The decision of the Court of Appeals is affirmed. The judgment of the circuit court is reversed, and the case is remanded to the circuit court for further proceedings.
Balmer, C. J., dissented and filed an opinion, in which Landau, J., joined.*61BREWER, J.
In this negligence action, plaintiff, the personal representative of the estate of Martha Delgado, alleged in her complaint1 that Delgado — a foreign exchange student staying in this country under the supervision of defendants Rotary International and Rotary International District 5100 (the Rotary defendants) — was shot and killed by an assailant while standing in line on a public sidewalk outside the Zone, a teenage nightclub in Portland owned by several business entities (the Zone defendants).2
On defendants’ motion, the trial court dismissed plaintiff’s complaint on the ground that plaintiff had failed to state facts sufficient to constitute a claim for relief with respect to the issue whether Delgado’s death was a foreseeable result of defendants’ conduct. See ORCP 21 A(8) (providing for dismissal for “failure to state ultimate facts sufficient to constitute a claim”). A divided panel of the Court of Appeals reversed the ensuing judgment dismissing this action. Piazza v. Kellim, 271 Or App 490, 354 P3d 698 (2015). On review, we conclude that plaintiff alleged facts that — if proved — were sufficient to permit a reasonable juror to find that Delgado’s death was a reasonably foreseeable result of defendants’ conduct. Accordingly, we affirm the decision of the Court of Appeals, and we reverse the judgment dismissing this action and remand to the trial court.
I. FACTS AND PROCEDURAL HISTORY
A. The Complaint
Because this is an appeal from a trial court judgment dismissing a complaint under ORCP 21 A(8), we assume that all well-pleaded facts are true and give plaintiff the benefit of all favorable inferences that reasonably may be drawn from those factual allegations. See, e.g., Caba v. Barker, 341 Or 534, 536, 145 P3d 174 (2006) (stating standard of review). Plaintiff alleged that Delgado, age 17, was staying with a host family in White Salmon, Washington, as *62part of an international exchange program sponsored and managed by the Rotary defendants. On January 24, 2009, a group of approximately 14 Rotary exchange students, including Delgado, gathered at the home of the host parents of one of the students for a birthday celebration. Later in the evening, the host parents drove the group to the Zone. The Zone was an underage nightclub that admitted people ages 16 to 21 and featured dancing. The Zone was located in the “Old Town/Chinatown neighborhood” bordering downtown Portland; it also was part of what is referred to by police as the “downtown entertainment district,” which consists of several streets where nightclubs and bars are located. The Zone frequently had a “cordoned-off line of customers on the sidewalk outside the club waiting to get in.” The host parents dropped the students off near the Zone without a chaperone and planned to pick them up at 1:00 a.m.
Delgado and the other exchange students were waiting on the sidewalk to get into the Zone when an assailant, Ayala, shot Delgado twice. Delgado died as a result of her injuries. Ayala, who previously had been diagnosed with schizophrenia and had exhibited obvious signs of mental illness and depression to coworkers, “went to the Zone nightclub looking to shoot ‘preppies’ or ‘pop tweens,’ against who[m] he may have held a grudge.” Before shooting Delgado, Ayala had stood in front of the Zone and loaded his pistol.
Delgado’s shooting was not the first at that particular location. In July 2002, a shooter fired into a crowd of people standing outside the nightclub, striking three people. At that time, the club had a different name, but it was owned and operated by the Zone defendants. There also had been a “history of fights and assaults in the line outside the nightclub.” In addition, the area surrounding the Zone had experienced violent crimes before the 2009 shooting. In the years preceding the 2009 shooting, the downtown entertainment district was “plagued by recurrent incidents of violence,” which were “linked by police to gang activity and to clubs in the district exceeding capacity and serving too much alcohol.” In 2005, a series of shootings left two people dead and four injured. “It was known by Portland police and club owners alike that there was a high probability that more *63shootings would take place in the downtown entertainment district.” As a consequence, police “blanketed the downtown entertainment district with police officers to ease fear.” The effort, which was called “Operation Safe Streets,” included at least two dozen police officers patrolling on foot and horseback on Friday and Saturday nights through early morning, and gang enforcement officers, traffic safety officers, parole officers, and liquor control investigators were also on patrol. Owners of area nightclubs were asked to close early on weekend nights and were advised to have adequate security, cut off all intoxicated customers, and respond swiftly to problems or notify police.
After violence continued in the downtown entertainment district, in August 2006, Portland police called a “bar summit” with owners and managers of downtown bars and nightclubs to help them adopt policies to reduce violence. At the meeting, which included a representative of one of the Zone defendants, businesses and police addressed, among other issues, the shootings in the Old Town/Chinatown neighborhood. At the summit, police and businesses also addressed whether the violence on downtown sidewalks and in parking lots was related to intoxicated club-goers (as police believed) or drug dealers (as clubs contended).3
By January 2009, the Old Town/Chinatown neighborhood “had approximately fourteen agencies serving addicted, mentally ill, and homeless people, more than any other neighborhood in the city, with thousands of clients coming to the area every day. Many of those clients bought drugs.” “Drug dealers, drug users and gang members, all of whom frequented the area where the Zone nightclub was located, frequently carried weapons ***.” “[S]ome club owners, realizing that their clients were in danger of violent assault, increased security at their bars and nightclubs.” In 2006, a principal of one of the Zone defendants “acknowledged downtown safety problems, but said they were created by *64‘a few bad apples,’ and that police, liquor control, and others should “work together for the benefit of club-goers.”
Before the shooting, the Zone had undertaken some measures to provide security for customers inside and outside the club. Those measures included:
“(a) The Zone had an employee whose primary responsibility was monitoring customers as they come and go from the club. The employee was to assist in line control; to distance ‘undesirables,’ i.e. intoxicated persons, harassers, transients, known trouble makers, or gang affiliated persons from guests; and to monitor the parking lot and deter potential guests from loitering in or around their cars;
“(b) The Zone nightclub had a security camera that monitored the outside of their establishment;
“(c) The Zone nightclub had a security guard or door-person at the door outside of the premises who frisked everyone before entry, checking for drugs, alcohol, firearms, or other weapons;
“(d) In the past, the Zone had hired off-duty police officers to provide security for their customers; [and]
“(e) On or about 2006, [t]he Zone nightclub remodeled its facilities in the hopes of attracting a better clientele and reducing problems at the club including violence.”
Thus, plaintiff alleged, the Zone defendants “knew about the risk of violence that its customers faced.”
Plaintiff further alleged that underage nightclubs generally are understood to pose “inherent” risks of violence because of the high proportion of young male patrons, high noise levels, crowding, competitive environment, and underage drinking. “A metropolitan nightclub may see as many as three or four assaults on staff each night and are often confronted with armed patrons. Shootings, stabbings, felonious assaults, drug violations, and/or murders are commonplace in metropolitan nightclubs across the country.” In fact, several cities around the country “have banned or heavily regulated teen dance clubs.” According to plaintiffs allegations, the instances of violence that “occurred nationally at underage nightclubs and locally in the area around the Zone nightclub were all publicized in local and/or national *65media, and [the Rotary defendants] knew, or in the exercise of reasonable care, should have known, about the dangers of leaving [Delgado] at the Zone nightclub on the date and at the time in question.”
Plaintiff alleged that the Zone defendants owed a duty to Delgado, who was their business invitee, to exercise reasonable care to make the premises safe for her, including protection from criminal acts by third parties. Plaintiff further alleged that the Zone defendants breached that duty:
“(a) In failing to take reasonable measures to protect their customers from the criminal acts of third parties;
“(b) In making their customers stand in line in front of the club;
“(c) In failing to have sufficient security personnel to protect their customers;
“(d) In failing to properly train their staff to identify threats to their customers;
“(e) In failing to identify * * * Ayala as a threat while he was in front of [t]he Zone;
“(f) In failing to have adequate emergency response procedures in place to protect customers once a threat was identified; [and]
“(g) In failing to warn their customers or the parents of their customers about the risk of being assaulted while patronizing [t]he Zone nightclub.”
