Todd & Hokanson, by William A. Helsell and Robert N. Gellatly, for petitioner.
Fred C. Pflanz, Beverly L. Anderson, Carl E. Hueber, and Winston & Cashatt, for respondents.
William R. Hickman and Sandra K. Pailca on behalf of United Services Automobile Association, amicus curiae for petitioner.
Sidney R. Snyder, Jr., and Ronald S. Dinning on behalf of PEMCO, Safeco, and Unigard Insurance Companies, amici curiae for petitioner.
J. Tucker Miller on behalf of Safeco Insurance Company, amicus curiae for petitioner.
Mark S. Cole, Mark Edward Mills, and Thomas Lether on behalf of Farmers Insurance Group and American States Insurance Company, amici curiae for petitioner.
BRACHTENBACH, J. – This case concerns the amount payable under the replacement clause of a homeowners insurance policy when the destroyed building is not repaired or replaced and the insured has no intent to repair or replace. This is the first occasion for this court to interpret such a replacement clause.
The facts are stipulated. Defendant, North Pacific Insurance Company, insured a summer cabin of plaintiffs Timothy K. and Georgianne H. Hess. The cabin was destroyed by fire. The agreed actual cash value was $20,000. The agreed replacement cost was $43,182.10. The insureds did not replace the cabin, nor do they intend to do so. Defendant paid the actual cash value to plaintiffs. Clerk's Papers, at 114-15. Plaintiffs sued for $23,182.10, the difference between the full replacement cost and the actual cash value.
The sole issue is whether, under the terms of the policy, the insureds can collect the full replacement cost when they have not replaced the destroyed insured cabin and stipulated they do not intend to replace it.
The trial court granted summary judgment for said $23,182.10 to the insured plaintiffs, plus prejudgment interest. The Court of Appeals affirmed. Hess v. North Pac. Ins. Co., 67 Wn. App. 783, 841 P.2d 767 (1992), review granted, 121 Wn.2d 1008 (1993). We reverse, and thereby join the virtually unanimous holdings in other jurisdictions which have considered the same or similar replacement clauses.
Before analyzing the policy clauses, a brief history of replacement clauses is helpful. Historically, the underlying purpose of property insurance is indemnity. Traditional coverage was for the actual or fair cash value of the property. The owner was indemnified fully by payment of the fair cash value, in effect the market value, which is what the owner lost if the insured building was destroyed. 6 J. & J. Appleman, Insurance § 3823 (1972).
However, it was recognized that an owner might not be made whole because of the increased cost to repair or to rebuild. Thus, replacement cost coverage became available. "Replacement cost coverages . . . go beyond the concept of indemnity and simply recognize that even expected deterioration of property is a risk which may be insured against." Jordan, What Price Rebuilding?, 19 The Brief 17 (Spring 1990) (cited hereafter as the Jordan report).
A Washington statute prohibits "overinsurance", i.e., insurance in excess of the "fair value" (defined as cost of replacement less depreciation), RCW 48.27.010. However, replacement insurance is authorized specifically by RCW 48.27.020:
[T]he insurer may in connection with a special provision or endorsement made a part of the policy insure the cost of repair or replacement of such property, if damaged or destroyed by a hazard insured against, and without deduction of depreciation . . ..
In this case, the relevant provisions of plaintiffs' policy are as follows:
3. Loss Settlement. Covered property losses are settled as follows:
a. (1)-(3) [relate to property not involved here].
b. Buildings under Coverage A or B at replacement cost without deduction for depreciation, subject to the following:
(1) If, at the time of loss, the amount of insurance in this policy on the damaged building is 80% or more of the full replacement cost of the building immediately before the loss, we will pay the cost to repair or replace, after application of deductible and without deduction for depreciation, but not more than the least of the following amounts:
(a) the limit of liability under this policy that applies to the building;
(b) the replacement cost of that part of the building damaged for like construction and use on the same premises; or
(c) the necessary amount actually spent to repair or replace the damaged building.
[Subparagraphs (2) and (3) relate to determination of the 80 percent of full replacement coverage and are not relevant here.]
(4) We will pay no more than the actual cash value of the damage unless:
(a) actual repair or replacement is complete; or
(b) the cost to repair or replace the damage is both:
(i) less than 5% of the amount of insurance in this policy on the building; and
(ii) less than $1000.
(5) You may disregard the replacement cost loss settlement provisions and make claim under this policy for loss or damage to buildings on an actual cash value basis. You may then make claim within 180 days after loss for any additional liability on a replacement cost basis.
Clerk's Papers, at 59-60.
 A careful examination of these clauses, read together and in context as we must do, does not reveal any ambiguity. Stated generally, subparagraphs 3.b.(1)(a), (b), and (c) set the limits of maximum liability, i.e., the lesser of (a) or (b) or (c). Those amounts reflect (a) the policy limits, (b) the replacement cost of like construction and use on the same premises, more fully explained hereafter, or (c) the amount actually spent to repair or replace the damaged building.
The Jordan report, cited above, cogently explains:
The first measure, of course, limits the amount available for replacement to policy limits, while the second relates to a theoretical or hypothetical measure of loss: that is, the replacement cost of rebuilding the identical structure as one limit of the company's liability. This particular limitation does not require repair or replacement of an identical building on the same premises, but places that rebuilding amount as one of the measures of damage to apply in calculating liability under the replacement cost coverage. The effect of this limitation comes into play when the insured desires to rebuild either a different structure or on different premises. In those instances, the company's liability is not to exceed what it would have cost to replace an identical structure to the one lost on the same premises. Although liability is limited to rebuilding costs on the same site, the insured may then take that amount and build a structure on another site, or use the proceeds to buy an existing structure as the replacement, but paying any additional amount from his or her own funds.
Finally, the third limitation of liability strengthens the requirement that liability of the company does not exist until repair or replacement is made. The purpose of this limitation is to limit recovery to the amount the insured spent on repair or replacement as yet another measure of the loss liability of the insurer. This third valuation method is intended to disallow an insured from recovering, in replacement cost proceeds, any amount other than that actually expended.
(Footnotes omitted.) Jordan, at 19-20.
The Court of Appeals somehow concluded that the insurer's interpretation, i.e., pay actual cash value unless replaced or repaired, "clearly implies that an insured who elects not to rebuild is entitled to no settlement at all." Hess v. North Pac. Ins. Co., 67 Wn. App. 783, 787, 841 P.2d 767 (1992), review granted, 121 Wn.2d 1008 (1993). The Court of Appeals found that to be an ambiguity when related to subparagraph 3.b.(4): "We will pay no more than the actual cash value of the damage unless: (a) actual repair or replacement is complete". Hess, at 787.
It appears quite clear that despite the measures of possible liability set forth in 3.b.(1)(a), (b), and (c), subparagraph (4)(a) conditions payment of any one of those amounts upon completion of "actual repair or replacement". (Italics ours.) The insurer has never contended that it does not owe, at a minimum, the actual cash value of the destroyed insured building, and, indeed, promptly paid that amount after the parties agreed that such in fact was the actual cash value.
The insurer well answers the Court of Appeals reasoning in this manner:
Paragraph 3.b.(1) deals with alternative measurers of replacement cost. When an insured does not replace, the least of the three alternative measures of loss (amount actually spent) is zero. Thus, the insured is not entitled to replacement cost. He is, however, entitled to actual cash value under paragraph 3.b.(4).
Supplemental Brief of Appellant, at 3-4.
The Jordan report summarizes the purpose of this clause:
This requirement for actual repair or replacement by the insured does not affect the company's liability to pay for actual cash value loss, but only for the difference between that figure and the higher replacement cost. The purpose of that limitation, obviously, is to prevent an insured from directly profiting through the receipt of cash funds beyond the actual cash value of the loss, thus forcing the insured to rebuild in order to recover amounts withheld as depreciation.
Jordan, at 19.
The Court of Appeals seemed to find some ambiguity because the policy does not define "replacement cost" or "actual cash value". However, the policy does provide an appraisal method if the parties do not agree on the amount of the loss (Clerk's Papers, at 60). In any event, the issue seems irrelevant here because the parties stipulated to both the replacement cost and actual cash value.
[2-4] We turn to general principles of interpretation of insurance policies.
The interpretation of insurance policies is a question of law. In construing the language of an insurance policy, the entire contract must be construed together so as to give force and effect to each clause. If the language in an insurance contract is clear and unambiguous, the court must enforce it as written and may not modify the contract or create ambiguity where none exists.
(Citations omitted.) Transcontinental Ins. Co. v. Washington Pub. Utils. Dists.' Util. Sys., 111 Wn.2d 452, 456, 760 P.2d 337 (1988). That opinion goes on to summarize the applicable rules of interpretation if the provisions of a policy are ambiguous. Transcontinental, at 456-57. Complexity or the necessity to interrelate policy provisions does not alone render a policy ambiguous. McDonald v. State Farm Fire & Cas. Co., 119 Wn.2d 724, 734, 837 P.2d 1000 (1992).
