Soon after appellant Johnnie Coleman asserted his right to a jury trial, the government filed an amended information reducing the charge against him from stalking 1 to attempted stalking. 2 The case was transferred to a misdemeanor calendar, and Mr. Coleman was convicted of attempted stalking after a bench trial. On appeal, Mr. Coleman argues that he was denied his right to a jury trial and that his conviction was not supported by constitutionally sufficient evidence.

One of the elements of stalking is that the defendant "purposefully engage in a course of conduct" involving at least two "occasions" of certain statutorily specified types of behavior-for example, following the complainant or communicating with him or her. D.C. Code §§ 22-3132 (8), - 3133(a). As part of his sufficiency claim, Mr. Coleman argues that the government failed to prove that he possessed the requisite mental state-namely, that he "should have known" that a "reasonable person in the [complainant's] circumstances" would fear for her or another's safety, "[f]eel seriously alarmed, disturbed, or frightened," or "[s]uffer emotional distress"-during at least two of the occasions that allegedly comprised his course of conduct. D.C. Code § 22-3133 (a)(3). We agree that the government was required to prove that Mr. Coleman possessed the requisite mental state on at least two occasions. Accordingly, although we reject Mr. Coleman's jury trial arguments and conclude that the government introduced sufficient evidence to meet its burden, we vacate Mr. Coleman's conviction and remand to allow the trial court to evaluate the evidence under the principles set forth in this opinion.

I.

During the time period relevant in this case, Mr. Coleman lived in a group home for "men who had mental[ ] or behavioral *1131 issues." The home was located on the Kansas Avenue side of a triangular block bounded by Kansas and Eastern avenues and Tuckerman Street in Northeast Washington, D.C. The complainant lived with her husband and adult son in a house on the Tuckerman Street side of the block, and the backyard of their home abutted the backyard of the group home. The allegation of stalking arises out of a series of four incidents. 3

A. The Two "Staring" Incidents

The first two incidents occurred in the spring of 2015. One morning, on her way to work, the complainant boarded a bus at the corner of Tuckerman Street and Kansas Avenue. A man whom the complainant later identified as Mr. Coleman boarded the bus at the next stop, at the corner of Kansas and Eastern avenues. According to the complainant, Mr. Coleman "practically just stared at [her] the entire" five- to ten-minute bus ride. She "never would have recalled" the incident of "Mr. Coleman staring at [her] on the bus[,] except for the fact that maybe a week or two later" a second incident occurred. The complainant was with her family in her backyard and saw Mr. Coleman "standing in the backyard of the group home," "staring at [her] and [her] family." Because the complainant recognized Mr. Coleman from the bus ride a week or two earlier, she found his behavior "a little alarming."

B. The October 12 Incident

After these two incidents, Mr. Coleman and the complainant had no further contact for several months. Around 8 a.m. on Monday, October 12, 2015, the complainant was walking to the Lamond Recreation Center, which is at the corner of Tuckerman Street and Kansas Avenue, "about half a block away from" her home. The center consists of an indoor facility and a baseball field with a walking path around it. The complainant was "heading up the sidewalk" on Tuckerman Street "to go on to the field" when Mr. Coleman approached from the opposite direction. According to the complainant, after they passed each other, Mr. Coleman "quickly turned around[ ] and ... ran back up and like stood directly in front of [her]," "stopp[ing] [her] in her tracks." Mr. Coleman said "something to the effect [of] like oh, how are you doing? You look really nice. Can I talk to you? That kind of thing." The complainant testified that she was "startled." She walked away from Mr. Coleman "[b]risk[ly]" and told him, "You know, you're my neighbor. I'm not trying to talk to you. You know I have a husband. I have a dog. Okay. Good bye. Have a good day."

The complainant proceeded to walk on the path around the baseball field. She soon noticed, however, that Mr. Coleman had taken a seat on a "bench ... at the baseball diamond." The complainant testified that she was "concerned" because it was "still fairly early in the morning" and because Mr. Coleman "was the only person out there." She was "a little bit worried that [Mr. Coleman] was prepared to ... cause [her] some harm" and that his presence "made [her] feel very uncomfortable." The complainant continued her walk, but she also called her husband, who had not yet left for work. When the complainant's husband arrived, he and Mr. Coleman had what "looked like ... a calm, pleasant conversation." Mr. Coleman walked away, and the complainant's husband told her *1132 that Mr. Coleman was "harmless" and that "he understands now that he doesn't need to be out here harassing people." Mr. Coleman left, and the complainant continued her walk around the baseball field. "[S]uddenly," she saw Mr. Coleman "across the street, waiting at the bus stop on Kansas [Avenue] and Tuckerman Street." The complainant continued "walking and watching, walking and watching," and saw two buses go by without Mr. Coleman boarding. The complainant then saw Mr. Coleman leave, and she assumed that he was going back to his group home.

When the complainant finished her walk, "something just told [her] that [she] should probably ... go down the street and knock on the group home door, just to alert them, as a concerned neighbor." The complainant there told a nurse that she was "concerned" because Mr. Coleman had been "wandering around"; the nurse responded that Mr. Coleman had not "take[n] his medication" and had "just left." The complainant began walking back home, and when she was about two houses away from the group home, she heard Mr. Coleman-who was standing in front of the group home-"yell[ ] out to" her. Mr. Coleman said, "I see you tried to get me in trouble. I'm not a bad person.... I read the Bible. It's not like I go around masturbating ...." The complainant "thought [this] was just odd," and she responded by telling Mr. Coleman to "go inside and take [his] medicine." She recalled that this interaction made her "really uncomfortable."

The complainant testified that as a result of the October 12, 2015, incident, she "became a lot more wary" and decided not to go "walking that early anymore."

C. The October 26 Incident

The complainant's next and final encounter with Mr. Coleman occurred two weeks later, on October 26, 2015, around noon. She was walking on the path around the baseball field when she "notice[d] that [Mr. Coleman] was standing up at the doorway to the recreation center." There were some children and their teachers on the baseball field, so the complainant "didn't feel, at that moment, ... like there was any issue." After walking several laps around the field, the complainant saw the "kids and the teachers ... leaving the field." The complainant was on the opposite side of the baseball field from the recreation center, and "before [she] kn[e]w it, [Mr. Coleman] literally sprinted across the entire length of the field, and he was standing about eight feet in front of [her]," on the walking path. 4 The complainant testified that Mr. Coleman had "this scary-looking grin on his face."

At that moment, the complainant testified, she "knew it was like fight or flight," so she "stopped on [her] heel, ... turned in the other direction," and told Mr. Coleman to "leave [her] alone." The complainant quickly headed towards the baseball field's exit-she was "almost running"-and she "started cussing at [Mr. Coleman]": "[L]eave me alone. Stop acting weird. Stop bothering me. Get the fuck away from me. ... [L]eave me the fuck alone." Mr. Coleman followed her and said, "I'm just out here trying to get some exercise." The complainant proceeded quickly toward her home, and Mr. Coleman continued following. 5 The complainant "screamed" at Mr. Coleman that she was going to "get [her] husband to fuck [him] up," and the complainant's neighbor also told Mr. Coleman to leave the complainant *1133 alone, but Mr. Coleman continued following, shouting things like "F you bitch." When the complainant arrived at her house, she told her son that Mr. Coleman had been following her. The complainant's son went outside and told Mr. Coleman "to get the fuck out of here." Mr. Coleman walked away, stopping briefly at the corner and "erratically shaking a [street] sign" before leaving the area. The complainant's neighbor called the police while these events were unfolding, and Mr. Coleman was arrested soon thereafter.