With regard to the Rotary defendants, plaintiff alleged that Delgado lived with a host family under the “auspices, sponsorship, and control” of the Rotary defendants and that the host parents were the Rotary defendants’ agents. Plaintiff alleged that the Rotary defendants were authorized to exercise “independent supervisory and parental responsibility to safeguard [Delgado’s] physical and mental well-being,” and, thus, that a special relationship between Delgado and the Rotary defendants “gave rise to a heightened duty of care beyond the general duty to avoid foreseeable risk of harm.” According to plaintiff, the Rotary defendants, directly or acting through their agents, breached that duty:
*66“(a) In failing to provide host parents with sufficient training so as to reasonably ensure the safety of the students;
“(b) In leaving the students unsupervised, late at night in a high crime area;
“(c) In failing to adopt and enforce reasonable rules and regulations to ensure the safety of the students;
“(d) In failing to adopt reasonable rules and regulations regarding travel to prevent the students from being left alone in dangerous areas; [and]
“(e) In failing to properly and adequately warn all the students and their parents of the dangers of being left unsupervised in a high crime area.”4
B. The Motions to Dismiss
The Zone defendants and Rotary defendants each moved under ORCP 21 A(8) to dismiss plaintiffs claims against them on the ground that plaintiff had failed to allege a foreseeable risk of harm to Delgado. Defendants argued that plaintiffs negligence claims reduced to the following theory: (1) criminals were more likely to commit crimes in the area in which the Zone was located; (2) defendants knew or should have known of that fact; and (3) Delgado was killed in that “high crime area.” Those allegations, defendants argued, failed to establish a foreseeable risk of the particular crime that killed Delgado, which defendants characterized as a “random spree shooting” against a particular class of person (“preppies”). According to defendants, such a shooting was just as likely to occur at a mall, a movie theater, or an ice cream parlor; thus, “the occurrence of a random spree shooting in a high crime area does not make it any more foreseeable than if it occurred in a place one would normally deem safe, such as a movie theater or a school.” Indeed, the “randomness” of shootings like the one that killed Delgado, defendants argued, “means that when and *67where they occur are not foreseeable, as that term is used in Oregon tort law.”
Plaintiff, meanwhile, argued that defendants’ arguments misapprehended what it means for a risk of harm to be reasonably foreseeable under Oregon case law. In plaintiffs view, “[fjoreseeability in this case does not turn on being able to foresee ‘random spree shootings,’ but rather on foreseeing the risk of a violent criminal assault” to Delgado. “[T]he dangerous character of the nightclub, its location and its violent history,” plaintiff asserted, “would allow a jury to conclude that the harm in this case was foreseeable.” She explained that “[tjhere are many foreseeable ways in which the assault might have taken placet.] * * * Even if an assault by a mentally ill person is somehow unusual or unexpected, the resulting harm was legally foreseeable because it was within the class of reasonably foreseeable hazards” at the Zone.
The trial court ultimately agreed with defendants, ruling that the shooting that killed Delgado was unforeseeable as a matter of law. The court subsequently entered limited judgments that dismissed all claims against both the Zone defendants and the Rotary defendants. Plaintiff appealed those limited judgments and argued before the Court of Appeals, as she did before the trial court, that her complaint alleged facts that, if proved, were sufficient to establish a reasonably foreseeable risk of the type of harm that befell Delgado.
C. The Court of Appeals Decision
As noted, the Court of Appeals reversed. As to the Zone defendants, relying on several of this court’s previous decisions, the Court of Appeals concluded that plaintiff had alleged sufficient facts that, if proved, would permit a trier of fact to find that (1) it was reasonably foreseeable to the Zone defendants that people standing in line outside the Zone late at night were at risk of harm from violent assault; and (2) Delgado’s death was within that general class of harm. Piazza, 271 Or App at 509, 510-11. The court further concluded that the circumstances alleged in the complaint were not “so highly unusual, or the sequence of events so attenuated, that *68no reasonable person in the Zone defendants’ position could have anticipated the harm to Delgado.” Id. at 512. Indeed, the court opined, “[a] jury could find, based on the facts alleged in the complaint, that the link between Ayala’s crime, the Zone defendants’ conduct, and Delgado’s death was relatively straightforward”; defendants exposed Delgado to a risk of harm from violent assault when they left her to line up and wait on the street in a high-crime area, where a criminal could approach the line and shoot at the Zone’s business invitees. Id. Therefore, the Court of Appeals reasoned, plaintiff had satisfied her pleading burden with respect to the issue of foreseeability as to the Zone defendants.
For similar reasons, as to the Rotary defendants, the Court of Appeals concluded that the complaint sufficiently alleged foreseeability of the risk of harm in at least one of plaintiff’s specifications of negligence — leaving the students in a high-crime area — to withstand a motion to dismiss plaintiffs claim under ORCP 21 A(8).5 Piazza, 271 Or App at 513-14. Plaintiff alleged that prior crimes in and around the Zone were “publicized in local and/or national media, and [the Rotary defendants] knew, or in the exercise of reasonable care, should have known, about the dangers of leaving [plaintiff] at [the Zone] on the date and at the time in question.” Id. The court concluded, from those facts, that a reasonable juror could find that the host parents’ conduct had exposed Delgado to a reasonably foreseeable risk of violent assault and that the type of harm that actually befell her was within that general class. Id. at 514.
Judge Edmonds dissented. As he saw the case,
“the very tragic specific harm that befell plaintiffs decedent was the result of a random shooter who was mentally ill and who undertook to shoot ‘preppies.’ The randomness of his criminal target demonstrates that it could have just as well have been committed at a soccer or football game, an outside youth church or synagogue gathering, a mall, a public or private school event, or any place where young people typically gather. Giving plaintiff the benefit of all reasonable inferences, it was a mere happenstance that *69the shooter chose the line of young people outside the Zone defendant’s club as his target. For purposes of reasonable foreseeability, a specific risk of harm is only within the scope of a general risk of harm if the specific risk of harm resulting in the injury is qualitatively similar to other foreseeable specific risks of harm that fall within a general risk of harm. That test is not satisfied by plaintiffs allegations. The circumstances of the random shooting spree in this case are unrelated to plaintiffs ‘high crime area’ allegations. For these reasons, the specific risk of harm from a random shooting spree like the one in this case is not within the general risk of harm alleged by plaintiff.”
Piazza, 271 Or App at 522-23 (Edmonds, S. J., dissenting).
Echoing Judge Edmonds’ dissent, all defendants assert on review that the type of crime that occurred— which they describe as a “random spree shooting” — was unforeseeable as a matter of law. They argue that the Court of Appeals majority erred by describing the scope of the risk of harm too generally and by relying on the occurrence of prior crimes that were dissimilar to the circumstances of the crime at issue. In short, they assert that the Court of Appeals expanded the concept of general risk of harm beyond the practical realities of what business operators, parents, and adult supervisors should reasonably anticipate as risks of harm. Defendants also assert that courts in other jurisdictions have limited liability in circumstances such as those present here.
Plaintiff responds that this court’s prior decisions have held that a plaintiff need not allege and prove the predictability of the actual sequence of events that caused the harm in question, but rather, must plead and prove facts showing a generalized risk of the types of incidents and injuries that occurred. Plaintiff urges that, under that framework — where all reasonably foreseeable risks of harm are relevant, not merely the particular harm that actually befell Delgado — plaintiffs factual allegations were sufficient to state a claim with respect to foreseeability.
II. ANALYSIS
The concept of foreseeability embodies a prospective judgment about a course of events; it “therefore ordinarily *70depends on the facts of a concrete situation” and, if disputed, is a jury question. Fazzolari v. Portland School Dist. No. 1J, 303 Or 1, 4, 734 P2d 1326 (1987).6 Foreseeability plays a role in at least two overlapping common-law negligence determinations: (1) whether the defendant’s conduct unreasonably created a foreseeable risk of harm to a protected interest of the plaintiff such that the defendant may be held liable for that conduct — formerly described in terms of “duty” and “breach” as measures of negligent conduct; and (2) whether, because the risk of harm was reasonably foreseeable, the defendant may be held liable to the plaintiff for the particular harm that befell the plaintiff — a concept that traditionally was referred to as “proximate” cause and which, in our current analytical framework, operates as a legal limit on the scope of a defendant’s liability for negligent conduct. See generally Fazzolari, 303 Or at 11-18 (describing role of foreseeability in negligence determinations formerly described in terms of duty-breach and proximate cause); see also Deckard v. Bunch, 358 Or 754, 761, 370 P3d 478 (2016). In synthesis,
*71“unless the parties invoke a status, a relationship, or a particular standard of conduct that creates, defines, or limits the defendant’s duty, the issue of liability for harm actually resulting from defendant’s conduct properly depends on whether that conduct unreasonably created a foreseeable risk to a protected interest of the kind of harm that befell the plaintiff. The role of the court is what it ordinarily is in cases involving the evaluation of particular situations under broad and imprecise standards: to determine whether upon the facts alleged or the evidence presented no reasonable factfinder could decide one or more elements of liability for one or the other party.”
Fazzolari, 303 Or at 17.