 Applying the quoted provisions of Transcontinental, we must give effect to subparagraph 3.b.(4): "We will pay no more than the actual cash value of the damage unless: (a) actual repair or replacement is complete". (Italics ours.) Clerk's Papers, at 60. When the definitions of the limits of liability which immediately precede the clearly stated condition and limitation of 3.b.(4) are read together and each clause given meaning, it is evident that only the actual cash value is owed unless actual repair or replacement is undertaken and completed. It should be noted that paragraph 3.b., which refers to replacement cost, contains the phrase "subject to the following". One of the following subparagraphs is 3.b.(4). In other words, the very paragraph upon which the insureds rely makes itself subject to the limitation of 3.b.(4).
The following subparagraph, 3.b.(5), buttresses this conclusion by providing the insured "may disregard the replacement cost loss settlement provisions and make claim . . . for loss or damage to buildings on an actual cash value basis." Clerk's Papers, at 60. Obviously, rebuilding or replacement recovery remains an option for the insured because that subparagraph goes on to provide: "You may then make claim within 180 days after loss for any additional liability on a replacement cost basis." Clerk's Papers, at 60. This optional method of loss settlement would be superfluous if an insured were entitled to replacement cost without making the replacement.
Our holding is consistent with the holdings of other courts. In a recent annotation, the rule is summarized: "Generally, actual replacement of damaged or destroyed property has been held to be a prerequisite to collection of proceeds under a replacement cost endorsement of an insurance policy . . .." Annot., Construction and Effect of Property Insurance Provision Permitting Recovery of Replacement Cost of Property, 1 A.L.R.5th 817, 829 (1992). The annotation cites only two cases to the contrary which will be discussed later.
A recognized text makes the following comment: "Certainly, considering the purpose of [the replacement cost provisions], it is reasonable to deny recovery for replacement costs where the insured is not going to replace the property as he would then make a profit out of his loss . . .." 6 J. & J. Appleman, Insurance § 3823 n.66.57 (Supp. 1992).
The cases reach the same result. In Huggins v. Hanover Ins. Co., 423 So. 2d 147 (Ala. 1983), the relevant policy clauses were virtually identical to those in this case, including a provision that "we will pay no more than the actual cash value of the damage until actual repair or replacement is completed." Huggins, at 149. The court stated:
Provisions like those contained in subparagraph c.(4) [quoted immediately above] have been interpreted as providing a condition precedent to an insurer's duty to pay repair or replacement costs of an insured building. A party who has not repaired or replaced his insured building has not complied with the condition precedent to recovery under the policy and so cannot recover. See Kolls v. Aetna Casualty & Surety Co., 503 F.2d 569 (8th Cir.1974); Bourazak v. North River Insurance Company, 379 F.2d 530 (7th Cir.1967).
Huggins, at 150. Accord, Hilley v. Allstate Ins. Co., 562 So. 2d 184, 1 A.L.R.5th 1167 (Ala. 1990).
The replacement cost clause in Paluszek v. Safeco Ins. Co. of Am., 164 Ill. App. 3d 511, 515-16, 517 N.E.2d 565 (1987) contained this language: "'[T]his company shall not be liable for more than the actual cash value of the damaged property unless and until actual repair or replacement is completed.'" The court held:
Unless and until an actual loss is sustained and proved by the insured, he is not entitled to reimbursement by the insurer.
In the present case, defendant paid plaintiff the actual cash value of her house . . .. Under the terms of the policy, this was all defendant was obligated to pay unless and until repairs were completed.
Paluszek, at 516.
Snellen v. State Farm Fire & Cas. Co., 675 F. Supp. 1064 (W.D. Ky. 1987) is in point. The three amounts of possible recovery were essentially identical to those in this case. A following paragraph provided: "We will pay the cash value of the damage, up to the policy limit, until actual repair or replacement is completed." Snellen, at 1066. The insured did not replace the home and evidenced no intent to do so, but she claimed the policy was ambiguous so she should recover the replacement cost. The court held:
The policy in question is not ambiguous with respect to the insurer's obligation to pay replacement cost. Actual replacement, and the incurrence of costs in that endeavor, is clearly a condition precedent to such a claim.
. . . .
To hold otherwise in this case would necessitate ignoring the plain terms of the policy. It would also be contrary to the obvious provisions of the loss settlement clause taken as a whole, and would result in enlarging the coverage of the policy beyond its natural and obvious meaning.
(Citations omitted.) Snellen, at 1067.
The policy in Higgins v. Insurance Co. of North Am., 256 Or. 151, 469 P.2d 766, 66 A.L.R.3d 871 (1970) did not contain a specific limitation, as in this case, i.e., that the company would only pay the actual cash value until repair or replacement was completed. However, after carefully tracing the history and purpose of replacement coverage, the court held:
We conclude that since plaintiffs have not expended anything in repairing or replacing the insured building they are not eligible to recover under the "Replacement Cost" extension of the policy. For a similar conclusion see Bourazak v. North River Insurance Company, 379 F2d 530 (7th Cir. 1967).
(Footnote omitted.) Higgins, at 166-67.
Plaintiffs-insureds dismiss this very substantial line of authorities with the unhelpful statement that "[t]he cases from other jurisdictions have no bearing on the interpretation of this ambiguous policy." Brief of Respondent, at 9.
The annotation cited above cites only two cases in the United States which purportedly hold replacement is not a prerequisite to collection of replacement proceeds. Those two cases are Reese v. Northern Ins. Co., 207 Pa. Super. 19, 215 A.2d 266 (1965) and National Fire Ins. Co. v. Solomon, 96 Wn.2d 763, 638 P.2d 1259 (1982). Because of a broad statement by Justice Dore in National Fire Ins. Co. v. Solomon, supra, examination of that opinion is necessary. That statement is:
The replacement cost method of payment does not require the rebuilding of the structure as a condition precedent to the payment of the proceeds under such policy. We rely on the rationale of Reese v. Northern Ins. Co. of N.Y., 207 Pa. Super. Ct. 19, 22, 215 A.2d 266 (1965).
Solomon, at 770.
Before examining the policy in Solomon to determine its relevance here, it is necessary to reevaluate the reliance on Reese, the Pennsylvania Superior Court case which is the only authority cited by Justice Dore as author of the Solomon opinion. Reese has been cited by only one other court, outside of Pennsylvania. In that case, Higgins v. Insurance Co. of North Am., supra, the Oregon Supreme Court, in reaching a contrary holding, simply said: "We think [Reese] . . . was wrongly decided . . .." Higgins, at 167 n.5. In the 28 years since Reese was decided, it has never been cited by the Pennsylvania Supreme Court. It has been cited only three times in superior court cases and only in passing. In one of those cases the trial court had held that the insurance policy provision providing for payment of only the cash value until replacement was completed was consistent with Pennsylvania statutes and judicial interpretations. Obviously such holding is directly contrary to Reese. On appeal, the Superior Court held only that the trial court ruling was but a partial and interlocutory resolution of the issues, and thus not appealable. Canulli v. Allstate Ins. Co., 315 Pa. Super. 460, 462 A.2d 286 (1983). In another case, Reese was cited solely for the definition of "replacement costs". Ditch v. Yorktowne Mut. Ins. Co., 343 Pa. Super. 22, 493 A.2d 782 (1985). In the third case, Reese was cited solely for the general rule of construction that ambiguous policy language is to be construed against the insurer and in favor of the insured and coverage. Slate Constr. Co. v. Bituminous Cas. Corp., 228 Pa. Super. 1, 5, 323 A.2d 141 (1974).
Careful reading of the Reese opinion reveals it does not support the broad rule for which it is cited in Solomon. The policy in Reese contained the same three measures of loss as in this case, but there was no clause providing for the payment of cash value only until repair or replacement was completed. Even though the company had paid the cash value, it argued that since the insured had not repaired or rebuilt the lesser amount under the replacement clause was zero. The court noted that under such interpretation, carried to its logical conclusion, the insured would not even be entitled to the cash value, already paid. The court distinguished a New Jersey case where the policy required actual replacement before there was any sum due. Significantly, the court said:
That clause [in the New Jersey case] expressly requires the insured to replace and [then] he is entitled to recover replacement value. Had such a clause been included in the present policy, there would be some justification for the defendant's contention but such was not the case.
(Italics ours.) Reese v. Northern Ins. Co., supra at 24-25. Without further analysis, the court held that the insured was entitled to the replacement cost without rebuilding.