The complainant testified that this incident was a "traumatic experience" for her and that she had "lost some sleep" as a result.

D. The Trial Court's Verdict

The trial court credited the complainant's testimony and made the following factual findings. The court found that the complainant first noticed Mr. Coleman in the spring of 2015, when he stared at her on the bus. She encountered him a second time about one week later, when he again stared at her while she was in her backyard. With respect to the October 12 encounter, the court found that Mr. Coleman ran back to the complainant after walking past her and asked to talk to her, a request she rejected. Mr. Coleman then sat on the bleachers and watched the complainant as she walked around the field. When she finished her walk, the complainant noticed Mr. Coleman standing at the bus stop without boarding any of the buses that went by. She then walked to Mr. Coleman's group home to tell them that a resident was wandering around in the early morning hours, and when she left, Mr. Coleman "yell[ed] at her, saying I see you're trying to get me in trouble."

The court also found that on October 26, Mr. Coleman "sprinted across the field" toward the complainant and "started following her" after she told him to leave her alone. Mr. Coleman crossed the street to continue following her and yelled and cursed at her, "saying F you, bitch[,] and words to that effect." Mr. Coleman stayed in the area after the complainant got home and shook a street sign on the corner. The trial court rejected Mr. Coleman's testimony that he was not following the complainant and was just trying to go home, noting that his testimony was "flatly contradicted by the record."

After making these factual findings, the trial court concluded:

I believe the Government has satisfied each and every element beyond a reasonable doubt, when you look at the totality of the circumstances as well as the constant staring at her, and basically engaging in harassment, and I believe she was fearful of him, and so much so that she went to the group home to tell them, you know, about his conduct. And also spoke on the 9-1-1 call telling them about who the defendant was and where he was found.

So the Court is satisfied the Government has proved in each area and element beyond a reasonable doubt and I'm going to enter a judgment of guilty to the charges, attempted stalking.

During the sentencing stage, which immediately followed the verdict, the court added that it considered the case "rather disturbing." The judge explained that he did not "believe for one moment that [Mr. Coleman] didn't understand how much he was bothering this woman," because he "was persistent in following her" after being told to stop.

II.

Mr. Coleman first claims that he was denied his constitutional and statutory right to a jury trial when the government amended the charge from stalking to attempted stalking. Although Mr. Coleman initially requested a jury trial, he did not object to the amended information and *1134 made no jury demand on the attempted stalking charge. We thus review this claim for plain error, and we will reverse only if Mr. Coleman demonstrates that the filing of the amended information "was (1) error, (2) that is plain, (3) that affects substantial rights, and (4) that seriously affects the fairness, integrity or public reputation of judicial proceedings." Jones v. United States , 124 A.3d 127 , 129 (D.C. 2015) (internal quotation marks omitted). Mr. Coleman's challenge in this regard boils down to a claim that the trial court plainly erred in failing to grant Mr. Coleman a jury trial sua sponte . See Fretes-Zarate v. United States , 40 A.3d 374 , 376 (D.C. 2012) (applying plain error review to a non-citizen appellant's claim that trial judge should have advised her sua sponte of her right to a jury trial).

"It has long been settled that 'there is a category of petty crimes or offenses which is not subject to the Sixth Amendment jury trial provision.' " Blanton v. City of N. Las Vegas , 489 U.S. 538 , 541, 109 S.Ct. 1289 , 103 L.Ed.2d 550 (1989) (quoting Duncan v. Louisiana , 391 U.S. 145 , 159, 88 S.Ct. 1444 , 20 L.Ed.2d 491 (1968) ). To determine whether the constitutional right to a jury trial attaches to a particular offense, we examine "objective indications of the seriousness with which society regards the offense," the most relevant of which is the "maximum penalty set by the legislature." United States v. Nachtigal , 507 U.S. 1 , 3, 113 S.Ct. 1072 , 122 L.Ed.2d 374 (1993) (quoting Blanton , 489 U.S. at 541 , 109 S.Ct. 1289 ). "[O]ffenses for which the maximum period of incarceration is six months or less are presumptively 'petty.' " Id. A defendant may overcome this presumption " only if he can demonstrate that any additional statutory penalties, viewed in conjunction with the maximum authorized period of incarceration, are so severe that they clearly reflect a legislative determination that the offense in question is a 'serious' one." Blanton , 489 U.S. at 543 , 109 S.Ct. 1289 (emphasis added).

Mr. Coleman acknowledges that attempted stalking is a presumptively petty offense, carrying a maximum term of incarceration of 180 days. D.C. Code § 22-1803 . He argues, however, that the D.C. Council's unambiguous intent to have stalking charges tried by a jury, and not by an individual judge, overcomes the presumption that the legislature did not consider attempted stalking a serious offense. Mr. Coleman's point is not without force. The current version of the stalking statute was enacted as part of the Omnibus Public Safety and Justice Amendment Act of 2009, D.C. Law 18-88, 56 D.C. Reg. 7413 (Dec. 10, 2009). Citing the "subjective nature" of stalking, the Council's Committee on Public Safety and the Judiciary deemed it an offense for which "the community, not a single judge, should sit in judgment" and found it "highly appropriate that a jury of [the defendant's] peers ... judge whether the behavior is acceptable or outside the norm and indicative of escalating problems." D.C. Council, Comm. on Pub. Safety & Judiciary, Rep. on Bill 18-151, at 33 (June 26, 2009), http://lims.dccouncil.us/Download/22306/B18-0151-CommitteeReport1.pdf (Committee Report). The Council expressly set the maximum penalty for stalking at a level that guaranteed the defendant's right to a jury trial. Id. (explaining that the penalty of twelve months for first-time stalking offenders was established "so that a defendant will have a right to a jury of [his] peers").

But even assuming the statements in the Committee Report are probative of the Council's view of the seriousness of attempted stalking, the Supreme Court has clearly instructed courts to consider only the additional statutory penalties, and this court has accordingly refused to rely on legislative history in evaluating whether there is a constitutional right to a jury *1135 trial for a particular offense. See, e.g. , Day v. United States , 682 A.2d 1125 , 1129 (D.C. 1996) ("The [Supreme] Court does not ... require a search of the legislative history to find the legislature's attitudes; it points only to the penalties that the legislature imposes."); Burgess v. United States , 681 A.2d 1090 , 1095 (D.C. 1996) ("We decline on several grounds Burgess'[s] invitation to explore the legislative history.... We note that with all the opportunities the Supreme Court has had to consider this issue, it has never suggested that legislative history might be considered as a gauge of whether a particular offense is 'serious' or 'petty.' "); Stevenson v. District of Columbia , 562 A.2d 622 , 623 n.1 (D.C. 1989) ("In light of this plain focus on statutory penalties, we decline appellant's invitation to search the committee report or other legislative history for additional indications that the Council believed driving under the influence to be a "serious" offense."). We likewise decline to do so here. Having failed to identify any "additional statutory penalties" which, viewed in conjunction with the maximum penalty of 180 days, are "so severe" as to indicate that the Council considered the offense a serious one, see Blanton , 489 U.S. at 543 , 109 S.Ct. 1289 , Mr. Coleman cannot overcome the presumption that attempted stalking is a petty offense to which the Sixth Amendment right to a jury trial does not attach. Facing at most 180 days' imprisonment and a $ 1,000 fine, Mr. Coleman had no constitutional right to a jury trial for the charge of attempted stalking. 6