Under that framework, in effect, the more traditional duty-breach and proximate-cause analyses in a common-law negligence claim are subsumed in the question whether the defendant’s conduct resulted in a reasonably foreseeable and unreasonable risk of harm to a protected interest of the kind that the plaintiff suffered. Towe v. Sacagawea, Inc., 357 Or 74, 86, 346 P3d 1207 (2015).7 But, as the passage quoted above from Fazzolari observes, the duty-breach analysis is not always subsumed in a foreseeability analysis. Rather, the nature and scope of the duty owed by the defendant to the plaintiff can be created, defined, or limited based on, among other things, the relationship between or status of *72the parties. Towe, 357 Or at 86. So understood, our threshold inquiry is whether plaintiff has invoked a particular status or relationship that affects our analysis.
In her complaint, plaintiff invoked two special relationships that informed defendants’ obligations toward Delgado, which defendants have not challenged at this point in the litigation. The first, which plaintiff invoked as to the Zone defendants, is the relationship between possessors of land and persons whom they invite onto their premises. As possessors of premises, the Zone defendants had an obligation to take reasonable steps to protect the nightclub’s visitors from reasonably foreseeable criminal acts by third persons. See, e.g., Uihlein v. Albertson’s, Inc., 282 Or 631, 639, 580 P2d 1014 (1978) (noting that this court previously had adopted standard set out in Restatement (Second) of Torts section 344 comment f (1965) “as being a part of the law of this state”).8 The second special relationship, which plaintiff invoked with respect to the Rotary defendants, is the relationship between a child and a person entrusted with that child’s care. For example, schools have a special duty to students “apart from any general responsibility not unreasonably to expose people to a foreseeable risk of harm,” and the “scope of th[at] obligation does not exclude precautions against risks of crime or torts merely because a third person inflicts the injury.” Fazzolari, 303 Or at 19, 20.
In both special relationships asserted by plaintiff, the obligation to protect against risks of harm — including the risk of harm from third-party criminal acts — is broader than that which might apply in the absence of such a *73relationship, but, because those relationships do not prescribe a particular scope of duty, that obligation does not extend to risks that are not reasonably foreseeable. Id. at 16-17, 20.9 Thus, although each defendant is alleged to have had a special relationship with Delgado and a corresponding duty to take reasonable precautions to protect her against criminal conduct by third parties, those duties extended to only reasonably foreseeable criminal conduct. See id.
In this case, defendants do not dispute, based on the facts alleged in plaintiffs complaint, that they held the asserted special relationships toward Delgado, under which they had obligations to take reasonable precautions to protect Delgado against foreseeable risks of harm. Nor have defendants asserted that plaintiffs’ allegations are insufficient, if proved, and if the risk of harm to Delgado otherwise was foreseeable, to establish that defendants failed to use reasonable care to protect Delgado from harm. Stated differently, apart from their general foreseeability challenge, defendants do not separately assert that plaintiffs’ allegations were insufficient to establish, if proved, that their conduct was negligent.10 Instead, as framed by defendants’ motions, the only question before us is whether plaintiffs complaint alleged sufficient facts to withstand an ORCP 21 A(8) motion with respect to the issue of foreseeability as a legal limit on the scope of defendants’ liability.
The question is whether plaintiff has alleged facts sufficient, if proved, to permit a jury determination that reasonable persons in defendants’ positions would have *74foreseen a risk to Delgado’s safety “of the kind of harm that befell” her. See Fazzolari, 303 Or at 17. “The community’s judgment, usually given voice by a jury, determines whether the defendant’s conduct met that threshold in the factual circumstances of any particular case.” Chapman v. Mayfield, 358 Or 196, 206, 361 P3d 566 (2015). “The court intervenes only when it can say that the actor’s conduct clearly meets the standard or clearly falls below it.” Fazzolari, 303 Or at 18 (quoting Stewart v. Jefferson Plywood Co., 255 Or 603, 607, 469 P2d 783 (1970)).
In Stewart, this court elaborated on the nature of a court’s role in the foreseeability determination. In that case, the plaintiff volunteered to assist in putting out a fire that the defendant had started at one plant and that then had spread to an adjacent warehouse. The plaintiff and others were hoisted to the roof of the warehouse by a forklift. In attempting to put out the fire, plaintiff fell through a covered skylight — covered by corrugated plastic and a film of dust so that it was indistinguishable from the rest of the roof — to the warehouse floor, suffering injuries. The question, the court said, was “whether plaintiffs injury and the manner of its occurrence was so highly unusual that we can say as a matter of law that a reasonable man * * * would not have reasonably expected the injury to occur.” Stewart, 255 Or at 609.
The answer required identification of a court’s proper gate-keeping role:
“The temptation here is to leave the question to the jury where the problem can be solved by an intuitive process, thus relieving us from the judicial task of reaching a reasoned conclusion. Unfortunately, however, we have inherited the duty to exercise control over the jury and to keep it within the bounds set for it, vague as they may be.”
Id. at 607. On the facts presented, the court held:
“A reasonable man could foresee that a fire started by him might spread to his neighbor’s building and that in the effort to extinguish the fire his neighbor or a third person coming to his aid might be injured as a result of a variety of possible circumstances — by being burned, by falling off a ladder, by falling off a roof, by falling through a burned *75portion of the roof, or by other similar risks normally associated with fighting fires.
“It is less likely that an injury would occur as a result of falling through a skylight and even less likely that one would fall through a skylight because it was covered in a manner that made it appear to be a solid part of the roof. But, although such conditions may not be a common cause of injury, on the other hand it cannot be said that this setting for possible injury is so uncommon that a jury could not reasonably describe as foreseeable the risk of harm which it creates.”
Id. at 610. In short, the circumstances, although unusual, were not “out of the range within which a jury could determine that the injury was reasonably foreseeable.” Id. at 609-10.
In a later case involving third-party criminal conduct, this court recognized a limit on how generally the type of harm at risk can be described. In Buckler v. Oregon Corrections Div., 316 Or 499, 853 P2d 798 (1993) — a case on which defendants here heavily rely — this court considered whether the state could be held liable in negligence for the death and serious injury of victims of violence inflicted by an escaped prisoner. The prisoner was a member of a work crew in a remote location. The crew supervisor had left a state van’s keys in the ignition. The prisoner escaped in the van, drove 50 miles to his mother’s house, stole a gun, and, two days after the escape, shot two people, killing one of them. The trial court granted the state’s summary judgment motion, and, on review, the plaintiffs’ claims raised several issues concerning foreseeability relating to the plaintiffs’ various theories of negligence.
One foreseeability issue in Buckler concerned the plaintiffs’ claim that the state had facilitated the harm to the victims by leaving the van’s keys in the ignition. Id. at 507. The court explained that describing the type of harm at risk too generally — such as stating that criminals commit crimes or that escaped prisoners may commit crimes while at large — makes criminal acts the legal responsibility of everyone who may have contributed in some way to the criminal opportunity. Id. at 511. Such a conception would *76sweep too broadly because “mere ‘facilitation’ of an unintended adverse result, where intervening intentional criminality of another person is the harm-producing force, does not cause the harm so as to support liability for it.” Id. at 511-12. The court in Buckler ultimately concluded as a matter of law that the harm that the plaintiffs suffered was not a reasonably foreseeable consequence of the risk created by leaving the keys in the van. Id. at 514. In doing so, the court clarified that, for a jury question to arise with respect to foreseeability, the relevant risk of harm must be reasonably foreseeable. Id.
In keeping with its focus in Stewart, Fazzolari, and Buckler, this court recently has emphasized the centrality— in determining whether a triable issue has been established with respect to foreseeability — of the plaintiffs description of the injury-producing factual circumstances in the context of her theory of liability:
“[A]s part of determining whether the defendant’s conduct was unreasonable, there is nothing surprising about a conception of foreseeability that assesses the overall ‘setting for possible injury’ under the plaintiffs theory of liability. Drawing on the plaintiff’s theory of liability, the court in Stewart determined that it was not unlikely that an ‘injury would occur in this manner in the course of fighting a fire.’”
Chapman, 358 Or at 208 (citations omitted); see also Fazzolari, 303 Or at 22 (stating that, where school knew of assaults at specific locations on school property, trier of fact could find that school should have warned students of that general risk, which included risk of rape).