Analysis of the 2-1/2-page Reese opinion and its subsequent history casts substantial doubt upon the validity of the statement in National Fire Ins. Co. v. Solomon, supra at 770, that: "The replacement cost method of payment does not require the rebuilding of the structure as a condition precedent to the payment of the proceeds under such policy." We disapprove of the statement as a general rule of interpretation of insurance policies. The first thing wrong with it is that it ignores the possibility of different provisions which may exist in other policies, as in this case. Second, the statement ignores all the substantial, well-reasoned authorities to the contrary. Third, the Reese case is not persuasive authority, but is the only authority cited.
The facts in National Fire Ins. Co. v. Solomon, 96 Wn.2d 763, 638 P.2d 1259 (1982) mandate limiting whatever its holdings may be to those facts and the policy involved. The case arose from the trial court's denial of the insurer's motion to enforce the appraisal provisions in the policy. Yet the court considered the merits. Relying on a California statute, the court held that actual cash value meant fair market value without depreciation. Another state's statutory definition should not control our interpretation.
Most significantly, Solomon simply does not apply or control in this case because the policy here specifically provides that the insurer will pay no more than actual cash value unless actual repair or replacement is complete. This clause alone completely distinguishes Solomon.
We hold that the policy at issue is not ambiguous; its conditions plainly limit recovery to actual cash value under these facts.
The Court of Appeals is reversed. The matter is remanded to the trial court with directions to enter summary judgment in favor of the defendant.
ANDERSEN, C.J., and UTTER, DOLLIVER, DURHAM, SMITH, JOHNSON, and MADSEN, JJ., concur.[Nos. 58492-3, 58733-7, 58734-5. En Banc. September 9, 1993.]
THE STATE OF WASHINGTON, Appellant, v. DAVID K. TALLEY, Respondent.
THE STATE OF WASHINGTON, Appellant, v. DANIEL MYERS, Respondent.
THE STATE OF WASHINGTON, Appellant, v. BRANDON STEVENS, Respondent.
 Constitutional Law – Police Power – Validity – Effect on Individual Rights. The State's police power to restrain harmful conduct is limited by the constitutional protection afforded certain personal liberties.
 Constitutional Law – Freedom of Speech – Content Restriction – Validity – In General. Generally, speech may not be regulated on the basis of its content.
 Malicious Harassment – Validity – Intimidating or Harassing Conduct – Freedom of Speech – Effect on Speech. RCW 9A.36.080(1), which defines the crime of malicious harassment as engaging in certain conduct with an intent to intimidate or harass the victim based on the victim's perceived membership in a protected category, regulates harmful conduct and only incidentally affects speech. It is the perpetrator's selection of the victim, rather than the perpetrator's thoughts, beliefs, or motives underlying the criminal act, that results in increased punishment.
 Malicious Harassment – Validity – Intimidating or Harassing Conduct – Freedom of Speech – Victim Selection. RCW 9A.36.080(1), which criminalizes a defendant's discriminatory conduct in selecting a victim for a crime, does not violate the First Amendment freedom of speech guaranty.
 Constitutional Law – Freedom of Speech – Content Restriction – Validity – Test. The regulation of proscribable speech based on its content is permissible under the First Amendment if: (1) the basis for the content discrimination consists of the very reason the entire class of speech is proscribable; (2) the content is not the object of the regulation so that any content discrimination is inadvertent; or (3) the content discrimination presents virtually no danger of eroding First Amendment rights.
 Criminal Law – Punishment – Sentence – Enhanced Penalty – Status of Victim. Criminal conduct may be punished more severely where it is directed at individuals because of their perceived membership in a protected category.
 Malicious Harassment – Proof – Racial Intolerance – First Amendment – Effect. Evidence of racial intolerance, relevant to the crime of malicious harassment, is not inadmissible merely because the person's beliefs and associations are protected by the First Amendment.
 Criminal Law – Statutes – Overbreadth – Speech – In General. A criminal statute is unconstitutionally overbroad if it prohibits a substantial amount of constitutionally protected speech within its legitimate scope and there is no way to sever the statute's unconstitutional applications.
 Malicious Harassment – Validity – Intimidating or Harassing Conduct – Overbreadth – In General. RCW 9A.36.080(1), which defines the crime of malicious harassment as engaging in certain conduct with an intent to intimidate or harass the victim because of the victim's perceived membership in a protected category, is not unconstitutionally overbroad.
 Malicious Harassment – Validity – Intimidating or Harassing Conduct – Overbreadth – Speech as Evidence of Victim Selection. In a prosecution for malicious harassment, the fact that the defendant's speech may be used to prove victim selection does not render the statute unconstitutionally overbroad and violative of the First Amendment.
 Constitutional Law – Construction – State and Federal Provisions – Independent State Interpretation – Argument – Necessity. A court will not consider whether the state constitution provides greater protection than the federal constitution absent briefing of any of the six nonexclusive factors set forth in State v. Gunwall, 106 Wn.2d 54.
 Criminal Law – Statutes – Vagueness – Test. A statutory definition of a crime is not unconstitutionally vague if it provides both adequate notice and adequate standards to prevent arbitrary enforcement.
 Statutes – Validity – Burden and Degree of Proof – In General. A party challenging the validity of a statute must prove its unconstitutionality beyond a reasonable doubt.
 Statutes – Construction – Meaning of Words – Absence of Statutory Definition – In General. Terms that are not defined in a statute will be given their plain and ordinary meaning.
 Statutes – Construction – Considered as a Whole – In General. Courts interpret statutes as a whole.
 Statutes – Construction – Meaning of Words – Ejusdem Generis. Under the doctrine of ejusdem generis, a court will give a statutory term in a sequence a meaning consistent with the other words in the sequence.
 Malicious Harassment – Validity – Intimidating or Harassing Conduct – Vagueness. RCW 9A.36.080(1), which defines the crime of malicious harassment as engaging in certain conduct with an intent to intimidate or harass the victim based on the victim's perceived membership in a protected category, is not unconstitutionally vague.
 Criminal Law – Crimes – Discretion To Charge – Equal Protection – Different Elements. When a criminal defendant's conduct violates more than one criminal statute and where each statute redresses substantially different harms and has different elements, the prosecutor's discretion to choose which statute to prosecute under does not violate the defendant's right to equal protection of the laws.
 Malicious Harassment – Validity – Intimidating or Harassing Conduct – Equal Protection. RCW 9A.36.080(1), which defines the crime of malicious harassment as engaging in certain conduct with an intent to intimidate or harass the victim based on the victim's perceived membership in a protected category, does not violate equal protection.
 Statutes – Construction – Meaning of Words – Different Words. The use of different language within the same statute gives rise to a presumption that the Legislature intended to address different concerns.
 Constitutional Law – Freedom of Speech – Hate Speech. Symbolic hate speech, while offensive, is protected by the First Amendment.
 Malicious Harassment – Validity – Per Se Violations – Content-Based Regulation of Protected Speech. RCW 9A.36.080(2), which establishes per se violations of the crime of malicious harassment, regulates protected symbolic speech based on its content and is unconstitutional.
DOLLIVER, J., did not participate in the disposition of this case.
Nature of Action: Defendant Talley was charged with six counts of malicious harassment. In separate prosecutions, defendants Myers and Stevens were each charged with one count of malicious harassment. All the charges stemmed from cross burnings.
Superior Court: The Superior Court for King County, No. 91-1-02001-5, Patricia H. Aitken, J., on October 8, 1991, dismissed the charges against defendant Talley. The Superior Court for King County, Nos. 91-8-02436-9, 91-8-02530-6, Marsha J. Pechman, J., on October 9, 1991, dismissed the charges against defendants Myers and Stevens.
Supreme Court: Holding that former RCW 9A.36.080(1) regulates conduct and only incidentally affects speech, is neither unconstitutionally overbroad nor vague, and does not violate equal protection, but that former RCW 9A.36.080(2) unconstitutionally regulates protected symbolic speech based on content, the court affirms the dismissal of the charges against defendant Talley and reverses the dismissal of the charges against defendants Myers and Stevens.
Norm Maleng, Prosecuting Attorney, and Donna L. Wise, Senior Appellate Attorney, for appellant.
Browne & Ressler and Allen M. Ressler, for respondent Talley.
Debra K. Hannula and Ken Scearce of Associated Counsel for the Accused, for respondent Myers.
Ronald R. Piper, for respondent Stevens.
Phillip H. Ginsberg (Tzivia Schwartz, Ruth Lansner, and Steven M. Freeman, of counsel) on behalf of the Anti-Defamation League, amicus curiae for appellant.
Paul J. Lawrence and Adam W. Gravley on behalf of the American Civil Liberties Union of Washington, amicus curiae.