With respect to Mr. Coleman's claim that he was denied his statutory right to a jury trial, D.C. Code § 16-705 (b) provides that a defendant is entitled to a jury trial on all offenses (except contempt of court) that are punishable by more than 180 days' imprisonment. A maximum penalty of 180 days, however, "is not sufficient to trigger one's statutory right to a jury." Jones , 124 A.3d at 132 . Though he acknowledges that attempted stalking carries a maximum penalty of 180 days, Mr. Coleman urges us to read the term "offense" in that statute as referring to the substantive offense-here, stalking-where the elements of the attempted offense still require a determination of reasonableness under prevailing community norms and where the Council has determined that a jury is best equipped to make that determination. This argument is foreclosed by our decision in Evans v. United States , where we held that "[t]he existence of a right to a jury trial depends on the maximum punishment for the offense that is charged, not on the maximum punishment for an offense that could be charged but is not." 779 A.2d 891 , 895 (D.C. 2001). 7 Mr. Coleman was ultimately charged with *1136 attempted stalking, so we look to the maximum penalty for that offense to determine whether he had a statutory right to a jury trial. According to the plain language of D.C. Code § 16-705 (b), he did not.

For these reasons, the trial court did not err in failing to afford Mr. Coleman a jury trial sua sponte .

III.

In a related argument, Mr. Coleman contends that reversal is required because the government violated Rule 7(e) of the Superior Court Rules of Criminal Procedure by filing an amended information charging attempted stalking. He acknowledges that, because he did not object to the amendment when it was filed, this claim is also subject to plain error review.

Rule 7(e) states that the trial court "may permit an information to be amended at any time before the verdict or finding," so long as no "additional or different offense is charged" and no "substantial right of the defendant is prejudiced." Super. Ct. Crim. R. 7(e). Because the amendment charged the lesser included offense of attempted stalking, 8 we need only determine whether Mr. Coleman suffered "prejudice substantial enough to reverse under the plain error standard." Jones , 124 A.3d at 132 . 9

The purpose of a criminal information is twofold: "to apprise a defendant of the charge against him so he may properly prepare a defense, and to spell out the offense clearly enough to enable the accused to plead the judgment as a bar to a subsequent prosecution for the same crime." Dyson v. United States , 485 A.2d 194 , 196 (D.C. 1984) (internal quotation marks omitted). Our cases addressing whether a defendant has shown prejudice under Rule 7(e) have examined the particular language and timing of the challenged amendment with those purposes in mind, focusing on whether the defendant had adequate notice of the new charge and whether the amendment unfairly impaired his ability to prepare and present his defense. See Sutton v. United States , 140 A.3d 1198 , 1203-05 (D.C. 2016) ; Jones , 124 A.3d at 132 ; Dyson , 485 A.2d at 197 . 10

*1137 Here, the government filed the amended information in open court more than three months before Mr. Coleman's trial eventually began. Mr. Coleman does not claim that he was surprised by the new charge or that the timing interfered with his ability to prepare a defense. Nor does he claim that he would have altered his defense in any way had he presented the case to a jury instead of a judge. The concerns typically implicated by the filing of an amended information were therefore not present in this case.

Mr. Coleman's sole argument is that he was prejudiced by the loss of his right to a jury trial because the D.C. Council expressly intended for stalking charges to be tried before a jury. The effect of the amended information, however, was to charge him with attempted stalking, and as we determined in Part II, Mr. Coleman had no constitutional or statutory right to a jury trial for that offense. To the extent that Mr. Coleman claims more broadly that any amendment that has the effect of extinguishing a jury trial right is prejudicial, that argument also must fail under a plain error standard. We have never decided whether a defendant may demonstrate prejudice under Rule 7(e) by showing only that the amendment, by charging an offense with a shorter maximum penalty, "result[ed] in his losing his right to a jury trial," see Jones , 124 A.3d at 132 n.12, and we cannot say that our cases applying Rule 7(e) obviously compel such a result. 11 The trial court thus did not plainly err by permitting the amendment despite its effect of eliminating Mr. Coleman's right to be tried by a jury.

IV.

Finally, Mr. Coleman contends that there is insufficient evidence to support his conviction for attempted stalking because the government failed to prove that he "should have known" that his conduct on October 12, 2015, would have "cause[d] a reasonable person in the [complainant's] circumstances to" fear for her safety, feel seriously alarmed, or suffer emotional distress. 12 D.C. Code § 22-3133 (a)(3). The court reviews this claim de novo , "view[ing] the evidence in the light most favorable to the government, mindful of the [factfinder's] right to determine credibility, weigh the evidence, and draw justifiable inferences of fact." Hughes v. United States , 150 A.3d 289 , 305 (D.C. 2016) (citation omitted). Although this standard of review is "deferential," it is not "toothless," and "[w]e have an obligation to take seriously the requirement that the evidence in a criminal prosecution must be strong enough that a [factfinder] behaving rationally really could find it persuasive *1138 beyond a reasonable doubt." Rivas v. United States , 783 A.2d 125 , 134 (D.C. 2001) (en banc).

Mr. Coleman's argument rests on two premises: (1) that to establish a course of conduct, the government was required to prove at least two independent instances in which Mr. Coleman possessed the "should have known" mens rea, and (2) that the government failed to do so because no reasonable factfinder could have found that Mr. Coleman "should have known" his October 12, 2015, conduct would seriously alarm a reasonable person. We examine each premise in turn.

A. Course of Conduct

A person commits the crime of stalking when he or she

purposefully engage[s] in a course of conduct directed at a specific individual: (1) With the intent to cause that individual to ... [f]eel seriously alarmed ... ; (2) That the person knows would cause that individual reasonably to ... [f]eel seriously alarmed ... ; or (3) That the person should have known would cause a reasonable person in the individual's circumstances to ... [f]eel seriously alarmed. 13

D.C. Code § 22-3133 (a). The stalking statute states that " '[t]o engage in a course of conduct' means ... on 2 or more occasions[ ] to ... [f]ollow, monitor, place under surveillance, threaten, or communicate to or about another individual" or engage in certain other statutorily specified conduct. D.C. Code § 22-3132 (8). Mr. Coleman argues that these two provisions, taken together, require proof of at least two "occasions" on which he "should have known" that a reasonable person in the complainant's position would feel "seriously alarmed." The government argues, however, that "the plain language only requires the government to prove that the entire course of conduct, which is the totality of the individual occasions, would cause those feelings" and that the defendant "reasonably should have known" that the conduct would cause those feelings.

In interpreting a statute, we begin with its text. "[I]f the statute's ... language is 'plain' and allows for no other meaning, we will generally look no further and give the 'words used the meaning ordinarily attributed to them.' " Whitfield v. United States , 99 A.3d 650 , 656 (D.C. 2014) (quoting Sullivan v. District of Columbia , 829 A.2d 221 , 224 (D.C. 2003) ). We consider "the statute's full text [and] language[,] as well as punctuation, structure, and subject matter." Hood v. United States , 28 A.3d 553 , 559 (D.C. 2011) (quoting Cook v. Edgewood Mgmt. Corp. , 825 A.2d 939 , 946 (D.C. 2003) ).