In Ckapman, this court held, in reviewing a grant of summary judgment, that the trial court had properly dismissed the plaintiffs action against a tavern that allegedly had over-served a visibly intoxicated patron who, in turn, left the tavern on foot, walked down the street to another business establishment, and unintentionally fired a concealed handgun through the door, injuring two people inside. 358 Or at 198. To establish foreseeability, the plaintiff primarily relied on (1) expert opinion evidence that intoxicated persons frequently become violent and that medical, scientific, and lay journals have documented a connection between violence *77and alcohol for decades; and (2) experiential observations by a bartender at another tavern in the vicinity about the link between alcohol and violence. Id. at 220. We concluded that the plaintiffs pleading and evidence did not create a triable issue of fact with respect to foreseeability. We explained:
“[E]ven though the precise mechanism of harm need not be foreseeable, it is necessary to describe the type of harm at risk and the class of plaintiffs at risk with reference to the particular factual circumstances of the case, as gleaned from the pleadings and evidence in the record. Based on the circumstances of this case, we conclude that the appropriate characterization of the type of harm at issue is an unintentional attack by a visibly intoxicated patron after he had left defendant’s premises. Cf. Fazzolari, 303 Or at 21-22 (characterizing type of harm from failing to provide adequate warning or security as an assault on students at school); Stewart, 255 Or at 610 (describing type of harm from negligently setting fire as injuries that may occur while attempting to extinguish fire). We further conclude that plaintiffs’ evidence was insufficient because it described the risk of harm too generally.
“As explained, evidence that it is common knowledge that intoxicated people have impaired judgment and may, therefore, behave improperly is too general to establish that a person who serves a visibly intoxicated person reasonably should expect that that person will commit an assault. Evidence making the bare assertion that it is common knowledge that visibly intoxicated persons frequently become violent is no more sufficient. Such evidence does not create a permissible inference that a particular defendant should have been aware of an unreasonable risk of violent harm or that a particular plaintiff was within the class of persons at risk of such harm.”
Chapman, 358 Or at 220-21.
Other, more specific evidence — such as evidence showing the rate of incidence of violence among intoxicated drinkers, the types of intoxicated drinkers who become violent, or the class of persons at risk of violent harm from a visibly intoxicated person — could have been adduced in Chapman that may have permitted different inferences so as to preclude summary judgment. The omission of such evidence, we held, distinguished the circumstances in Chapman *78from those in Stewart, where this court was willing, in the absence of more direct evidence, to use general knowledge to supply an inference that the risk of harm was reasonably foreseeable. Id. at 221; see Stewart, 255 Or at 611 (although record did not contain statistics on frequency of injuries resulting from falling through concealed skylights, “general knowledge” of manner in which injuries occur supported foreseeability of how injury could occur in such manner). The plaintiffs evidentiary proffer in Chapman did not include facts that would support an inference — based on general knowledge — that it was reasonably foreseeable to the defendant that the visibly intoxicated patron would unintentionally attack the plaintiffs at a different location. Chapman, 358 Or at 222; see Buchler, 316 Or at 511 (disavowing broad conception of foreseeability based on purported “common knowledge” connecting thieves, guns, and crimes).11
Before summing up the principles that govern our analysis, we briefly turn to one of defendants’ primary points of emphasis. There is an understandable thirst in this area of the law for bright-line rules that — at least in theory — can produce predictable results from case to case in negligence claims. In urging this court to require a high degree of similarity between the nature and circumstances of prior criminal acts and the criminal act at issue here, defendants note that some courts have analyzed whether a particular criminal act was reasonably foreseeable based on prior similar acts, and others have looked to the place and character of the location of the act. See, e.g., McKown v. Simon Property Group, Inc., 182 Wash 2d 752, 768-74, 344 P3d 661, 667-70 (2015) (discussing the two approaches). Courts following the “prior similar acts” approach have held that prior acts must be sufficiently similar in character to the act at issue before the latter can be deemed to have been foreseeable.12
*79Under the “place and character” test, courts have tended to focus on whether the place and character of a location or business “invited” the criminal behavior. See, e.g., Early v. N.L.V. Casino Corp., 100 Nev 200, 204, 678 P2d 683, 685 (1984), partially overruled on other grounds by Moody v. Manny’s Auto Repair, 110 Nev 320, 871 P2d 935 (1994). In response to a claim that an unprovoked attack at a large music festival was foreseeable based on incidents at previous festivals, and because crime is foreseeable when large groups of people gather, the New York Court of Appeals held that “[a] random criminal attack of this nature is not a predictable result of the gathering of a large group of people.” Maheshwari v. City of New York, 2 NY3d 288, 294, 810 NE2d 894 (2004). In the same vein, the California courts have emphasized that “it is difficult if not impossible in today’s society to predict when a criminal might strike,” and “if a criminal decides on a particular goal or victim, it is extremely difficult to remove his every means for achieving that goal.” Wiener v. Southcoast Childcare Centers, 32 Cal 4th 1138, 1147, 1150, 88 P3d 517, 524 (2004).
The concerns expressed in the cited cases are reasonable; in fact, they animated in part this court’s rejection of the facilitation rationale that the plaintiff in Buckler presented. We also recognize the relevance of the factors that other courts have identified in the foreseeability analysis. In many sets of factual circumstances, multiple considerations — including the place and character of the location of a criminal act, and the occurrence of prior similar acts — are relevant. However, as we repeatedly have *80stated, our preference for giving voice to the community’s judgment through a jury determination prevails, except in extreme cases, where no reasonable person could find that the harm that befell the plaintiff was reasonably foreseeable. Fazzolari, 303 Or at 17-18; Stewart, 255 Or at 607. Under that framework, a narrow focus on the actual sequence of events that led to a particular injury to a particular person “misunderstands foreseeable risk.” Towe, 357 Or at 106 n 17. A defendant need not have been able to precisely forecast a specific harm to a particular person to be held liable. See id. This court punctuated that point in Fazzolari when it made the following observations about Stewart:
“[T]he Stewart opinion acknowledged that ‘[wjhether the harm in a particular case is deemed foreseeable may depend upon the manner in which we describe “harm.” This is even more true of “risk.” Stewart itself made clear that the court meant generalized risk of the types of incidents and injuries that occurred rather than the predictability of the actual sequence of events. One reasonably might foresee that a fire might place others seeking to protect a neighboring building in danger of injury by falling as well as by being burned, even though one might not foresee that the fall would be through an insecurely covered and hidden skylight. So, at least, a jury might find.’”
Fazzolari, 303 Or at 13 (citations omitted).
Nor is it necessary that a particular quantifiable mass of similar types of incidents involving a narrow class of victims have occurred at a precise location for a triable issue of fact with respect to foreseeability to arise. In Fazzolari, for example, a jury question was created by generalized and somewhat limited evidence concerning the defendant school’s prior experience with criminal activity:
“Here there was evidence that a woman reportedly had been sexually assaulted on the school grounds 15 days before the attack on plaintiff, and plaintiff attempted to introduce evidence of various other kinds of attacks. Obviously a school’s responsibility for students’ safety against assault is not limited to the risk of rape, and evidence of foreseeability will differ depending on whether the risk of injury is claimed to be specific to a school, or schools *81generally, or a neighborhood, or a class of potential victims such as women or particular ethnic groups.”
Id. at 21 (citation omitted).
But that does not mean that the foreseeability determination is untethered to principle. As our previous decisions show, negligence claims arising from third-party criminal acts often involve the defendant’s failure to monitor or screen a dangerous third person or the placement of another person at unreasonable risk of criminal harm at a location and in circumstances that are unsafe. In either situation, there is a common requirement: a trier of fact must be able to find from concrete facts that a reasonable person in the position of the defendant reasonably would have foreseen that the person or location and circumstances posed a risk of criminal harm to persons such as the plaintiff. See generally Towe, 357 Or at 86; Oregon Steel Mills, 336 Or at 340; Fazzolari, 303 Or at 17. Pacts pertaining to the similarity, frequency, and recency of prior criminal acts, committed under the same or similar circumstances, or at or near the same location, and involving the same or similar types of victims, as well as the place and character of the location of the current criminal act, are all relevant to the determination. See Chapman, 358 Or at 220-22; see also Buckler, 316 Or at 511-12; compare Uihlein, 282 Or at 640-41 (store not liable for shopper assaulted in supermarket when little evidence of unsafe location), with Brown v. J.C. Penney Co., 297 Or 695, 710, 688 P2d 811 (1984) (store liable for shopper attacked in parking lot where there was ample evidence of criminal activity in area). Defendants have not persuaded us that those principles require further refinement or that the circumstances of this case present a compelling reason to do so.