MADSEN, J. – Nearly every state has passed what has come to be termed a "hate crimes statute". See Hate Crimes Stat-utes: A Status Report, Anti-Defamation League Legal Affairs Department, Civil Rights Division (Washington, D.C. 1991). Some of these statutes enhance penalties where crimes are bias motivated or target certain victims. Other statutes punish various forms of hate speech, such as cross burning. Hate Crimes, at 1-2. The primary inquiry in this case is whether Washington's hate crimes statute, malicious harassment, RCW 9A.36.080, is an unconstitutional infringement of First Amendment rights.
State v. Talley
In April of 1991, Phillip and Renee Smith, a mixed race couple, their 3-year-old son, their friend Jeanne Tyler, her son, and their realtor Stacy Litzenberger, visited the house they were purchasing in Maple Valley in King County. The Smiths had recently signed an earnest money agreement, and they were visiting the house to measure it for carpets and drapes.
David Talley, who lived next door, saw the Smiths and complained to bystanders that "having niggers next door" would ruin his property values. He built a 4-foot-tall cross, planted it in his own yard, set it on fire with gasoline, and began to "hoot and holler". These actions attracted the attention of the Smith party. They noticed that Talley was wearing fatigues, combat boots, a Harley-Davidson T-shirt, and that he had a clean-shaven head. Renee Smith approached Talley and asked if there was a "problem". He replied that he did not talk with her "kind" and ordered her off his driveway. The Smiths were disturbed and frightened by the incident and opted not to purchase the house.
The State charged Talley with six counts of malicious harassment. Talley argued that the statute was unconstitutional because it violated his right to free speech. King County Superior Court Judge Patricia Aitken held that subsection (2)(a) of the malicious harassment statute was unconstitutionally overbroad because it infringed on First Amendment guaranties of free speech. She reasoned that although subsection (1) could be construed as constitutional, Talley's conduct did not fit within the statute under that construction. Judge Aitken granted Talley's motion to dismiss.
State v. Myers and Stevens
In April 1991, Daniel Myers, Brandon Stevens, and several other teenagers were at a party at one of the teenagers' homes. Conversation at the party turned to Chris Elion, an African-American classmate of the boys. The group decided to burn a cross in the Elion family's yard because some of the boys felt that Chris had been acting "too cool at school".
The teens fabricated a cross, which they planted in the Elions' front yard, and attempted to set it on fire. Mr. and Mrs. Elion had just returned from church when Mrs. Elion noticed the group of teenagers outside. She thought they were friends of her son so she flashed the porch lights on and off, opened the door and called out. The group fled. Mrs. Elion saw the cross and screamed to her husband. Mr. Elion ran outside and found the smoldering 8-foot cross. Only the base of the cross had burned, singeing some nearby shrubbery.
The State charged Myers and Stevens each with one count of malicious harassment. King County Superior Court Judge Marsha J. Pechman concurred with Judge Aitken regarding subsection (2)(a), but held that subsection (1)(b) was unconstitutionally overbroad because its vague language carried it into "constitutionally treacherous waters".
The State appealed to this court and the three cases were joined to determine the constitutionality of Washington's malicious harassment statute.
RCW 9A.36.080 states:
(1) A person is guilty of malicious harassment if he maliciously and with the intent to intimidate or harass another person because of, or in a way that is reasonably related to, associated with, or directed toward, that person's race, color, religion, ancestry, national origin, or mental, physical, or sensory handicap:
(a) Causes physical injury to another person; or
(b) By words or conduct places another person in reasonable fear of harm to his person or property or harm to the person or property of a third person. Such words or conduct include, but are not limited to, (i) cross burning, (ii) painting, drawing, or depicting symbols or words on the property of the victim when the symbols or words historically or traditionally connote hatred or threats toward the victim, or (iii) written or oral communication designed to intimidate or harass because of, or in a way that is reasonably related to, associated with, or directed toward, that person's race, color, religion, ancestry, national origin, or mental, physical, or sensory handicap. However, it does not constitute malicious harassment for a person to speak or act in a critical, insulting, or deprecatory way unless the context or circumstances surrounding the words or conduct places another person in reasonable fear of harm to his or her person or property or harm to the person or property of a third person; or
(c) Causes physical damage to or destruction of the property of another person.
(2) The following constitute per se violations of this section:
(a) Cross burning; or
(b) Defacement of the property of the victim or a third person with symbols or words when the symbols or words historically or traditionally connote hatred or threats toward the victim.
(3) Malicious harassment is a class C felony.
RCW 9A.36.080. The Legislature attached a severability clause in 1989. Laws of 1989, ch. 95, § 4.
In this appeal, the State of Washington argues that RCW 9A.36.080 regulates criminal conduct and words tantamount to criminal conduct rather than speech. The State argues further that if the statute does affect speech then it only reaches unprotected fighting words. The three respondents, Talley, Myers, and Stevens, contend that RCW 9A.36.080 is unconstitutional because it directly regulates the communicative impact of speech. They also argue that the statute is overbroad because protected free speech falls within the statute's legitimate scope and because it chills free speech. Finally, respondent Stevens argues that RCW 9A.36.080(1) violates equal protection and is impermissibly vague.
We find that RCW 9A.36.080 subsection (1) withstands constitutional scrutiny because it is aimed at criminal conduct and only incidentally affects speech. Conversely, subsection (2) is overbroad because it inhibits free speech on the basis of its content. Since our analysis brings us to differing conclusions, we address RCW 9A.36.080(1) and (2) separately.
 States have a legitimate interest in restraining harmful conduct and are empowered to do so under their police powers. Seattle v. Hill, 72 Wn.2d 786, 797, 435 P.2d 692 (1967), cert. denied, 393 U.S. 872 (1968); Minnesota ex rel. Whipple v. Martinson, 256 U.S. 41, 45, 65 L. Ed. 819, 41 S. Ct. 425 (1921). A state's police power is limited, however, by constitutional protection afforded certain personal liberties. Olympic Forest Prods., Inc. v. Chaussee Corp., 82 Wn.2d 418, 435, 511 P.2d 1002 (1973). Boddie v. Connecticut, 401 U.S. 371, 379, 28 L. Ed. 2d 113, 91 S. Ct. 780 (1971). In this case, the personal liberty allegedly implicated is freedom of speech. The threshold question, therefore, is whether RCW 9A.36.080 regulates speech or conduct.
CONDUCT VERSUS SPEECH REGULATION
Relying on the Supreme Court's recent decision in R.A.V. v. St. Paul, Minn., ___ U.S. ___, 120 L. Ed. 2d 305, 112 S. Ct. 2538 (1992), respondents argue that Washington's malicious harassment statute is a content-based speech regulation and, thus, an unconstitutional restriction on First Amendment rights. We do not agree that R.A.V. is controlling on the issue presented by RCW 9A.36.080(1) which we conclude regulates conduct, not speech.
 Unlike RCW 9A.36.080(1), the statute considered in R.A.V. was, on its face, a content-based speech regulation.
Whoever places on public or private property a symbol, object, appellation, characterization or graffiti, including, but not limited to, a burning cross or Nazi swastika, which one knows or has reasonable grounds to know arouses anger, alarm or resentment in others on the basis of race, color, creed, religion or gender commits disorderly conduct and shall be guilty of a misdemeanor.
St. Paul, Minn. Legis. Code § 292.02 (1990); R.A.V., 112 S. Ct. at 2541. The Minnesota Supreme Court upheld the statute by construing it to apply only to unprotected fighting words which it held were beyond the realm of constitutional scrutiny. In re Welfare of R.A.V., 464 N.W.2d 507 (Minn. 1991) (citing Chaplinsky v. New Hampshire, 315 U.S. 568, 571-72, 86 L. Ed. 1031, 62 S. Ct. 766, 769 (1942)). While accepting the Minnesota court's construction, Justice Scalia, writing for the majority, rejected the notion that fighting words are exempt from constitutional scrutiny. The majority opinion held that, even in the area of proscribable speech with low First Amendment value, First Amendment doctrine prohibits speech regulation on the basis of its content. Scalia called this the "'content discrimination' limitation." R.A.V., 112 S. Ct. at 2545.
This holding is a logical extension of the rationale in flag burning cases. E.g., Texas v. Johnson, 491 U.S. 397, 405-06, 105 L. Ed. 2d 342, 109 S. Ct. 2533, 2540 (1989). Under this rationale, a statute prohibiting speech based on offensive content cannot survive First Amendment challenge. The St. Paul ordinance was aimed at doing precisely this. St. Paul admittedly enacted its ordinance to inhibit certain types of symbolic hate speech. R.A.V., 112 S. Ct. at 2549. «1»
«1» The defendant in R.A.V. was also convicted under the Minnesota equivalent of RCW 9A.36.080(1) which enhances punishment for crimes where the criminal selects the victim based on the victim's membership in a protected class. See generally Minnesota Criminal Code, Minn. Stat. ch. 609 (1992). The defendant, however, only challenged his conviction under the St. Paul ordinance because it regulated pure symbolic speech protected by the First Amendment. R.A.V., 112 S. Ct. at 2541 n.2.