Here, the text of § 22-3133 is ambiguous as to whether the government must prove that the defendant had the requisite "should have known" mental state during at least two of the occasions that comprised the course of conduct. The subsection that sets forth the "should have known" mental state, D.C. Code § 22-3133 (a)(3), begins with the word "[t]hat" and has the form of a restrictive clause, but it is not clear what word or phrase this restrictive clause modifies: it could be read as modifying either "course of conduct" or just "conduct" in subsection (a). The former reading (the one the government prefers) would require only proof that the defendant possessed the requisite mens rea with respect to the course of conduct as a whole, but the latter reading (the one Mr. Coleman prefers) would seem to require proof that the defendant possessed the requisite mens rea when committing the "conduct"-that is, the "occasions" or *1139 acts that comprised the course of conduct. 14

Reading § 22-3133 in conjunction with § 22-3132, the "Definitions" section, reduces this ambiguity in favor of Mr. Coleman's interpretation. See Eaglin v. District of Columbia , 123 A.3d 953 , 956 (D.C. 2015) ("[S]tatutory interpretation is a holistic endeavor, and, at a minimum, must account for a statute's full text, language as well as punctuation, structure, and subject matter." (internal quotation marks omitted) ). The phrase "to engage in a course of conduct," as relevant here, means " on 2 more occasions , to ... [f]ollow, monitor, place under surveillance, threaten, or communicate to or about another individual[.]" D.C. Code § 22-3132 (8) (emphasis added). 15 When the definition of "to engage in a course of conduct" is inserted into D.C. Code § 22-3133 (a)(1), for example, the statute makes it unlawful for a person, on two or more occasions, to follow, monitor, place under surveillance, threaten, or communicate to or about another individual, with the intent to cause that individual to experience fear, serious alarm, or emotional distress. Substituting "to engage in a course of conduct" with its statutory definition makes clear, at least with respect to intentional stalking, that the government must prove both the act and the intent for at least two separate incidents. While the same substitution reads less smoothly with respect to the knowing and "should have known" mental states, it would not make sense to interpret the act element of the statute any differently for subsections (a)(2) and (a)(3). Intentional stalking differs from knowing and "should have known" stalking only in terms of the mental state required; structurally they are the same, and the "course of conduct" language is shared by all three. In sum, when read together with § 22-3132(8), the stalking statute appears to support Mr. Coleman's reading in that it prohibits a person from, on two or more occasions , doing a particular thing while possessing a particular mental state.

Other aids to construction, including applicable canons of statutory interpretation and legislative history, see District of Columbia v. Reid , 104 A.3d 859 , 868 (D.C. 2014), also support Mr. Coleman's interpretation. Under the last-antecedent rule, "qualifying words and phrases, where no contrary intention appears, [are typically understood as] refer[ring] solely to ... the last word, phrase, or clause that can be made an antecedent without impairing the meaning of the sentence." 2A Norman Singer & Shambie Singer, Sutherland Statutes and Statutory Construction § 47.33 (7th ed. 2016) (internal quotation marks omitted); see also Barnhart v. Thomas , 540 U.S. 20 , 26, 124 S.Ct. 376 , 157 L.Ed.2d 333 (2003) ; Richman Towers Tenants' Ass'n, Inc. v. Richman Towers LLC , 17 A.3d 590 , 614 (D.C. 2011). Here, "conduct"-not *1140 "course of conduct"-is the last word or phrase in D.C. Code § 22-3133 (a) that can fairly be understood as the antecedent of the restrictive clause in § 22-3133(a)(3). The last-antecedent rule thus suggests that the requisite mens rea must be proved with respect to the conduct (the "occasions" or acts) comprising the course of conduct, not merely with respect to the course of conduct as a whole.

The legislative history of the stalking statute also supports requiring the government to prove that a defendant possessed the requisite mental state on at least two occasions. The Committee on Public Safety and the Judiciary stated that the purpose of revising the stalking statute was "to enable the justice system's intervention before stalking escalates into violence-yet avoid inadvertent criminalization of legal behaviors." Committee Report at 32. The Committee's intent to avoid inadvertent criminalization could best be implemented by requiring that the defendant possess a culpable mental state when he or she commits at least two of the acts that constitute the course of conduct. Without this requirement, ordinarily innocuous acts like "communicat[ing]" with the complainant could in the aggregate constitute a "course of conduct." Indeed, in United States v. Smith , this court, considering an earlier version of the stalking statute, 16 noted that there would be "potential vagueness problems" if the statute were read to require only a single act of "conduct with the intent to cause emotional distress in another person." 685 A.2d 380 , 384 (D.C. 1996). The court avoided the constitutional issue by reading the statute to require that "the elements of 'willfully, maliciously [,] and repeatedly following or harassing' " be proved for "emotional distress" stalking. Id. (emphasis added); see Zadvydas v. Davis, 533 U.S. 678 , 689, 121 S.Ct. 2491 , 150 L.Ed.2d 653 (2001) (" 'It is a cardinal principle' of statutory interpretation, however, that when an Act of Congress raises 'a serious doubt' as to its constitutionality, 'this Court will first ascertain whether a construction of the statute is fairly possible by which the question may be avoided.' " (quoting Crowell v. Benson , 285 U.S. 22 , 62, 52 S.Ct. 285 , 76 L.Ed. 598 (1932) ) (brackets removed) ).

Reading the statute to require that the defendant possess the prohibited mental state during at least two occasions does not undermine the Council's other stated objective-to enable early "intervention before stalking escalates into violence," Committee Report at 32-for two reasons. 17 First, the most dangerous kinds of isolated acts will often satisfy the elements of other offenses. See, e.g. , D.C. Code § 22-404 (assault); id. § 22-407 (threats to do bodily harm); id. § 22-1810 (threats to kidnap or injure a person or damage his or her property). The government is thus able to intervene even in many situations in which it is unable to prove a two-occasion course of conduct. 18

*1141 And second, a reasonable factfinder can certainly consider all of the previous acts in a defendant's course of conduct in assessing whether he or she possessed the requisite mental state when he or she committed the two acts of following, monitoring, surveilling, threatening, or communicating that the government says support a conviction for stalking. Where a defendant has committed a series of alarming acts, the defendant's prior acts will usually justify a conclusion that he or she should have known that at least two of the acts would cause a reasonable person to feel seriously alarmed.

The rule of lenity provides another interpretive resource that tips the balance in favor of Mr. Coleman's interpretation of the statute. The rule of lenity states that "criminal statutes should be strictly construed and that ambiguities should be resolved in favor of the defendant." Whitfield , 99 A.3d at 656 (internal quotation marks omitted). To the extent that the statutory definitions, the last-antecedent rule, and the legislative history fail to decisively establish that the legislature intended to require proof that the defendant possessed the requisite mental state on at least two occasions, the rule of lenity requires that we resolve any remaining ambiguity in favor of that interpretation. See Holloway v. United States , 951 A.2d 59 , 65 (D.C. 2008) ("The rule of lenity [ ] can tip the balance in favor of criminal defendants only where, exclusive of the rule, a penal statute's language, structure, purpose and legislative history leaves its meaning genuinely in doubt." (internal quotation marks omitted) ).