III. APPLICATION
As discussed, defendants repeatedly recount the details of this particular shooting, emphasizing the mental illness of the shooter, his grudge against teens, and characterizing the attack as a “random spree,” and the shooting as “indiscriminate.” Defendants then fault plaintiffs allegations of foreseeability because they involve different facts. To the extent that their challenge focuses on the precise *82sequence of events that occurred in this case, defendants invite a taxing level of comparison that this court consistently has rejected. As noted, contrary to defendant’s focus, foreseeability has reference to the “generalized risk of the types of incidents and injuries that occurred!,] rather than the predictability of the actual sequence of events ” Fazzolari, 303 Or at 13 (discussing Stewart). Moreover, plaintiffs allegations support an inference that the attack in this case was not random in the sense of being indiscriminate or lacking a plan or purpose. The complaint alleged that the shooter specifically targeted the Zone because it was a teenage nightclub that attracted young people like Delgado. Although the shooter could have chosen other locations to carry out his plan, that fact does not make the shooting any less foreseeable to defendants.
We agree with defendants that, among the numerous facts alleged in plaintiffs complaint, some, viewed in isolation, would be too general to sustain her threshold pleading burden with respect to foreseeability. For example, the allegations that violent crimes have occurred in downtown Portland over a period of several decades, and in some instances, many blocks away from the Zone, and a recitation of general crime statistics for “these neighborhoods,” alone add little to the analysis.13 Defendants are right to observe that such generalized allegations — substantially separated in time and distance from the pertinent events *83here — closely resemble the sort of attenuated circumstances that this court eschewed in Buckler.
However, at the pleading stage of this case, the dis-positive issue is whether the allegations of foreseeability in plaintiffs complaint can withstand a motion to dismiss. In making that determination, we conclude that the proper level of generality at which to compare the criminal history on which plaintiff relies and the shooting in this case is to treat both as falling within the category of violent assaults. This court’s previous decisions generally prescribe a relatively broad level of generality in assessing the reasonable foreseeability of criminal conduct as a limit on liability. See, e.g., Fazzolari, 303 Or at 21 (treating sexual assault and other types of assaults at a school as falling in same category for purposes of foreseeability). That is especially true where, as here, the risk of harm at a particular location is at issue. See, e.g., Uihlein, 282 Or at 641 (“[T]here must be something to alert the storekeeper to the likelihood of harm of some kind from a criminal agency.”). That only makes sense. When a business owner is on notice that — in the absence of reasonable precautions — assaultive conduct can be anticipated where people are waiting in line to enter its premises, it rings hollow to suggest that a specific crime pattern triggers the owner’s duty to protect patrons only with regard to those specific crimes (such as gang-related homicide) that have occurred. Only if we can say that the types of historical crimes are so dissimilar, or so remote in time and location from the current crime that no reasonable person could determine that the current crime was foreseeable, will a court intervene to remove the matter from the jury as a matter of law. Fazzolari, 303 Or at 13, 17-18, 21.
Here, plaintiff alleged a repeated — if somewhat unevenly spaced — history of violent assaults, including gun violence, at and in the neighboring vicinity of the Zone, and a known risk of such violence in the future. In the years preceding the 2009 shooting, the downtown entertainment district — a nine block area that included the Zone — had been “plagued by recurrent incidents of violence,” which were “linked by police to gang activity and to clubs in the district exceeding capacity and serving too much alcohol.” *84In 2005, a series of shootings in that district left two people dead and four injured. “It was known by Portland police and club owners alike that there was a high probability that more shootings would take place in the downtown entertainment district.” Pertinent to the foreseeability of that risk are plaintiff’s allegations asserting, in effect, that the Zone defendants knew that teenage nightclubs such as the Zone are especially susceptible to incidents of violent assault. Plaintiff alleged that the Zone defendants had in the past acknowledged the risks of harm posed by that history at the Zone and in the downtown entertainment district. Plaintiff also alleged that a gunman stood and loaded his gun on that sidewalk, a gunman who was attracted to that location because he could find young people standing where he found and shot Delgado.
Those allegations would permit a reasonable trier of fact to determine that a reasonable person in the position of the Zone defendants reasonably would have foreseen a risk of violent assault on the public sidewalk outside the nightclub, where underage patrons were queued up to enter. They also are sufficient to permit a reasonable trier of fact to find that the harm that befell Delgado was within the class of harms at risk.14
The foregoing allegations would not, by themselves, permit a finding that the risk of harm to Delgado was reasonably foreseeable to the Rotary defendants. However, plaintiff further alleged, with respect to those defendants, that the prior violent assaults in and around the Zone were “publicized in local and/or national media, and [the Rotary defendants] knew, or in the exercise of reasonable care, should have known, about the dangers of leaving [Delgado] at the Zone nightclub on the date and at the time in question.” Assuming, as we must at this stage of the proceedings, the truth of those allegations, we cannot say that no trier of fact reasonably could determine that a reasonable person *85in the position of the host parents would have foreseen a risk of violent assault on the public sidewalk where they left Delgado. The question whether plaintiff can adduce evidence sufficient to create a triable issue of fact with respect to those allegations, of course, is not presently before us.15
Moreover, on the facts alleged, we are unable to say that a reasonable trier of fact would have to find that the criminal harm that befell Delgado was categorically different from the criminal harms of which the Rotary defendants had notice, and that no reasonable factfinder could find that the harm to Delgado fell within the class of harms of which the Rotary defendants are alleged to have been aware. The link between defendants’ allegedly negligent conduct and the actual harm that befell Delgado was direct: Plaintiff alleged that the Rotary defendants knew that there had been a significant history of violent assault late at night at or near the Zone, they took Delgado to the Zone and left her there, and she was the victim of a violent assault. That chain of events is not the sort of “concatenation of highly unusual circumstances” that compels the conclusion that Delgado’s death was not reasonably foreseeable as a matter of law. See Stewart, 255 Or at 609.
IV. CONCLUSION
It bears repeating that, as framed by defendants’ motions, the issue before us is not whether plaintiffs’ allegations were sufficient, if proved, to show that defendants’ conduct was unreasonable in light of a reasonably foreseeable risk of harm to Delgado. Thus, we are not concerned— for example — with whether any precautions that defendants took or failed to take were reasonable in relation to *86the nature or degree of foreseeable risk of harm to persons in Delgado’s circumstances. Instead, our exclusive focus is on reasonable foreseeability as a legal limit on the scope of defendants’ liability.
In Fazzolari, this court stated:
“Another person’s crime was once thought to lie beyond a defendant’s responsibility on grounds of ‘proximate cause,’ Aune v. Oregon Trunk Railway, 151 Or 622, 630-31, 51 P2d 663 (1935), but more recent decisions have dealt with the behavior of others, lawful or otherwise, as part of the general analysis of foreseeable risks.”
303 Or at 20. The court’s separation of foreseeability from causation arose from a concern that proximate cause not be invoked to improperly cut off claims by substituting a judicial veto for what ordinarily is a jury determination. As this court previously had stated:
“[W]e cannot lose sight of the fact that we are simply making an allocation of the appropriate functions of the court and jury. The sole function of a rule of limitation in these cases is to tell the court that it must not let the case go to the jury. The jury should be given wide latitude in setting these limits of liability, whether it be done under a formula of negligence or causation.”
Dewey v. A.F. Klaveness & Co., 233 Or 515, 535, 379 P2d 560 (1963) (internal quotation marks and citation omitted).
In this case, plaintiffs allegations accomplished — at the pleading stage — what the evidentiary proffers in cases such as Buckler and Chapman did not. Plaintiff alleged concrete facts showing a history of violent assault at and in the vicinity of the Zone nightclub and related facts that, if proved, would support a determination that reasonable persons in defendants’ positions would have foreseen a risk of violent assault against patrons lined up on the sidewalk outside the nightclub. Cf. Buckler, 316 Or at 511 (evidence of generic fact that criminals commit crimes insufficient to establish foreseeability).
Moreover, the type of harm that befell Delgado generally corresponded to the risk of violent assault to patrons of the Zone in Delgado’s position that was reasonably *87foreseeable to defendants. Cf. Chapman, 358 Or at 219-22 (evidence involving violence by or among visibly intoxicated patrons on premises at other bars did not support inference that it was reasonably foreseeable that defendant tavern’s well-behaved but visibly intoxicated patron would leave tavern premises and unintentionally shoot victims at different location). As our prior decisions teach, it is the general nature of the harm at risk, not the precise nature of the harm suffered by a particular victim, that controls the analysis.16 Plaintiffs allegations were sufficient to withstand defendants’ motion to dismiss.
The dissent’s disagreement with our conclusion requires some discussion but, in the end, it presents similar arguments to those of the dissent in the Court of Appeals.
The dissent’s most insistent disagreement with our ultimate conclusion is grounded in our characterization of the risk of harm in this case as violent assault. The dissent would describe the relevant risk more narrowly as “gang-, drug-, and alcohol-related violence.” 360 Or at 105 (Balmer, C. J., dissenting). According to the dissent, the relevant risk in this case is defined in plaintiffs complaint “by the type of criminal,” not “the type of crime.” Id. But, such narrow conceptions of the relevant risk are not supported by our previous decisions, including those on which the dissent relies.