 Unlike the St. Paul ordinance considered in R.A.V., RCW 9A.36.080(1) does not, on its face, regulate speech. Rather, the statute is aimed at criminal conduct and enhances punishment for that conduct where the defendant chooses his or her victim because of their perceived membership in a protected category. The statute punishes the selection of the victim, not the reason for the selection. It increases punishment where the perpetrator acts on particularly offensive beliefs, not the beliefs themselves. The statute is triggered by victim selection regardless of the actor's motives or beliefs.
Absent prohibited victim selection, the conduct described in subsections (1)(a), (b), and (c) is punishable elsewhere in state law and in some municipal criminal codes as misdemeanor violations. However, when the victim is targeted because of perceived membership in one of the enumerated categories, then the criminal conduct is punishable as a felony. RCW 9A.36.080(3). Under subsection (1)(a), simple assault is punished as a felony when the State proves the additional element of unlawful victim selection. See RCW 9A.36.041 (assault in the fourth degree). In subsection (1)(b), proof of harassment or assault, with victim selection, constitutes malicious harassment. Guffey v. State, 103 Wn.2d 144, 690 P.2d 1163 (1984) (assault is committed by putting another in apprehension of harm); Seattle v. Huff, 111 Wn.2d 923, 925, 767 P.2d 572 (1989) (crime of telephone harassment is defined as threatening physical injury or property damage); Seattle v. Camby, 104 Wn.2d 49, 701 P.2d 499 (1985) (City must prove substantial risk of assault as an element of the crime of harassment); State v. Smith, 48 Wn. App. 33, 737 P.2d 723 (1987) (harassment proscribes only threats of bodily injury and physical confinement or restraint), rev'd on other grounds, 111 Wn.2d 1, 759 P.2d 372 (1988); see RCW 9A.46.020 (harassment). Conviction under subsection (1)(c) requires proof of victim selection along with malicious mischief. See RCW 9A.48.090 (malicious mischief in the third degree).
In subsection (1)(b), the Legislature provided examples of words or conduct which may constitute malicious harassment. These examples, which include (i) cross burning, (ii) hate symbols, and (iii) oral or written communication designed to harass or intimidate, do not themselves constitute the crime unless they place another in reasonable fear of harm to his/her person or property or the person or property of a third person. Section (1) of the statute is not violated absent conduct amounting to a crime of assault, harassment, or malicious mischief coupled with prohibited victim selection.
To ensure that RCW 9A.36.080(1) only applies to conduct, the Legislature specified the following exclusion:
However, it does not constitute malicious harassment for a person to speak or act in a critical, insulting, or deprecatory way unless the context or circumstances surrounding the words or conduct places another person in reasonable fear of harm to his or her person or property or harm to the person or property of a third person . . .[.]
When the Legislature amended the statute in 1984 to add this language, it did so to clarify that "words or acts must convey a threat of harm to the body or property of another person" to constitute malicious harassment. Final Legislative Report 163 (1984). Thus, from both the language of the statute and the legislative history, we conclude that RCW 9A.36.080(1)(a), (b), and (c) are intended to regulate only harmful conduct.
This court has upheld the constitutionality of other harassment laws where those statutes sought to regulate harmful conduct rather than the content of speech. Huff; Camby; Smith. Like the statutes under review in those cases, RCW 9A.36.080(1) specifically regulates conduct and only incidentally touches speech.
Several other state courts, since R.A.V., have upheld enhancement statutes against First Amendment challenge as conduct regulations rather than speech or thought regulations. In re Joshua H., 13 Cal. App. 4th 1734, 17 Cal. Rptr. 2d 291 (1993); State v. Plowman, 314 Or. 157, 838 P.2d 558 (1992), cert. denied, ___ U.S. ___, 125 L. Ed. 2d 666, 113 S. Ct. 2967 (1993); Dobbins v. State, 605 So. 2d 922 (Fla. Dist. Ct. App.), jurisdiction accepted, 613 So. 2d 3 (Fla. 1992); People v. Mulqueen, 155 Misc. 2d 632, 589 N.Y.S.2d 246 (Dist. Ct. 1992); People v. Miccio, 155 Misc. 2d 697, 589 N.Y.S.2d 762 (Crim. Ct. 1992).
In Plowman, the Oregon Supreme Court upheld a statute similar to RCW 9A.36.080. The court read R.A.V. to distinguish laws "directed against the substance of speech from laws that are directed against conduct." Plowman, at 168. The court determined that the Oregon statute was "directed against conduct" because it was not "directed against the substance" of any speech. Plowman, at 168. The court noted that in R.A.V., the Court "expressly did not rule on the constitutionality . . . of a statute like the one that we consider here." Plowman, at 168.
The Florida District Court of Appeals in Dobbins held that its enhancement statute did not run contrary to R.A.V. because "[i]t is only when one acts on such [hate-based] opinion to the injury of another that the statute permits enhancement." Dobbins, at 924. Thus, any speech implicated by the statute was sufficiently tied to criminal conduct to prevent the erosion of free speech. Dobbins, at 924.
Similarly, in Mulqueen, the New York District Court distinguished between statutes regulating criminal conduct and those regulating speech. Again, in Miccio, the New York Criminal Court held that its statute did not run contrary to R.A.V. because of a criminal conduct requirement. "In R.A.V., [conversely,] the activity of the defendant became criminal only when his actions amounted to the specifically proscribed fighting words." Miccio, 155 Misc. 2d at 701.
In Joshua H., the California Court of Appeal reviewed a juvenile court finding that a youth had assaulted an individual because of his sexual orientation thereby violating the California equivalent of RCW 9A.36.080(1). Like the defendant in this case, the defendant in Joshua H. argued that the statute was unconstitutional because it punished racist beliefs protected by the First Amendment. The court reasoned:
While we agree that bigotry is a protected class of expression, we disagree that the hate crime statutes punish bigotry. Rather, they punish the discriminatory act of selecting a crime victim based on his or her race or other status.
Joshua H., 13 Cal. App. 4th at 1743. The California Court of Appeal held that the statute was directed at conduct, not speech, and therefore did not violate the First Amendment.
These courts all agreed that the enhancement statutes in question were directed not at speech, but at conduct, and that they punished not thought or belief, but rather victim selection. This tight nexus between criminal conduct and the statutes sufficiently protected free speech guaranties.
Only two states have overturned enhancement statutes under R.A.V. State v. Wyant, 64 Ohio St. 3d 566, 597 N.E.2d 450 (1992), vacated and remanded, 113 S. Ct. 2954 (1993); State v. Mitchell, 169 Wis. 2d 153, 485 N.W.2d 807, rev'd, 113 S. Ct. 2194 (1993). «2»
«2» On June 11, 1993, the United States Supreme Court reversed the Wisconsin court's holding in State v. Mitchell, supra. Wisconsin v. Mitchell, ___ U.S. ___, 124 L. Ed. 2d 436, 113 S. Ct. 2194 (1993). The Wisconsin statute at issue in Mitchell punishes the same criminal conduct more harshly where the victim is selected because of race or other protected status. Specifically, the statute provides:
"(1) If a person does all of the following, the penalties for the underlying crime are increased as provided in sub. (2):
"(a) Commits a crime under chs. 939 to 948.
"(b) Intentionally selects the person against whom the crime under par. (a) is committed or selects the property which is damaged or otherwise affected by the crime under par. (a) because of the race, religion, color, disability, sexual orientation, national origin or ancestry or that person or the owner or occupant of that property."
Mitchell, 113 S. Ct. at 2197 n.1.
Citing Dawson v. Delaware, ___ U.S. ___, 117 L. Ed. 2d 309, 112 S. Ct. 1093 (1992) and Barclay v. Florida, 463 U.S. 939, 77 L. Ed. 2d 1134, 103 S. Ct. 3418 (1983), the Supreme Court held that the Wisconsin Legislature has the authority to determine, as a general matter, that bias-motivated offenses warrant greater penalties. Mitchell, 113 S. Ct. at 2196. The legislative finding that the harm caused by hate crimes to both society and the individual is greater provided adequate justification to enhance penalties designed to redress the greater harm. In response to the concern that the government was punishing motive, the Court pointed out that the motive underlying a particular crime has traditionally been a factor considered in the sentencing process. Mitchell, 113 S. Ct. at 2199. Since it is the role of government to set criminal penalties, the Court concluded that it is within the state's authority to enhance punishment across the board for hate-motivated crimes. Mitchell, 113 S. Ct. at 2200.