Finally, we note that although it does not appear that this court has previously addressed the question currently before us with respect to the current version of the stalking statute, the court's discussion in Whylie v. United States , 98 A.3d 156 (D.C. 2014), suggests that the court accepted Mr. Coleman's favored interpretation as a given. See id. at 163 n.7 ("[B]oth the current and the predecessor statute criminalized a 'course of conduct' made up of multiple acts of harassing or threatening behavior ." (emphasis added) ). Moreover, as stated above, the court understood the previous version of the stalking statute to require repeated acts of willful and malicious behavior. See Smith , 685 A.2d at 384-86 . Given that "nothing in the legislative history of the current statute indicates an intent by the Council ... to change the substantive meaning of the term 'course of conduct,' " Whylie , 98 A.3d at 163 n.7, there is good reason to think that a similar requirement applies under the current version of the statute. Other state courts have interpreted their comparable stalking statutes similarly. See, e.g. , Hawes v. State , 335 P.3d 1073 , 1077 (Wyo. 2014) (accepting the state's "conce[ssion] that there can be no 'course of conduct' if there was no intent to harass" in the first of two incidents); Morgan v. Commonwealth , 189 S.W.3d 99 , 112 (Ky. 2006) (finding that the state failed to prove a course of conduct where the first of two incidents could not have caused mental *1142 distress due to the complainant's unawareness of the defendant's entry into her home during that incident), overruled on unrelated grounds by Shane v. Commonwealth , 243 S.W.3d 336 (Ky. 2007). 19

In sum, although the text of D.C. Code § 22-3133 is ambiguous as to whether a defendant can be convicted of stalking absent proof that he or she possessed a culpable mental state (an intentional, knowing, or "should have known" mental state) during at least two of the occurrences that comprise the course of conduct, the statutory definitions, the last-antecedent rule, the legislative history, and the rule of lenity lead us to conclude that the defendant cannot. We are unable to discern from the transcript of the proceedings below whether the trial court applied this requirement in reaching its guilty verdict. A remand is therefore required so that the trial court may consider the evidence under the appropriate legal standard. See Carrell v. United States , 165 A.3d 314 , 326-28 (D.C. 2017) (en banc).

*1143 B. Mental Element

Although we have already determined that a remand is in order, we nevertheless address Mr. Coleman's sufficiency challenge because a determination that the government presented insufficient evidence of stalking would obviate the proceedings on remand. Mr. Coleman was convicted of attempted stalking based on the theory that he "should have known" that his conduct "would cause a reasonable person in the [complainant's] circumstances to" fear for her safety, feel seriously alarmed, disturbed, or frightened, or suffer emotional distress. D.C. Code § 22-3133 (a)(3). As explained above, the government was required to prove beyond a reasonable doubt that Mr. Coleman "should have known," during at least two of the occurrences that comprised his course of conduct, that his conduct would produce such mental harm. Mr. Coleman does not seriously contest that there was sufficient evidence that he had the requisite "should have known" mental state during the final, October 26, 2015, occasion, but he argues that there was insufficient evidence that he possessed the requisite mental state on October 12. 20

As a threshold matter, the parties dispute whether the "should have known" language refers to the defendant's subjective mental state or to what an objectively reasonable person in the defendant's position would have known. Mr. Coleman argues that the standard is a subjective one and points to Owens v. United States , in which our court understood statutory language allowing conviction for receiving stolen property when the defendant "ha[s] reason to believe" that the property is stolen as referring to the defendant's actual, subjective mental state. 90 A.3d 1118 , 1123 (D.C. 2014). The government responds that Owens does not control here due to the different statutory language, arguing that the mens rea element for "should have known" stalking is a completely objective one.

We agree with the government that "should have known" is an objective standard. In Owens , the court recognized that although the D.C. Council "plainly" intended to "reach beyond actual knowledge," a subjective standard would "give effect to this legislative purpose without permitting a jury to convict a defendant for mere negligence," which " ordinarily [does not] form[ ] the basis of felony liability." Id. at 1121 (emphasis added) (quoting DiGiovanni v. United States , 580 A.2d 123 , 126 (D.C. 1990) (Steadman, J., concurring) ). Because the phrase "having reason to believe" is susceptible to a subjective interpretation-in which jurors would inquire whether the defendant actually possessed information that would suggest to him that the property might be stolen (even if he did not know for sure)-this discussion in Owens merely reflects courts' longstanding reluctance to read a negligence standard into a criminal statute in the absence of "a clear statement from the legislature." Carrell , 165 A.3d at 320 (citing Elonis v. United States , --- U.S. ----, 135 S.Ct. 2001 , 2011, 192 L.Ed.2d 1 (2015) ). A "should have known" standard, on the other hand, is necessarily objective; when applying such a standard, we assume that the defendant did not know a particular thing, and we determine whether he should have known that thing by reference to whether someone else (a reasonable person) who is aware of the same facts and circumstances as the defendant would have known it. 21 The "should have known"

*1144 language represents just the type of clear legislative statement not present in Owens , and it evinces the Council's intent to allow a conviction for stalking based on what an objectively reasonable person would have known. 22

In addressing whether a reasonable factfinder could conclude that Mr. Coleman should have known that his behavior during the two staring incidents and the October 12 encounter would be seriously alarming to a reasonable person, we analyze the statutory language describing the mental harm that a reasonable person in the complainant's position must experience to trigger criminal liability. In interpreting this language, we rely first and foremost on the text of the statute itself, but also on the legislative history and the Model Stalking Code Commentary. See Nat'l Ctr. for the Victims of Crime, The Model Stalking Code Revisited: Responding to the New Realities of Stalking (2007), http://victimsofcrime.org/docs/default-source/src/model-stalking-code.pdf.

The first type of mental harm listed in the stalking statute, "[f]ear for ... safety," is not defined in the statute. D.C. Code § 22-3133 (a)(3)(A). The legislative-intent section of the stalking statute, however, states that the law was designed to prevent "severe intrusions on [an individual's] personal privacy and autonomy" and conduct that "creates risk to the security and safety of the [individual]." D.C. Code § 22-3131 (a) ; see also Committee Report at 33 ("[T]he purpose [of the law] is to enable law enforcement to intercept behaviors that potentially lead to violence, a loss in the quality of life, or even death."). Further, the Model Stalking Code Commentary provides the following examples of fears that would constitute fear for one's safety: "fear of death or serious physical harm," fear of sexual assault, fear that a child will be kidnapped or harmed, and "[f]ear of the unknown." Model Stalking Code Commentary at 39-40. Together, these sources indicate that fear for safety means fear of significant injury or a comparable harm. Moreover, they indicate that the stalking statute is meant to prohibit seriously troubling conduct, not mere unpleasant or mildly worrying encounters that occur on a regular basis in any community.

The statute expressly defines "emotional distress," another form of mental harm listed in the statute, as "significant mental suffering or distress that may, but does not necessarily, require medical or other professional treatment or counseling." D.C. Code § 21-3132 (4). This language indicates that the type of emotional distress that the victim must experience is high, reaching a level that would possibly *1145 lead to seeking professional treatment. See also Committee Report at 32 ("Stalking is a serious crime that often involves intimidation, psychological terror, and escalating severity."). The Model Stalking Code Commentary cites with approval Wallace v. Van Pelt , in which the court explained that "emotional distress" was "a general or specific feeling of mental anguish ," "something markedly greater than the level of uneasiness, nervousness, unhappiness or the like which [is] commonly experienced in day to day living." 969 S.W.2d 380 , 386 (Mo. Ct. App. 1998) (emphases added), cited by Model Stalking Code Commentary at 49. Further, the Model Stalking Code Commentary sets forth the following examples of conduct that would cause "emotional distress": "making repeated telephone calls to a victim at a workplace, possibly endangering her job, ... engaging in conduct that destroys the victim's credit history," and "plac[ing] [the victim] under constant surveillance." Model Stalking Code Commentary at 41, 49.