It is true that, in determining the appropriate level of generality with which to characterize the risk of harm, we “[d]raw[] on the plaintiffs theory of liability” and “view[j the defendant’s conduct through the lens of the particular factual circumstances of the case.” Chapman, 358 Or at 208. However, as we also reiterated in Chapman, in the context of the pertinent factual setting, we “describ[e] the type of harm at risk more generally, rather than predicting an actual sequence of events.” Id. at 208-09. The problem with the plaintiffs evidentiary proffer on summary judgment in *88Chapman was that it relied on “common knowledge” that intoxicated tavern patrons engage in violence, an assertion for which there was no factual support in the record and which did not square up with the plaintiffs pleaded theory that the patron in that case unintentionally shot the plaintiffs. Id. at 220-21. In short, there was a failure of proof that might have been remedied by the presentation of concrete evidence that applied to the circumstances of the case. Id. at 221. Although the dissent acknowledges, as it must, that in assessing foreseeability, “we focus on generalized risks of harm,” 360 Or at 100 (Balmer, C. J., dissenting), it never fully comes to grips with that standard in appraising the allegations of plaintiffs pleading: A “generalized” risk of harm inherently connotes an assessment at a relatively broad level of generality.
The dissent also disagrees with our characterization of the relevant risk of harm at issue in Fazzolari. As the dissent sees it, Fazzolari’s discussion of the foreseeability of different kinds of assaults, in particular, its statement that “a school’s responsibility for students’ safety against assault is not limited to the risk of rape,” 303 Or at 21, must be understood as focused on “the reasonableness of the school’s precautions,” 360 Or at 107 (Balmer, C. J., dissenting), rather than on foreseeabilty as a limit on a defendant’s liability. We do not share the dissent’s reading of Fazzolari. When the quoted statement from Fazzolari is read in context, it is apparent that the court was discussing foreseeability as it relates to the risk of harm. Before making that statement, the court discussed the “issue of foreseeability *** [and whether] no reasonable factfinder could find thé attack on [the] plaintiff to have been a foreseeable risk.” 303 Or at 21. The court then referred to the “generalized risks of the types of incidents and injuries” discussed in Stewart. Fazzolari, 303 Or at 21. Applying that concept to the facts of the case, the court noted the evidence of the sexual assault on school grounds 15 days before the attack on plaintiff and that the plaintiff had attempted to introduce evidence of various other kinds of attacks. Id. The court then stated:
“Obviously a school’s responsibility for students’ safety against assault is not limited to the risk of rape, and evidence of foreseeability will differ depending on whether the *89risk of injury is claimed to be specific to a school, or schools generally, or a neighborhood, or a class of potential victims such as women or particular ethnic groups.”
Id. Only after discussing the foreseeability of the risk of harm did the court refer, as an additional point, to foreseeability as a measure of the defendant’s unreasonable conduct:
“Also, the character and probability of the risk that is claimed to be foreseeable bears on the steps administrators reasonably should take to avert it.”
Id. at 21-22. In any event, that reference does nothing to diminish the fact that, as the court held, the proper level of generality with which to view the risk of harm under the plaintiff’s theory of liability in Fazzolari was the risk of assault in general, not a particular type of assault. Id.
And, that was no aberration. As early as Danner v. Arnsberg et al, 227 Or 420, 362 P2d 758 (1961), this court held that, if a plaintiffs injury was “of the same general character” as an injury that could be anticipated, the injury was not necessarily unforeseeable. Id. at 427-28. In Stewart, we noted that “liability is confined to harms * * * that are of the general kind to be anticipated from the [allegedly tortious] conduct.” 255 Or at 608-09 (quoting Harper & James, The Law of Torts Introduction xl (1956) (emphasis and brackets added)). Fazzolari adopted the same standard. 303 Or at 21. In short, the question is whether, in the factual setting of the case, the harm suffered by the plaintiff is of the same general kind to be anticipated from the defendants’ allegedly negligent conduct. Nothing that we have said since Fazzolari has altered that fundamental construct.
As that principle applies here, the fact that the history of violent assaults at and in the neighborhood surrounding the Zone included gang-, alcohol-, and drug-related conduct, and shootings, as well as other forms of physical violence, is not dispositive of the foreseeability inquiry.17 The facts that plaintiff alleged, if proved, would *90permit a determination that the location at which Delgado was attacked — because it attracted party-going young people at night — attracted violent assault in various forms. As plaintiff alleged, the shooter was drawn to that location precisely because he expected to find partying teens at that location and time. To call the attack random, senseless, or mass does not advance the analysis very far. The shooting in this case was random in the sense that the shooter could have picked other locations to find teen victims, but it was no more random than attacks in the area that were the product of gang-related violence, to the extent that gang violence involves members of one gang targeting another gang at a location where that gang might be found.18 Reasonable minds can and do differ as to whether, in light of the history of violent assaults at the Zone and in the surrounding neighborhood, the attack on Delgado was reasonably foreseeable.19
In a different argument, the dissent asserts that, in concluding that plaintiff has sufficiently stated a claim with respect to foreseeability, we have mistakenly failed *91to distinguish between “the manner of harm” and the “end result.” 360 Or at 108-13 (Balmer, C. J., dissenting). According to the dissent, our characterization of the harm risked as “violent assault,” is an “end result,” upon which we focus to a fault, while “ignoring the manner in which the harm occurred.” Id. at 108. In making that argument, the dissent relies on several decisions of this court, but that reliance is misplaced.20
Hefty v. Comprehensive Care Corporation, 307 Or 247, 766 P2d 1026 (1988), illustrates the type of “extreme case” that is beyond the general rule of foreseeable harm. In that case, K, a teenager, was voluntarily admitted to the defendant’s Adolescent Care Unit for the treatment of alcoholism. Id. at 250. Six days after she was admitted, she *92left the facility against the medical advice of the unit. Id. Neither K nor the hospital notified her parents that she had left. Id. While riding as a passenger on a friend’s motorcycle, she sustained severe head injuries in an accident with an automobile. Id. at 251. Neither drugs nor alcohol was a factor in the collision. Id.
Importantly, the court stated:
“The risk to be foreseen was not that [K] would ride on the back of a motorcycle and be injured in a collision with an automobile. *** The risk to be foreseen was more general — it encompassed those generalized incidents and injuries created by discharging [K], an impaired minor addicted to alcohol, without notice to parents or police, without the permission of her doctor, without providing alternative care, and with the knowledge that [K] could not care for herself.”
Id. at 252 (citation omitted).
As pertinent here, Hefty stands for the proposition that the risk of harm to be foreseen must be described more generally than the dissent would describe it. Instead of describing the risk of harm to be foreseen narrowly as the risk that the daughter “would ride on a motorcycle and be injured in a collision with an automobile,” the court carefully tied the risk to be foreseen to the more general risk encompassed by the “generalized incidents and injuries created by” the defendant’s unreasonable conduct. Id. Ultimately, the court held that, because neither alcohol nor drugs had been involved in the accident — that is, because the connection between the defendant’s negligent conduct and the harm that the plaintiff suffered was too attenuated — no reasonable trier of fact could find that K’s injuries fell within the scope of the foreseeable risk posed by the discharge. Id. at 252-53.
The dissent also relies on Oregon Steel Mills. In that case, there was no causal relationship between the defendant’s conduct (negligently preparing an accounting report) and the “force” that inflicted harm on the plaintiff (a declining stock market); the stock market did not take notice of the defendant’s negligence and use the opportunity to decline. Oregon Steel Mills, 336 Or at 345. The two events *93occurred in a temporal relationship to each other that were completely coincidental.
This court in Oregon Steel Mills took pains to describe the problem for the plaintiff as lack of foreseeability, not causation in fact. See id. at 344. However, the court focused on the lack of a causal nexus between the defendants’ conduct and the market forces that led to the price decline that harmed the plaintiff, when it ultimately concluded that “the intervening action of market forces on the price of plaintiffs stock was the ‘harm-producing force,’ and defendant’s actions did not ‘cause’ the decline in the stock price so as to support liability for that decline.” Id. at 345 (emphasis added). That is another way of saying, as in Buckler, that the defendant’s conduct merely facilitated the harm ultimately suffered by the plaintiff. See Buckler, 316 Or at 511-12 (“[M]ere ‘facilitation’ of an unintended adverse result, where intervening intentional criminality of another person is the harm-producing force, does not cause the harm so as to support liability for it.”).