The Court then dismissed the distinctions drawn by the Wisconsin court between The Civil Rights Act of 1964, 42 U.S.C. §§ 2000e et seq. (Title VII) and the penalty enhancement statute, citing to its recent decision, R.A.V., 112 S. Ct. at 2546, where the Court included Title VII as an example of a permissible content-neutral regulation of conduct. Mitchell, 113 S. Ct. at 2200. As with the Title VII regulations, the Court found the enhancement statute "in this case is aimed at conduct unprotected by the First Amendment." Mitchell, 113 S. Ct. at 2201.
Finally, dismissing the overbreadth argument, the Court held that the First Amendment does not prohibit the use of speech to prove either motive or intent nor to prove the elements of a crime. Mitchell, 113 S. Ct. at 2201-02.
In both Wyant and Mitchell, the courts reviewed statutes similar to RCW 9A.36.080(1). Those courts reasoned that even though the statutes punished selection of victims, the real intent of the Legislature was to punish bigotry. This, the courts said, could not be done without offending the First Amendment. Wyant, at 576 (citing Abood v. Detroit Bd. of Educ., 431 U.S. 209, 234-35, 52 L. Ed. 2d 261, 284, 97 S. Ct. 1782, 1799 (1977)); Mitchell, 169 Wis. 2d at 166-67.
Both the Wyant and the Mitchell courts distinguished enhancement statutes from Title VII provisions. The Civil Rights Act of 1964, 42 U.S.C. §§ 2000e et seq. (Title VII). They reasoned that enhancement statutes constituted the "outright criminalization of subjective bigoted thought." Mitchell, 169 Wis. 2d at 177. Title VII, conversely, did not punish the defendant's actual motive per se, but rather, punished only the objective act of discriminatory treatment.
We are not persuaded by this distinction of Title VII from victim selection statutes. As Justice Bablitch stated in his dissent in Mitchell, "it is pure sophistry" to distinguish Title VII from such statutes. Mitchell, 169 Wis. 2d at 184 (Bablitch, J., dissenting). One cannot make a sound argument that the government can make discrimination in the hiring process illegal, but it cannot criminalize discrimination in selecting a victim for a crime.
 We find that the malicious harassment statute is similar to the various antidiscrimination laws governing employment practices, public accommodation, and housing. Such laws punish discriminatory acts committed "because of" the victims' protected status and are directed at discriminatory conduct rather than discriminatory thought or speech. See, e.g., 18 U.S.C. §§ 242, 245(b)(2); 42 U.S.C. §§ 1981, 1982, 2000a, 2000e-2, 3601 et seq.
Title VII is not concerned with discriminatory thought. It operates whether the employment action is based on xenophobia, intolerance, or fear of losing other bigoted customers. While the employment decision may reflect some bias, Title VII is only concerned with the act of discriminating in employment practices against a person because of his or her minority status. The Supreme Court has held that the object of antidiscrimination laws is not to punish employers' motives or speech, but to guard against employment practices that "operate as 'built-in headwinds' for minority groups". Griggs v. Duke Power Co., 401 U.S. 424, 432, 28 L. Ed. 2d 158, 91 S. Ct. 849 (1971) (good intent or absence of discriminatory intent does not redeem discriminatory employment practices). "Congress directed the thrust of the Act [Title VII] to the consequences of employment practices, not simply the motivation." Griggs, at 432; see International Bhd. of Teamsters v. United States, 431 U.S. 324, 335-36, 52 L. Ed. 2d 396, 97 S. Ct. 1843 (1977).
Likewise, RCW 9A.36.080(1) concerns not bigoted speech or thought, but rather the act of victim selection. As Justice Stevens stated in his concurrence in R.A.V., "[c]onduct that creates special risks or causes special harms may be prohibited by special rules." R.A.V., 112 S. Ct. at 2561 (Stevens, J., concurring). RCW 9A.36.080(1), like Title VII, punishes discriminatory conduct, not the reason for the discrimination. Thus, penalty enhancement statutes, like antidiscrimination laws, regulate conduct rather than speech.
CONTENT-BASED REGULATION OF SPEECH
[5, 6] The respondents here argue that even if RCW 9A.36.080(1) applies facially to criminal conduct, it operates only where bigoted thought is present. Since punishment is enhanced only in these circumstances, they reason that the statute effectively punishes bigoted thought. Even if we were to agree with the respondents that enhancement statutes punish bigoted thought, RCW 9A.36.080(1) would still withstand constitutional scrutiny under R.A.V. Noting that "content discrimination among various instances of a class of proscribable speech often does not pose [a] threat", Justice Scalia found three exceptions to the prohibition against content discrimination. R.A.V., 112 S. Ct. at 2545.
The first exception in R.A.V. is where the "basis for the content discrimination consists of the very reason the entire class of speech at issue is proscribable". R.A.V., 112 S. Ct. at 2545. The examples the Court gave for this exception were (1) prohibitions against only that obscenity which is most offensive in its prurience, (2) threats of violence directed against the President, and (3) regulation of price advertising in only one industry because, in the State's view, the risk of fraud is greater in that industry. R.A.V., 112 S. Ct. at 2546.
In accord with this exception, RCW 9A.36.080(1) punishes defendants who act with an even more depraved intent, an intent to target a crime victim because of that person's protected status. The resulting harm is greater than the harm caused by that same conduct absent the special animus underlying hate crimes. The greater harm of such acts justifies the increased punishment.
Hate crimes are undeniably more damaging than their content-neutral counterparts. In adopting the malicious harassment statute, the Washington Legislature found that:
In recent months incidents of violence and intimidation based on racial or ethnic bias have increased throughout the state. Existing law protects minorities against discrimination in the areas of employment, public accommodation, credit, insurance and real estate transactions; the criminal code addresses assaultive conduct and malicious mischief. However, no current law provides criminal sanctions or civil redress for threats or vicious acts which are racially motivated. Such a law is believed necessary to prevent confrontations and to protect civil rights of all citizens.
1981 Final Legislative Report 155.
Moreover, there are other instances in the criminal code where the Legislature has similarly enhanced penalties for targeting a victim based on the victim's status. For example, threats against the Governor are prohibited, RCW 9A.36.090, likewise, intimidating witnesses, RCW 9A.72.110, jurors, RCW 9A.72.130, and judges, RCW 9A.72.160. When criminal conduct is aimed at individuals because of their particular status, the State may legitimately conclude that there is greater harm justifying greater punishment.
RCW 9A.36.080 also conforms with the second R.A.V. exception which permits inadvertent content regulation, i.e., where the content is not the object of the regulation. St. Paul had argued in the juvenile court that "[t]he burning of a cross does express a message and it is, in fact, the content of that message which the St. Paul Ordinance attempts to legislate." R.A.V., 112 S. Ct. at 2548. Unlike the St. Paul statute, RCW 9A.36.080 is not concerned with the listener's reactions to speech. As with Title VII, RCW 9A.36.080(1) is concerned with the additional harm to the victim of a hate crime and that crime's effect on society as a whole. If the conduct in question can be characterized as speech, RCW 9A.36.080 is concerned only with its secondary effects.
As the American Civil Liberties Union of Washington points out in its amicus brief, a malicious harassment statute is "a public necessity". Brief of Amicus Curiae, at 15. "Experience shows that threats of violence based on personal characteristics or group identity cause deep individual and societal harm and that they diminish, rather than enhance, the exchange of ideas in a free society." Brief of Amicus Curiae, at 15.
Scholars agree with the proposition that "hate crimes" carry harmful secondary effects. Gellman writes that:
When attacks are made on one group, members of other disempowered groups may feel threatened as well: a rash of attacks on African Americans by a racist group may well create apprehension among Asians, Jews, gays, or Hispanics in the neighborhood. . . . Majority group members may feel both relief that they are not targets of such attacks, and fear of victimization should they come to be viewed as minority sympathizers.
Gellman, Sticks and Stones Can Put You in Jail, But Can Words Increase Your Sentence? Constitutional and Policy Dilemmas of Ethnic Intimidation Laws, 39 U.C.L.A. L. Rev. 333, 342 (1991-1992); see also Delgado, Words that Wound: A Tort Action for Racial Insults, Epithets, and Name-Calling, 17 Harv. C.R.-C.L. L. Rev. 133 (1982-1983).
RCW 9A.36.080(1) also falls under the third exception for content regulations of speech that present virtually no danger of eroding First Amendment rights. See R.A.V., 112 S. Ct. at 2547.
Where the government does not target conduct on the basis of its expressive content, acts are not shielded from regulation merely because they express a discriminatory idea or philosophy.
R.A.V., 112 S. Ct. at 2546-47. The express language of subsection (1)(b) protects such discriminatory ideas and philosophies where they are not combined with criminal acts.