There is much more limited guidance available about what it means to "feel seriously alarmed, disturbed, or frightened," the remaining form of mental harm listed in the statute. D.C. Code § 22-3133 (a)(3)(B). But the D.C. Council's removal, in the current version of the statute, of liability for "seriously ... annoy[ing]" conduct tells us that serious annoyance is insufficient. D.C. Code § 22-404 (e) (2001) (defining "harass[ment]"-one of the ways in which a person could commit the crime of stalking under the old statute-as conduct that "seriously alarms, annoys, frightens, or torments"); see also District of Columbia Public Defender Service, June 2, 2009, Letter to Councilmember Phil Mendelson, attachment to Committee Report, at 3 (explaining that the United States Attorney's Office and the Office of Attorney General had "propos[ed] ... replacing 'annoy' with 'disturb' " and that the Public Defender Service "prefer[ed] 'disturb' to 'annoy' because [it thought it] conveys a more serious effect"); Model Stalking Code Commentary at 39 ("[T]he stalking conduct needs to address behavior that goes beyond merely annoying the victim ...."). And the principle of noscitur a sociis suggests that serious alarm, disturbance, and fright should be understood as mental harms comparable to fear for one's safety or significant emotional distress. See Burke v. Groover, Christie & Merritt, P.C. , 26 A.3d 292 , 303 n.8 (D.C. 2011) (" 'The maxim noscitur a sociis , that a word [or phrase] is known by the company it keeps, while not an inescapable rule, is often wisely applied where a word [or phrase] is capable of many meanings in order to avoid the giving of unintended breadth' to words in a statute." (quoting Jarecki v. G.D. Searle & Co. , 367 U.S. 303 , 307, 81 S.Ct. 1579 , 6 L.Ed.2d 859 (1961) ) (alterations in original) ).

To sum up, while it would almost certainly distort the meaning of the statute to attempt to synthesize an all-encompassing rule or principle about the three forms of mental harms, it is clear that to trigger criminal liability, the level of fear, alarm, or emotional distress must rise significantly above that "which [is] commonly experienced in day to day living," Wallace , 969 S.W.2d at 386 , and must involve a "severe[ ] intrusion on the victim's personal privacy and autonomy," D.C. Code § 22-3131 (a). Ordinary "uneasiness, nervousness, [and] unhappiness" are insufficient. Wallace , 969 S.W.2d at 386 . Turning to Mr. Coleman's claim, then, we examine whether a rational factfinder could find beyond a reasonable doubt that Mr. Coleman should have known that his conduct during at least one encounter prior to October 26 would have been seriously alarming, in the way explained *1146 above, to a reasonable person in the complainant's position.

Whether the government presented sufficient evidence to support such a finding regarding the October 12 interaction is a close question, but ultimately we conclude that the evidence was sufficient. A reasonable factfinder could have found, with respect to that encounter, that after Mr. Coleman expressed a romantic interest in the complainant and she rebuffed him, Mr. Coleman continued to linger around the baseball diamond and the bus stop in order to watch her as she walked. 23 A reasonable factfinder could further infer from the complainant's testimony-specifically, that her husband told her Mr. Coleman now understood that "he [didn't] need to be out [t]here harassing people"-that the complainant's husband in fact informed Mr. Coleman that his actions were scaring the complainant and asked him to stop. Having been asked twice in a short period of time to stop bothering the complainant, Mr. Coleman then "yelled out" to the complainant as she left his group home, accused her of "tr[ying] to get [him] in trouble," and denied being someone who "[went] around masturbating." In the context of the two prior staring incidents, the fact that Mr. Coleman knew where the complainant lived, the early morning hour, and the two unequivocal requests that he leave the complainant alone, a factfinder could reasonably conclude that Mr. Coleman should have known that his behavior on October 12 would be seriously alarming to a reasonable person in the complainant's position. 24

Because a rational factfinder could potentially conclude that Mr. Coleman possessed the requisite mental state on at least two of the four occasions in which he encountered the complainant, we decline to reverse his attempted stalking conviction on the basis of insufficient evidence.

V.

For the foregoing reasons, we vacate Mr. Coleman's conviction for attempted stalking and remand to allow the trial court to evaluate the evidence in light of the principles discussed in this opinion. See Carrell , 165 A.3d at 326-28 (remanding for *1147 application of correct legal standard despite conclusion that the evidence was sufficient to support a conviction under the correct standard). 25

Blackburne-Rigsby, Chief Judge, dissenting in the judgment in part, concurring in part:

The anti-stalking statute was amended by the Council of the District of Columbia based on an evolving understanding and recognition of the unique nature of the offense of stalking. In many cases, stalking consists of individual acts of non-overt threatening behavior that, when viewed as a whole "course of conduct," evidence a clear intent to alarm, disturb, or frighten the victim. 1

In this case, appellant Johnnie Coleman's conviction for stalking stems from two occasions in which he engaged in conduct towards the complainant that caused her to fear for her safety or to feel seriously alarmed. Specifically, on October 12, 2015, the complainant was alone walking to the Lamond Recreation Center, when suddenly appellant appeared before her and sought to engage her in conversation. Appellant told her that she looked "nice" and asked if he could talk to her. The complainant rebuffed appellant's advances and made clear that she was "not trying to talk to" him. Nevertheless, the complainant soon noticed appellant sitting on a nearby bench staring at her as she took her early-morning walk. The complainant testified that, because of her earlier encounter with appellant and because no one was around so early in the morning, she felt "very uncomfortable" and scared that he might do her "some harm." After finishing her walk, the complainant reported appellant's behavior to the group home where he lived. As she began walking home, appellant again approached her and yelled disturbing things at her, including "I see you tried to get me in trouble. I'm not a bad person.... I read the Bible. It's not like I go around masturbating ...." On the second occasion, on October 26, the complainant was again walking around the Lamond Recreation Center when she saw appellant suddenly run after her in what she believed to be a menacing manner. Despite the complainant telling him to leave her alone, appellant continued chasing her and yelled profanities at her, such as "F you bitch." The complainant ran inside her home as appellant stood outside her home shaking a street sign. The complainant testified that this was a "traumatic experience," and that she had lost sleep.

In my view, the government met its burden here in proving that appellant "should have known" that his "course of conduct" towards the complainant would have caused a reasonable person to fear for her safety or feel alarmed, disturbed, or frightened. 2

The District's anti-stalking statute is not ambiguous, nor is it susceptible to multiple *1148 interpretations that are plausible. Further, the majority's holding ignores the intent of the D.C. Council when it rewrote the anti-stalking statute in 2009 to broaden its reach and to "encourage effective intervention ... before stalking escalates into behavior that has even more serious or lethal consequences." 3 Finally, the majority also overlooks large sections of the Model Stalking Code upon which the District's revised anti-stalking statute is based, and disregards persuasive case law from other jurisdictions that have contemplated and rejected the majority's interpretation. Therefore, I would affirm appellant's conviction for attempted stalking and I respectfully dissent from the majority's interpretation of the anti-stalking statute and decision to remand.

I. Discussion

A.