But, here, the relationship between defendants’ conduct and the harm inflicted on Delgado is qualitatively different from the relationships in Hefty and Oregon Steel Mills that this court deemed to be too attenuated.21 Here, in contrast to the circumstances in those cases, as alleged in plaintiffs complaint, the assailant’s access to Delgado and the other teens outside the Zone was directly connected to the Zone defendants’ asserted negligence in failing to provide adequate security against violent assault for patrons lined up on the sidewalk. To the extent that the dissent’s “manner of harm/end result” argument more narrowly describes the pertinent risks of harm as “gang-, drug-, and alcohol-related violence,” 360 Or at 108 (Balmer, C. J., dissenting), it is bound up in the dissent’s conception of the appropriate level of generality with which to describe the presenting risk of harm, a conception that we do not share.
Finally, the dissent argues that we have not adequately explained our conclusion that plaintiffs complaint *94sufficiently alleged foreseeability. Although the dissent may not be satisfied, the analysis cannot be reduced to the application of Euclidean-like postulates. That is because, as previously explained, foreseeability in the pertinent sense ultimately is a blended factual and normative inquiry. Because reasonable foreseeability invokes a community’s application of its own values, whether a particular outcome of conduct is foreseeable is most appropriately decided by the community’s closest proxy in our civil justice system, that is, by a jury. Exceptions exist only for harm resulting from “the concatenation of highly unusual circumstances.” Stewart, 255 Or at 609. Fazzolari reiterated that standard: “[T]he issue ordinarily can be left to the jury,” except for conduct “at the outer margins.” 303 Or at 12.22 In an earlier case, Justice O’Connell put it this way:
“In any case where there might be reasonable difference of opinion as to the foreseeability of a particular risk, the reasonableness of the defendant’s conduct with respect to it, or the normal character of an intervening cause, the question is for the jury, subject of course to suitable instructions from the court as to the legal conclusion to be drawn as the issue is determined either way.”
Dewey, 233 Or at 536 (O’Connell, J., concurring) (quoting William L. Prosser, Handbook of the Law of Torts 282 (2d ed 1955)).
Precisely because it is driven by community values, judges have no upper hand over juries in the inquiry, except to act as gatekeepers at the outer margins.23 With that *95limited role in view, what we have here is not a suburban mall in daylight hours or some other workaday public place where people ordinarily go about their routines with only a remote consciousness of the risk of a violent attack. Instead, as alleged in plaintiffs complaint, the scene of this shooting was an inadequately secured sidewalk queue where teenagers were waiting to enter a nightclub, late in the evening, in a high-crime urban neighborhood that, based on publicized past experience, posed a risk of violent harm to persons present. The harm that befell Delgado, although thankfully uncommon, was within the range of risks of harms that a reasonable factfinder could find was reasonably foreseeable in the circumstances alleged in plaintiffs complaint.
The decision of the Court of Appeals is affirmed. The judgment of the circuit court is reversed, and the case is remanded to the circuit court for further proceedings.
Plaintiffs operative pleading was a second amended complaint; for ease of reference, we refer to that pleading as the complaint.
The Zone defendants are Five Stars, Inc., Concept Real Estate, LLC, and Concept Entertainment Group, Ltd.
The Zone was located in what nsed to be one of Portland’s “Drug Free Zones,” in which anyone convicted of a drug offense could be barred from returning. The drug-free-zone program expired in 2007, and “drug dealers and addicts took over Portland’s Chinatown/Old Town neighborhood,” which police officers came to refer to as “crack alley.” “Residents complained of being terrorized by an increasingly aggressive and confrontational breed of drug users and sellers.”
The Rotary defendants ask this court to take judicial notice, pursuant to OEC 201, of certain facts relating to the shooting incident in this case that are not alleged in plaintiff’s complaint. In light of the procedural posture of our review of the judgment dismissing this action based on a motion to dismiss under ORCP 21 A(8), we decline that request, which was made for the first time before this court.
The parties do not discuss plaintiffs remaining specifications of negligence as to the Rotary defendants. Accordingly, we do not consider them further.
Although often described as an “issue of fact,” id.., it is more precise to describe the foreseeability determination as a blended factual and normative— that is, value-laden — inquiry that ordinarily is committed to juries. As one commentator has explained,
“[W]hen ‘foreseeable’ is used in what seems to be a kind of factual sense, as a prediction or anticipation, there is no implicit degree of likelihood in the reference. It may run from possible through likely and probable and even on up to certainty. ⅜ ⅜ * When ‘foreseeable’ is used in the context of determining if a defendant acted unreasonably, not in the fashion of the reasonable person, then the degree of likelihood can be factored in among all the variables entering into the calculus of harm, thus giving it a pragmatic use. In this context ‘foreseeable’ is modified explicitly or implicitly by ‘reasonably.’
«* * :|: * *
“More subtle is the use of ‘foreseeable’ in a mode that is not purely factual or predictive. To say that something is or was foreseeable can trigger an interpretation that it is being used in the purely predictive mode because the verb ‘to be’ is so often used as if information is being provided, although often opinion or value judgments are involved. Comparison with a lay use of ‘foreseeable’ may be instructive. If a passenger in an automobile were to say to the driver, ‘If you keep driving like this, it’s foreseeable we’ll have a wreck,’ or just, ‘You’re risking our lives,’ it is not likely he is only conveying information. It is a warning and a call for different behavior. Likewise, after the wreck, for someone to say, ‘Well, it sure was foreseeable,’ is to imply that it could have been prevented and likely that it should have been.”
Walter Probert, Torts and Language, 48 Fla L Rev 841, 854-55 (1996) (footnotes omitted).
Although uniform jury instructions are not sources of law, it often is useful to consider how sometimes elusive legal principles are understood and applied in practice by the trial judges and lawyers who are charged with following them. To that end, Oregon’s Uniform Civil Jury Instructions describe the overlapping roles of foreseeability in the following way. Under UCJI 20.01, the jury is instructed to determine whether (1) “[t]he defendant’s conduct was negligent”; (2) the defendant’s negligence “was a cause of harm to the plaintiff’; and (3) “[t]he harm was reasonably foreseeable.” UCJI 20.02 provides that a “person’s conduct is negligent if that person fails to use reasonable care.” That instruction further provides:
“In deciding whether a person used reasonable care, consider the dangers apparent or reasonably foreseeable when the events occurred.”
A separate uniform jury instruction based on Fazzolari, UCJI 20.03, addresses the issue of foreseeability as a limit on liability:
“A person is liable only for the reasonably foreseeable consequences of his or her actions. There are two things that must be foreseeable. First, the plaintiff must be within the general class of persons that one reasonably would anticipate might be threatened by the defendant’s conduct. Second, the harm suffered must be within the general class of harms that one reasonably would anticipate might result from the defendant’s conduct.”
Restatement section 344 comment f states:
“Since the possessor is not an insurer of the visitor’s safety, he is ordinarily under no duty to exercise any care until he knows or has reason to know that the acts of the third person are occurring, or are about to occur. He may, however, know or have reason to know, from past experience, that there is a likelihood of conduct on the part of third persons in general which is likely to endanger the safety of the visitor, even though he has no reason to expect it on the part of any particular individual. If the place or character of his business, or his past experience, is such that he should reasonably anticipate careless or criminal conduct on the part of third persons, either generally or at some particular time, he may be under a duty to take precautions against it, and to provide a reasonably sufficient number of servants to afford a reasonable protection.”
Even when a special relationship is the basis for the duty of care owed by one person to another, this court has held that, if the special relationship (or status or standard of conduct) does not prescribe a particular scope of duty, then “[ejommon law principles of reasonable care and foreseeability of harm are relevant.” Cain v. Rijken, 300 Or 706, 717, 717 P2d 140 (1986) (quoted with approval in Fazzolari, 303 Or at 16-17); see also Oregon Steel Mills, Inc. v. Coopers & Lybrand, LLP, 336 Or 329, 342, 83 P3d 322 (2004).
It follows that issues pertaining to whether defendants were negligent, such as “the feasibility and cost of avoiding the risk,” which “bear on the reasonableness of defendant’s conduct,” are not before us. See Donaca v. Curry Co., 303 Or 30, 38-39, 734 P2d 1339 (1987) (explaining difference between general foreseeability and duty/breach issues; whether county was negligent for failing to maintain right-of-way depended on whether county’s action was reasonable in light of risk, including feasibility and cost of avoiding risk); see also Hughes v. Wilson, 345 Or 491, 502, 199 P3d 305 (2008) (applying principle).
This court has “cautioned against turning fact-specific decisions on foreseeability into rules of law.” Chapman, 358 Or at 222 (quoting Bailey v. Lewis Farm, Inc., 343 Or 276, 289, 171 P3d 336 (2007) (citing Fazzolari, 303 Or at 16)). In Chapman, we concluded:
“We do not depart from that precept here; rather, our decision turns on the specific facts in the record before us.”