Finally, respondents urge this court to follow the reasoning adopted by the Wisconsin and Ohio courts which have held that while punishment of "intent" is appropriate, statutes cannot constitutionally punish "motive". While we concede that there may be a distinction between intent and motive, we are persuaded that in this context it is a distinction without a difference. While the State must prove that the defendant selected a victim because of perceived membership in a specified group, it is not required to establish the "motive" underlying that choice. As discussed earlier, RCW 9A.36.080 punishes the intentional act of discrimination, not the motive underlying the act.
 Moreover, the Supreme Court has recognized that, where evidence of racial intolerance is relevant to the issues, the constitution does not prohibit such evidence simply because those "beliefs and associations are protected by the First Amendment." Dawson v. Delaware, ___ U.S. ___, 117 L. Ed. 2d 309, 112 S. Ct. 1093, 1097 (1992). See also Mitchell, 113 S. Ct. at 2200.
 Respondents also argue that RCW 9A.36.080 is overbroad because it implicates speech as circumstantial evidence of victim selection and because it chills protected speech. Washington courts apply the federal overbreadth analysis to such challenges. Huff, 111 Wn.2d at 923, 925. A statute is overbroad if it prohibits constitutionally protected speech within its plainly legitimate sweep. Thornhill v. Alabama, 310 U.S. 88, 97, 84 L. Ed. 1093, 60 S. Ct. 736 (1940); Seattle v. Eze, 111 Wn.2d 22, 31, 759 P.2d 366, 78 A.L.R.4th 1115 (1988).
A statute is void in its entirety for overbreadth only if it reaches a substantial amount of protected speech and there exists no way of severing the statute's unconstitutional applications. Huff, at 925; Houston v. Hill, 482 U.S. 451, 458, 96 L. Ed. 2d 398, 410, 107 S. Ct. 2502, 2508 (1987) (citing New York v. Ferber, 458 U.S. 747, 73 L. Ed. 2d 1113, 102 S. Ct. 3348 (1982)). The Court has held that the overbreadth doctrine is "strong medicine" when used to invalidate a statute on its face and it should be applied "only as a last resort." Broadrick v. Oklahoma, 413 U.S. 601, 613, 37 L. Ed. 2d 830, 93 S. Ct. 2908 (1973). In Broadrick, the Court required substantial regulation of speech before it invalidated a statute on overbreadth grounds. In that case, state employees challenged a statute which prohibited them from engaging in political fundraising. Within the statute's legitimate sweep were activities such as wearing campaign buttons or displaying bumper stickers. The Court held that the illegitimate applications of the statute were insubstantial when compared to the statute's legitimate sweep. Broadrick, at 615.
 In this case, respondents argue that RCW 9A.36.080(1) implicates protected speech since the State uses the defendant's speech and expressive conduct as circumstantial evidence. Applying the rationale in Broadrick, we conclude that the incidental impact on speech does not render subsection (1) unconstitutionally overbroad. First, it regulates primarily conduct. Second, its effect on speech is minimal. As the Court in R.A.V. explained, that an act expresses a discriminatory idea or philosophy does not mean that it is shielded from regulation. R.A.V., 112 S. Ct. at 2546-47. The nexus between criminal conduct and any speech implicated ensures that subsection (1) does not deter a substantial amount of protected expression.
As is evident in the statute itself, the Legislature ensured that, absent criminal conduct, bigoted speech and thought are protected. A person is free under the statute to make his or her odious bigoted thoughts known to the world so long as those words do not cross the boundary into criminal harassment, assault, or property damage. Any overbreadth apparent in the application of RCW 9A.36.080(1) is insufficient to justify facial invalidity.
[10, 11] Respondent Stevens asserts that RCW 9A.36.080(1) is overbroad because the defendant's speech may be used by prosecutors to prove victim selection. It is true that utterances by the defendant may offer circumstantial evidence of discrimination or victim selection, but as with employment discrimination, victim selection can be shown by a pattern of conduct absent any speech. Even if speech is used to prove victim selection, we can find no distinction between using speech to prove malicious harassment or any other crime. Only that speech relevant to proving the crime will be admitted. As is always the case, the trial judge will be required to balance the probative value of the evidence against the prejudice to the defendant. The State cannot simply produce evidence of bigoted beliefs. Before such evidence can be admitted, the State must establish the relationship between the speech and the act of victim selection. See Rules of Evidence 402, 403, 404; accord, Mitchell, 113 S. Ct. at 2200.
Respondents also argue that the statute is overbroad because it chills protected speech. The only chilling effect that RCW 9A.36.080 has is on lawless conduct or bigoted speech uttered during lawless conduct. To be sure, the actor committing a victim selection crime may be less likely to utter a racial epithet when committing the crime. But this is insufficient to warrant invalidation of the statute. Any danger that the statute's overbreadth will chill protected bigoted speech is sufficiently minimized by the language of subsection (1)(b)(iii) which specifically protects such speech. See RCW 9A.36.080(1)(b)(iii).
Respondent Myers asserts that RCW 9A.36.080 is unconstitutionally overbroad as a prior restraint as well. He argues that the Washington Constitution categorically prohibits prior restraints. In support of this proposition, he cites O'Day v. King Cy., 109 Wn.2d 796, 749 P.2d 142 (1988). Myers fails, however, to brief any of the six factors set out in State v. Gunwall, 106 Wn.2d 54, 720 P.2d 808, 76 A.L.R.4th 517 (1986) which we held are necessary in determining whether the Washington State Constitution provides greater protection than the federal constitution.
We conclude that RCW 9A.36.080(1) is not unconstitutionally overbroad because its objective is to regulate conduct rather than speech or thought. The implication of speech as circumstantial evidence is minor and does not warrant the statute's invalidity. As the Dobbins court stated, "[i]t doesn't matter that Dobbins hated Jewish people or why he hated them; it only mattered that he discriminated against [him] by beating him because he was Jewish." Dobbins, 605 So. 2d at 925.
VAGUENESS AND EQUAL PROTECTION
In addition to his other arguments, Stevens asserts that RCW 9A.36.080(1) is impermissibly vague. He argues that the meaning of the phrase "in a way that is reasonably related to, associated with, or directed toward" is unclear. He also questions whether the phrase "reasonable fear" is measured by a subjective or objective standard.
[12, 13] This court applies the federal due process test for vagueness; a statute must provide both adequate notice and standards to prevent arbitrary enforcement. Huff, 111 Wn.2d at 929 (upholding state telephone harassment statute); accord, Plowman, 314 Or. at 162. Further, the challenging party must prove unconstitutionality beyond a reasonable doubt. Smith, 111 Wn.2d at 17.
 Turning to the terms under challenge we note they are not defined in the statute. Where terms are not defined, the court will look to the plain, ordinary meaning of the words. American Legion Post 32 v. Walla Walla, 116 Wn.2d 1, 802 P.2d 784 (1991). In ordinary usage, the terms "related" and "associated" are synonymous and mean "connected" or "united" in purpose and interest. Webster's New World Dictionary of the American Language 89, 1227 (1968). "Directed toward" means aimed at achieving an objective. Webster's, at 414; see also State v. Hansen, 67 Wn. App. 511, 837 P.2d 651 (1992) (defining the verb "direct").
[15, 16] Additionally, when analyzing the wording of a statute, the court will read the statute as a whole. Service Employees Int'l Union, Local 6 v. Superintendent of Pub. Instruction, 104 Wn.2d 344, 348-49, 705 P.2d 776 (1985). Applying the doctrine of ejusdem generis to the terms in the paragraph at issue, RCW 9A.36.080(1), we interpret the allegedly vague terms in a manner consistent with the other words in the sequence, in this case, consistent with the words "because of". State v. Hutsell, 120 Wn.2d 913, 918, 845 P.2d 1325 (1993) (citing Dean v. McFarland, 81 Wn.2d 215, 221, 500 P.2d 1244, 74 A.L.R.3d 378 (1972)). In ordinary usage "because of" means "by reason of" or "on account of". Webster's, at 131; see Plowman, at 161-62. When read as a whole, this language is clear and provides adequate notice that the prohibited conduct is the selection of crime victims from certain specified categories.
We also find that the statute gives adequate standards to ensure proper enforcement. Requiring the State to prove that the threats placed the victim in "reasonable fear" provides an objective standard by which to evaluate the harm to the victim. While Judge Pechman expressed concern about the term "reasonable fear", we note that the term "reason-able" is used commonly in criminal law. In the defenses chapter of the Washington Criminal Code, RCW 9A.16.010, "necessary" is defined as meaning no "reasonably" effective alternative to force exists and the force used is "reasonable".