The language of the anti-stalking law is unambiguous and clear. Under well-established canons of statutory interpretation, we first look to the language of the statute to see if it is plain and unambiguous to determine the intent of the legislature. However, we read the statutory language holistically, considering the statute's full text and subject matter. We must also consider the plain meaning of the statute in the context of its stated policies and objectives. 4

Preliminarily, D.C. Code § 22-3132 (8) (2012 Repl.) gives a definition of "[t]o engage in a course of conduct" as meaning to engage in prohibited acts "on 2 or more occasions." The statute does not give a definition of "conduct" alone. Consequently, the fact that the D.C. Council specifically defined "course of conduct" as a term of art, but did not define "conduct," is powerful evidence that the Council meant what it said, and that it intended for the mens rea element to modify "course of conduct" as a whole, rather than just "conduct." Moreover, the term "course of conduct" is a well-understood principle that is used in many criminal contexts. For example, in the context of rape, we have said that "[a] continuing course of conduct" is "a single continuous episode rather than [ ] a series of individually chargeable acts," even if "there has been some special or temporal separation." Gray v. United States , 544 A.2d 1255 , 1258 (D.C. 1988) (citation omitted). In fact, in looking at the prior version of the anti-stalking statute, this court observed that "it is the continuing course of conduct which constitutes the offense, not the individual discrete actions making up the course of conduct." Washington v. United States , 760 A.2d 187 , 198 (D.C. 2000).

In reaching its conclusion that the statute is ambiguous, the majority first relies on a somewhat tortured reading of the language in D.C. Code § 22-3133 (a)(3) and specifically the word "[t]hat," which begins the subsection. The majority contends that because "[t]hat" has the form of a restrictive clause, it is unclear which word or phrase the "[t]hat" under subsection (a)(3) was intended to modify, either "course of conduct" or just "conduct." Ante at 1138. However, when all parts of the statute are read together as a whole, the statute is clear. Upon reading subsection (a)(1), the initial word "[w]ith" is a preposition that is intended to show the relationship between the defendant's mental state and the actus reus. Using the word "with," subsection *1149 (a)(1) cannot be read as simply modifying the word "conduct" because that would render superfluous the previous phrase "in a course of." 5 We strive to avoid interpretations of a statute that render any part of the statutory language superfluous, as the majority's interpretation would do. Interpreting the mens rea element as modifying "course of conduct" as a whole, however, does not render any portion of the statute ambiguous.

Alternatively, the majority claims that swapping and replacing a phrase from D.C. Code § 22-3133 , with the definition of "to engage in a course of conduct" found under D.C. Code § 22-3132 (8), would favor appellant's reading that a defendant must have the requisite mental state "on two or more occasions" and thereby supports the majority's interpretation. Ante at 1139. This argument further underscores my point that the majority's interpretation of the statute is unnecessarily tortured and cumbersome. Swapping and replacing the phrases between these two different provisions of the statute results in a contrived and strained reading of two other sections of the statute which describe the different types of stalking - "knows would cause" (§ 3133(a)(2) ) and "should have known would cause" (§ 3133(a)(3) ) type stalking. A plain reading of the statute, as I propose, does not strain beyond recognition subsections 3133(a)(2) and (a)(3). This cumbersome interpretation of the statute is not consistent with a plain reading of the statute. See Peoples Drug Stores, Inc. , supra note 4, 470 A.2d at 753. Moreover, if the D.C. Council intended to re-write the statute by swapping and replacing language from different sections, the Council would have done so. No part of the statute is strained or rendered superfluous if the statute is read as having the mental state modifying "course of conduct" as a whole.

The majority's interpretation also results in another anomaly that the D.C. Council could not have intended. D.C. Code § 22-3133 lays out three mental states that the government could charge in proving stalking: (1) "[w]ith the intent to cause"; (2) "knows would cause"; and (3) "should have known would cause." Id. § (a)(1)-(a)(3). The legislative history of the statute reveals that the D.C. Council envisioned "three types of stalking that may occur: when the stalker intends to stalk; when the stalker knows that a specific individual will feel stalked; and when the stalker should have known they would feel stalked." Committee on Public Safety and the Judiciary, Report on Bill 18-151, at 34 (Jun. 26, 2009). The D.C. Council further explained that the "only situation where knowledge is relevant is in the second scenario ... [and that] it is fundamental that the defendant be able to argue that he or she did not know that their actions would cause that specific individual to feel frightened." Id. For example, in this case, appellant was charged under subsection (a)(3) because he "should have known" that his actions would cause a reasonable person to fear for her safety. Under the majority's holding, these different "types of stalking" that the D.C. Council envisioned might be rendered inapplicable because the government must prove the mens rea for individual "occasions." There will surely be anomalies where the government will seek to prove "should have known" on one occasion, but "knows would cause" on another occasion. This defeats the D.C. Council's intent of prescribing three different "types" of stalking. It also *1150 unnecessarily injects confusion at trial because appellant would be able to assert a knowledge defense for one "occasion" of stalking, but not another.

Consequently, the majority's subsequent reliance on other secondary canons of statutory construction such as the last-antecedent rule or the rule of lenity have no place in the statutory analysis if their alternative reading of the statute fails under a plain language analysis. Where there is a clear and unambiguous interpretation of the statutory language that does not render any parts of the statute superfluous, the rule of lenity does not apply. "[T]he rule of lenity is a secondary canon of construction, and is to be invoked only where the statutory language, structure, purpose and history leave the intent of the legislature in genuine doubt." Cullen v. United States , 886 A.2d 870 , 874 (D.C. 2005) (citation and internal quotation marks omitted).

B.

The legislative history of the anti-stalking statute does not support the majority's holding that the D.C. Council intended to require proof of the requisite mental state on at least two individual "occasions" that make up the course of conduct. The majority's interpretation ignores key aspects of the D.C. Council's intentions as evidenced in the legislative history. In the Committee on Public Safety and the Judiciary's ("Committee") report, the Committee expressly stated that, "If [the jurors] are able to unanimously agree that overall conduct equates to stalking beyond a reasonable doubt, it seems overly burdensome to require that they agree upon the same acts." Report on Bill 18-151, supra , at 34 (emphasis added). The Committee's choice of words ("overall conduct equates to stalking") in this sentence makes clear its intention for the mens rea element to apply to the "course of conduct," i.e., "overall conduct," as a whole.

The majority states that its interpretation is consistent with the D.C. Council's intent because one of the purposes of revising the anti-stalking law is to "avoid inadvertent criminalization of legal behaviors." Ante at 1140. Yet, the majority does not explain how applying the mens rea element to "course of conduct" as a whole would more likely criminalize innocuous behavior than their preferred interpretation. In my view, quite the opposite would be true. By looking at the totality of the circumstances, the factfinder is able to better gauge whether the defendant's actions were truly innocuous, or whether he or she intended to conduct an escalating pattern of stalking behavior. For example, the majority contends that, under the government's interpretation, innocuous acts like "communicating" with the complainant could in the aggregate constitute a "course of conduct," subject to prosecution. Ante at 1140-41. But unwanted communications with the complainant, if done with the intent to alarm, disturb, or frighten, can constitute stalking under the statute. 6 There is nothing "innocuous" about such behavior when it harasses or alarms the victim in violation of the statute. In fact, the D.C. Council recognized this view in the Committee Report. The Committee gave the following hypothetical:

A familiar example of [stalking] is when a man and a woman go on a date. After the date, the man is interested and repeatedly contacts the woman for another *1151 date. The woman meanwhile is not interested and does not respond to his communications. At what point does the man's conduct become harassing to that woman? Annoying? Alarming? Disturbing? The answer is not found in a bright line distinction between strict definitions for acceptable and alarming.