Id.
See, e.g., Romero v. Giant Stop-N-Go of New Mexico, Inc., 146 NM 520, 212 P3d 408 (2009) (holding that evidence of prior robberies, theft, physical *79altercations, domestic violence, harassment, narcotics and suspicious persons at same location did not make targeted and deliberate shooting of plaintiffs decedents foreseeable). Under that approach, courts also have held that the prior acts must have happened in the vicinity of the location where the act at issue occurred before that act can be deemed foreseeable. See Sigmund v. Starwood Urban Inv., 475 F Supp 2d 36, 46 (D DC 2007), aff’d, 617 F3d 512 (DC Cir 2010) (looking at a five-block radius around area of the attack and concluding that sudden and unexpected attack was not foreseeable). In addition, courts using that formula have held that the prior acts must have happened relatively recently, Dwiggins v. Morgan Jewelers, 811 P2d 182, 183 (Utah 1991) (one robbery five years earlier insufficient to make subsequent robbery foreseeable), and with some degree of frequency. See Dudas v. Glenwood Golf Club, Inc., 261 Va 133, 140, 540 SE2d 129 (2001) (two robberies within the month preceding the attack on plaintiff was not enough for foreseeability because there was no imminent danger).
For example, plaintiff listed the crime statistics of “these neighborhoods” in her opening brief before the Court of Appeals:
“From September 2006 through December 2006 — 1 homicide, 282 assaults, 3 rapes, 67 robberies, 15 sex crimes and 2 kidnappings;
“From January 2007 through December 2007 — 3 homicides, 16 rapes, 716 assaults, 187 robberies, 34 sex crimes and 7 kidnappings;
“From January 2008 through December 2008 — 13 rapes, 709 assaults, 145 robberies, 45 sex crimes, and 5 kidnappings;
“From January 4, 2009 through January 18, 2009 (the two weeks preceding the incident at issue here) — 31 assaults, 7 robberies and 2 sex crimes.”
Plaintiff does not, however, specify exactly what she means by “these neighborhoods,” beyond the fact that she is referring to the Chinatown/Old Town and downtown neighborhoods. Plaintiff also fails to provide information regarding the proximity of those crimes to the Zone. The “neighborhood” in question spans 52 square blocks, and the “downtown entertainment district” comprises nine blocks.
We deliberately do not assign a particular degree of significance or lack of significance in the foreseeability determination to many of the facts alleged in plaintiffs complaint, in terms of tipping the scales in favor of the overall sufficiency of plaintiffs pleading. By emphasizing the core facts summarized above, we simply note what, in our view, fundamentally surmounts the sufficiency hurdle.
The Court of Appeals correctly understood that:
“Plaintiffs complaint does not rest on allegations that crime is generally foreseeable, or on a conclusory allegation that [t]he Zone is located in a high-crime area. Rather, plaintiff alleged specific facts that, if proved at trial, would establish that [t]he Zone nightclub had experienced violent crime, including homicidal violence. Assuming, as we must at this stage of the proceedings, the truth of those allegations, and that the host parents were aware of those facts and nevertheless left Delgado there, a reasonable juror could conclude that the host parents’ conduct exposed her to a foreseeable risk of violent assault.”
Piazza, 271 Or App at 514.
Again, that is not to say that differences between past criminal acts committed at or near the Zone and the shooting in this case would not be relevant to the issue of foreseeability as it relates to whether defendants’ conduct was unreasonable in light of the risk of harm to patrons such as Delgado. That, as previously noted, is a different matter.
We note that, rather than give plaintiffs pleading the benefit of all reasonable inferences in its foreseeability analysis, the dissent has failed to give appropriate weight to several relevant historical facts alleged in plaintiffs pleading, including the 2005 series of shootings in the Downtown Entertainment District, *90the “history of fights and assaults in the line outside” the. Zone, and the more general allegation that “shooting, stabbings, felonious assaults, drug violations, and/or murders are commonplace” in underage nightclubs. See 360 Or at 103-05 (Balmer, C. J., dissenting). Moreover, contrary to the dissent’s apparent assumption, id. at 105, and, giving, as we must, plaintiffs pleading the requisite benefit of all reasonable inferences, there is no indication that all those incidents involved drug-, alcohol-, and/or gang-related violence, nor is there an indication that the 2002 shooting at the crowd outside the predecessor of the Zone was related to gangs, drugs, or alcohol.
Here, the shooting was intentional — -just like a gang- or drug-related shooting would be — and it occurred in a high-crime area with a history and reputation for violence. The risk of violent assault that those circumstances created included an intentional shooting that may leave incidental, unintended victims. Gang members and drug dealers do not always aim with precision. In that sense, the type of shooting in this case is not too different from the “mass murder” scenario that the dissent implies.
The Court of Appeals put the matter well when it stated:
“Whether the circumstances of a particular assault — such as the shooter’s motive or mental state, the intended victim, or the weapon used — are ‘details,’ or instead serve to make the crime qualitatively different from other foreseeable harms, will vary from case to case. Based on the facts alleged in plaintiff’s complaint, a reasonable juror could conclude that this type of assault — a gunman firing at patrons waiting in line to enter [t]he Zone — fell within the category of risks that the Zone defendants should have anticipated, regardless of the shooter’s particular motive, mental state, intended targets, or weapon used.”
Piazza, 271 Or App at 511-12.
The dissent also relies on cases from other jurisdictions. As the dissent notes, 360 Or at 111-12 (Balmer, C. J., dissenting), the Seventh Circuit in Shadday v. Omni Hotels Management Corp., 477 F3d 611, 517 (7th Cir 2007), limited the defendant’s liability for criminal conduct in that case under District of Columbia law to circumstances “when it has some reason to think such crime likely.” Similarly, in Commonwealth v. Peterson, 286 Va 349, 749 SE2d 307 (2013), relied on by the dissent, 360 Or at 102 n 2 (Balmer, C. J., dissenting), the court followed the Virginia rule providing, unlike Oregon law, that
“where the special relationship was that of business owner/invitee or landlord/ tenant, we have imposed a duty to warn of third party criminal acts only where there was an imminent probability of injury from a third party criminal act.”
Peterson, 286 Va at 357, 749 SE2d at 312 (internal quotation marks and citation omitted; emphasis added). Because this court has never required that an injury be probable, let alone imminently probable, in order to satisfy the foreseeability requirement, those authorities are simply inapplicable here. Moreover, although it is unnecessary to pursue the matter in detail, as two commentators have noted, there is much room to question the reasoning of Shadday, in particular:
“The idea that a hotel owes no duty to its guests to take steps to protect them from other guests is quite untenable. Imagine that a hotel installed low-grade electronic locks that enabled guests easily to use their own room keys to open other guest’s rooms. If a guest were to take advantage of this inadequate security system to break into the room of another guest and attack the other, the victim’s claim that the hotel failed in a basic responsibility to provide security would be quite compelling. This would be a failure to provide the security that a guest reasonably expects from a hotel and that a hotel implicitly promises to its guests. The duty Shadday was owed was a duty to take ordinary care against her being attacked on the premises. Parsing that risk into the risk of attack by intruders versus guests is arbitrary and unmotivated.”
John C. P. Goldberg & Benjamin C. Zipursky, Civil Recourse Defended: A Reply to Posner Calabresi, Rustad, Chamallas, and Robinette, 88 Ind LJ 569, 588-89 (2013).
As we later explained, Oregon Steel Mills “turned on the specific facts before the court.” Bailey, 343 Or at 289-90.
The dissent misunderstands our reasoning in asserting that, “[i]n effect, the majority concludes that an assailant’s methods and motivations are irrelevant to the foreseeability analysis.” 360 Or at 108 (Balmer, C. J., dissenting). To the contrary, those factors are relevant to the analysis, but we do not share the dissent’s view that they are so conclusive in this case as to preclude a jury determination that the harm that befell Delgado was reasonably foreseeable to defendants.
In a thoughtful discussion of the underpinnings of the allocation of responsibility between judge and jury in foreseeability determinations, one commentator has stated:
“The genius of the jury is that it brings to each case multiple perspectives, both shared and diverse experiences, and (with the exception of the occasional attorney-juror) a legal tabula rasa. To put it simply — especially when considering a question like foreseeability that is part-analysis, *95part-community experience, and part-gestalt — perhaps twelve heads are better than one.”
W. Jonathan Cardi, Purging Foreseeability: The New Vision of Duty and Judicial Power in the Proposed Restatement (Third) of Torts, 58 Vand L Rev 739, 800 (2005) (footnote omitted).