"Reasonable fear" is an element of harassment under RCW 9A.46.020. In Smith, this court rejected a challenge to that statute noting that because the victim must "reasonably fear" that the threats will be carried out, the statute is not unconstitutional. Smith, 111 Wn.2d at 15-16. "Reasonable fear" by a victim that a threat will be carried out has been held to limit application of an overly broad intimidation statute. See Smith, 111 Wn.2d at 15 n.5 (citing Wurtz v. Risley, 719 F.2d 1438, 1441 (9th Cir. 1983)). We have likewise held that the term "unreasonable" increases objectivity, thus reducing vagueness. Eze, 111 Wn.2d at 29.
 We find the statute provides adequate notice and sufficient standards to prevent arbitrary enforcement because the average citizen can understand the proscription at hand. Huff, at 929.
Respondent Stevens' final argument against RCW 9A.36.080(1) is that it violates equal protection. He states that the act allows prosecutorial discretion to charge a defendant with either malicious harassment or a lesser assault charge for the same conduct depending on whether the State approves of the defendant's beliefs. Stevens asserts that a prosecutor's exercise of discretion in charging some persons and not others, all guilty of the same crime, violates the Fourteenth Amend-ment's equal protection clause and article 1, section 12 of the Washington Constitution.
[18, 19] The United States Supreme Court and this court have long held in favor of prosecutorial choice. Bordenkircher v. Hayes, 434 U.S. 357, 365, 54 L. Ed. 2d 604, 98 S. Ct. 663, reh'g denied, 435 U.S. 918, 55 L. Ed. 2d 511, 98 S. Ct. 1477 (1978); State v. Pettitt, 93 Wn.2d 288, 294, 609 P.2d 1364 (1980). Such discretion is, however, subject to constitutional constraints, and selective enforcement deliberately based on unjustifiable standards does raise equal protection concerns. United States v. Batchelder, 442 U.S. 114, 125, 60 L. Ed. 2d 755, 99 S. Ct. 2198 (1979) (citing Oyler v. Boles, 368 U.S. 448, 456, 7 L. Ed. 2d 446, 82 S. Ct. 501 (1962)); State v. Lee, 87 Wn.2d 932, 936, 558 P.2d 236 (1976) (also citing Oyler). This court has held that where there are differing elements between offenses "the prosecutor's discretion is limited by consideration of which elements [can be proved in the particular case]". Kennewick v. Fountain, 116 Wn.2d 189, 193, 802 P.2d 1371 (1991).
In this case, the increased punishment for assault under RCW 9A.36.080(1) is justified by the additional element of victim selection and its associated greater harm. Under the facts before us we find the prosecutor's charging decision was not based on a bias toward the defendant's thought. Rather, the prosecutorial decision was dictated by statutory consideration of the harm to the victim. Respondent Stevens provides no evidence of improper or unjustified prosecutorial discretion.
Stevens relies substantially on State v. Devine, 84 Wn.2d 467, 527 P.2d 72 (1974) for his argument. Devine holds that equal protection is violated if two statutes parade under different titles, with different penalties, but have identical elements for what constitutes the crime. In this case, for example, Stevens would argue that since RCW 9A.36.041 designates assault in the fourth degree as a gross misdemeanor, and those same actions with the additional element of victim selection result in a felony under RCW 9A.36.080(1), the statute in question must violate equal protection. This argument fails because the two statutes redress substantially different harm and have different elements.
Stevens next posits that the statute violates equal protection because a white defendant engaging in the same act toward a white victim would most likely be charged with misdemeanors. Stevens asserts that the prosecutor would factor in race and that a victim of color would warrant a higher charge against a white defendant. State v. Judge, 100 Wn.2d 706, 713, 675 P.2d 219 (1984). Stevens' use of Judge is misplaced. Judge stands for the proposition that a prosecutor may not charge defendants differently based on the defendants' classification. RCW 9A.36.080(1) is racially neutral and applies to any defendant who targets any victim on the basis of any of the listed characteristics. For example, the prosecutor charged Talley, the white defendant, with malicious harassment of a black man, three white women, and two biracial children.
We find no violation of equal protection.
OVERBREADTH AND CONTENT-BASED REGULATION
[20-22] RCW 9A.36.080(1) regulates only criminal conduct and any effect that it has on speech is only incidental. RCW 9A.36.080(2), on the other hand, clearly regulates protected symbolic speech based on content. Any effect it has on regulating conduct is merely incidental. Subsection (2) is, therefore, unconstitutional.
In defense of the regulation, the State argues that subsection (2) should be read in conjunction with subsection (1) thus requiring the prohibited symbolic speech be in combination with criminal conduct. However, as a general proposition, where the Legislature uses different language within the same statute, we presume that it intended to address different concerns. Automobile Drivers & Demonstrators Union Local 882 v. Department of Retirement Sys., 92 Wn.2d 415, 598 P.2d 379 (1979), appeal dismissed, cert. denied, 444 U.S. 1040 (1980); Greenwood v. Department of Motor Vehicles, 13 Wn. App. 624, 536 P.2d 644, 98 A.L.R.3d 566 (1975). RCW 9A.36.080(1)(b)(i) and (ii) apply to situations where such symbolic speech is a part of the underlying assault, harassment, or malicious mischief. Under the State's reading of the statute, subsection (2)(a) would be repetitive and superfluous. It is clear, therefore, that the Legislature intended to criminalize cross burning and depiction of hate symbols "per se". This is unconstitutional because symbolic hate speech, however offensive, is protected. R.A.V., 112 S. Ct. at 2548; Texas v. Johnson, 491 U.S. 397, 406, 105 L. Ed. 2d 342, 109 S. Ct. 2533 (1989); Brandenburg v. Ohio, 395 U.S. 444, 447, 23 L. Ed. 2d 430, 89 S. Ct. 1827 (1969).
In the alternative, the State argues that RCW 9A.36.080(2) is beyond constitutional scrutiny because it punishes only unprotected "fighting words" or "words . . . which by their very utterance inflict injury or tend to incite an immediate breach of the peace." Chaplinsky v. New Hampshire, 315 U.S. at 572. The State asserts that fighting words are excluded from First Amendment protection. Chaplinsky, at 571-72.
The State is correct. The Supreme Court has permitted content-based regulation of speech within certain well-defined categories of unprotected, low value speech. Police Dep't v. Mosley, 408 U.S. 92, 33 L. Ed. 2d 212, 92 S. Ct. 2286 (1972); Chaplinsky, at 571-72. Such categories of low value speech include, but are not limited to, obscenity, defamation, and "fighting words". However, a finding that a category of speech, such as fighting words, has low First Amendment value does not mean that the speech is wholly without constitutional protection. If a regulation of low value speech is content based, the court applies the same stringent standard of review that it applies to all other content-based regulations. See Stone, Content Regulation and the First Amendment, 25 Wm. & Mary L. Rev. 189, 197 (1983-1984). This general prohibition of content discrimination is premised on the danger that government may "effectively drive certain ideas or viewpoints from the marketplace" by regulating content. R.A.V., 112 S. Ct. at 2545.
Contrary to the State's assertions, however, subsection (2) falls squarely within the prohibitions of R.A.V. Like the St. Paul ordinance, RCW 9A.36.080(2) criminalizes symbolic speech that expresses disfavored viewpoints in an especially offensive manner. Even if construed to address only fighting words, as the Minnesota Supreme Court did with the St. Paul ordinance, the statute is still unconstitutional under the R.A.V. analysis because even fighting words may not be regulated based on their content. However objectionable, this speech is protected by article 1, section 5 of the Washington Constitution and the first and fourteenth amendments to the United States Constitution. Mosley, at 97; O'Day, 109 Wn.2d at 802.
"No one pretends that actions should be as free as opinions." J. Mill, On Liberty 119 (1st ed. 1859). RCW 9A.36.080(1) punishes the act, not the opinion. It enhances punishment where the crime is made more harmful because the actor selected the victim based on the victim's association in a protected class. We hold that RCW 9A.36.080(1) is a valid exercise of police power that does not abridge the First Amendment. RCW 9A.36.080(2), conversely, punishes symbolic speech because of its offensive message. It is, therefore, an unconstitutional content-based regulation of protected speech.
We affirm King County Superior Court Judge Aitken's dismissal of charges against Mr. Talley. While Talley targeted his victims within the meaning of RCW 9A.36.080, his other conduct did not constitute criminal harassment, assault, or property damage. Despite Talley's odious behavior, his actions do not fall within the prohibitions of RCW 9A.36.080(1).
Conversely, Myers' and Stevens' conduct arguably falls within subsection (1) of the statute either because they actually damaged the victims' property, or because they placed the victims in reasonable fear for their persons.
Accordingly, we reverse the trial court and remand for appropriate disposition in accord with this decision.
ANDERSEN, C.J., and UTTER, BRACHTENBACH, DURHAM, SMITH, GUY, and JOHNSON, JJ., concur.