Report on Bill 18-151, supra , at 32-33. The majority's interpretation seeks to inject a bright line distinction that the Committee found did not exist. Under the majority's interpretation of this hypothetical, the factfinder must gauge whether the man had the requisite mental state on the first call, the second call, the third call, etc. That kind of analysis, on its face, goes against the D.C. Council's intention that there is no "bright line" rule as to what is or is not acceptable behavior. Rather, the statute requires that there must be a "course of conduct," that is intended to harass or alarm, which must be evidenced by a review of the totality-of-the-circumstances.

The D.C. Council rewrote the District's anti-stalking statute as part of the Omnibus Public Safety and Justice Amendment Act of 2009 to conform to best practices as endorsed by the revised Model Stalking Code. See The National Center for Victims of Crime, The Model Stalking Code Revisited, Responding to the New Realities of Stalking (2007). The Committee Report explained that the D.C. Council sought to rewrite the anti-stalking statute to clear up any confusion in the statute and to expand the conduct that could be considered stalking behavior, including the use of technology such as the internet, email, and GPS. Report on Bill 18-151, supra , at 32.

The Committee explained that "the challenge is to enable the justice system's intervention before stalking escalates into violence - yet avoid inadvertent criminalization of legal behaviors." Id. These considerations stem from concerns expressed in the revised Model Stalking Code on the limitations of previous anti-stalking legislative attempts. In particular, the Model Stalking Code referenced studies by the National Center for Victims of Crime showing that "[f]ewer than 50 percent of victims reported being directly threatened by their stalkers," and that the "variability of stalking behaviors suggests that laws must be broad enough to address stalking in all its forms." Model Stalking Code , supra , at 14-15. The Model Stalking Code also made the following observations, among others, about then-existing anti-stalking legislation: "[t]he burden of proof is so high under many stalking laws that it is extremely difficult to secure convictions" and "[w]ithout a full appreciation of the role of context in a stalking situation ... many stalking behaviors can be viewed as harmless, when in fact the behaviors may terrify the victim ...." Id. at 17 (emphasis in original).

The Council viewed the role of "context" as critically important to a determination of whether a course of conduct is intended to harass or harm. Interpreting the mens rea element as modifying "course of conduct" as a whole, rather than individual acts, would fully support the policy goals of the District's anti-stalking legislation. In particular, reading the statute in this way is consistent with the D.C. Council's intent that intervention occur as soon as possible "before stalking escalates into" more serious behavior.

For example, consider the following hypothetical where there are two alleged "occasions." The first occasion occurred when the defendant approached the complainant at a nightclub and sought to hit on her. Although the complainant rebuffed the defendant, because he was a stranger, she did not necessarily feel alarmed by this first chance encounter. The second occasion occurs when, a few days later, the *1152 complainant notices the defendant following her around as she walked down the street. The complainant now feels alarmed because she recalls that this is the same guy who approached her at the club. Under the majority's view, the victim has to wait until a third incident before the government can prove the requisite mens rea on two occasions. This creates a perverse "one free stalk" situation that is clearly not what the D.C. Council and the Model Stalking Code envisioned in revising anti-stalking legislation and is contrary to the plain language of the statute, which states that stalking requires two occasions, not three.

The majority's rationale that dangerous isolated acts could possibly be prosecuted independently under other statutes is of little comfort to the majority of stalking victims. As the majority states, the factfinder can consider any previous act in determining whether the defendant possessed the requisite mens rea for a subsequent act, but it will be exceedingly difficult for the factfinder to determine the mens rea for the first act under the majority's interpretation.

The legislative history demonstrates that the D.C. Council intended for the requisite mental state to modify "course of conduct" as a whole. The D.C. Council understood that it is the overall conduct of the defendant, within context, that must result in harassing or alarming behavior in order to constitute stalking. The D.C. Council essentially considered and rejected the majority's view that the mens rea element should apply to each act as an independent element of the offense, rather than to the course of conduct as a whole. 7

C.

There is persuasive case law from other jurisdictions that have grappled with similar anti-stalking statutes, where courts have concluded that the mens rea element modifies "course of conduct" as a whole, rather than individual acts of stalking. In United States v. Shrader , 675 F.3d 300 , 311 (4th Cir. 2012), the Fourth Circuit considered an issue similar to the one before us. The defendant, Shrader, essentially made the same argument as appellant here. He argued that the federal stalking statute, 8 which shares many similarities with the District's statute, was unconstitutionally vague because it did not define "whether all acts included in the prohibited course of conduct must be done with the specific intent to cause harm required by the statute." Id. at 311 . The Fourth Circuit rejected this argument completely. The court stated that:

[T]he statute does not impose a requirement that the government prove that each act was intended in isolation to cause serious distress or fear of bodily injury to the victim, [instead] the government is required to show that the totality of the defendant's conduct "evidenc[ed]

*1153 a continuity of purpose" to achieve the criminal end. The specific intent requirement thus modifies the cumulative course of conduct as a whole.

Id. (emphasis added) (third brackets in original). The Fourth Circuit went on to observe that this interpretation makes sense because it "reflects a clear understanding on the part of Congress that while severe emotional distress can of course be the result of discrete traumatic acts, the persistent efforts of a disturbed harasser over a period of time ... can be equally or even more injurious." Id. at 311-12 (emphasis added). And that, "[t]o read in a requirement that each act have its own specific intent element would undo the law's protection for victims whose anguish is the result of persistent or repetitive conduct on the part of a harasser." Id. at 312 . The Fourth Circuit's statutory interpretation is therefore very persuasive. The federal and District statutes use the same terms of art, "course of conduct," and seek to protect the same interests.

Similarly, Washington State's Court of Appeals, in looking at Washington State's stalking statute, 9 which is similar to the District's, considered whether each individual act making up a course of conduct must independently constitute "harassment." See State v. Haines , 151 Wash.App. 428 , 213 P.3d 602 , 606 (2009). The court held that, "There is no basis whatsoever to suppose that each of the separate acts that comprise that course of conduct must be vexatious when taken in isolation." Id. at 606 . Instead, the court observed that "[i]t is the combination of separate acts - none of which is necessarily criminal in its own right - that must be 'seriously alarming, annoying, harassing, or detrimental to the victim in order for the perpetrator to have committed the criminal offense of stalking.' " Id. (original brackets omitted); see also State v. Kintz , 169 Wash.2d 537 , 238 P.3d 470 , 479 (2010) (en banc) (approving the observations made in Haines ). 10 Although both Shrader and Haines were interpreting somewhat differently worded statutes, the courts in both cases understood that the crime of stalking is a unique offense that requires looking at the course of conduct as a whole.

II. Conclusion

For these reasons, I must respectfully dissent from the majority's interpretation *1154 of the District's anti-stalking statute. There are real-life consequences between our differing views on the elements of stalking. There is no doubt that the context in which a "course of conduct" occurs is vitally important to determining whether the offense of stalking has occurred. The majority's interpretation makes it more difficult to convict a defendant of stalking. Fundamentally, stalking is a crime of repetition and parsing out the defendant's intent behind specific individual acts makes it more likely that a defendant can explain away as innocuous individual incidents that, when viewed together, demonstrate a disturbing pattern of harassing and alarming actions. The anti-stalking statute is clear and unambiguous. The D.C. Council enacted the 2009 amendments to the District's anti-stalking statute to make it easier, not more difficult, to convict defendants of stalking and to prevent further escalation of a defendant's harassing and alarming conduct.