Sheriff deputies stand outside a house in Cleveland, the day after three women who vanished a decade ago were found there. (AP Photo/Tony Dejak)   Spoken by Charles Ramsey in an interview with ABC News last week, the words heard around the country were a case study in the impact of bystander intervention:

“I heard screaming. I’m eating my McDonald’s, I come outside, I see this girl going nuts trying to get out of her house. So I go on the porch, I go on the porch and she says ‘help me get out, I been in here a long time’. So you know, I figured this is a domestic violence dispute…” Charles Ramsey, a resident of Cleveland, Ohio, had no idea that he was living next door to three young women who went missing more than ten years ago. He and his neighbor Angel Cortero (who, by some accounts was the first on the scene) helped Amanda Berry, Gina DeJesus and Michelle Knight escape the house where they were trapped. The three women, abducted as girls, were living in captivity and had managed that day to draw their neighbors’ attention while their kidnapper, Ariel Castro, was away from the home.

The story prompts many questions about how three captive women could have been living in such close proximity to so many who never realized they were there. Yet as a fuller story slowly emerges, one thing seems clear: this was an intervention by people who did not know the victims personally, who heard Amanda’s shouts and got her out.

The incident, and the intervention in particular, has been hailed a victory of feminism and a perfectly executed example of the adage “If you see something, say something.” It’s a victory when violence against women is acknowledged as a societal problem and not merely a family issue, operating under a notion privacy that places the importance of freedom from violence significantly below the protection of the patriarchal order. Until very recently, violence in the home was not widely considered a social issue. The idea of a man raping his wife didn’t even register as a problem, cultural or legally. Thanks to decades of advocacy—and a ton of hell-raising—we are building a cultural lexicon in which family and intimate partner violence are recognized as violence at all. That cultural shift, as slowly as it is happening, is no small achievement. A key element of this shift has been normalizing bystander intervention in instances when sexual harassment, sexual assault or intimate partner violence seem to be unfolding. The fruits of that work were evident in the clarity with which both Charles Ramsey and Angel Cortero expressed their decision to intervene.

Encouraging bystander intervention is an explicit challenge to the “bystander effect,” the acknowledgment that there are many obstacles in motivating bystanders to act. It takes a perfect storm of conditions that might enable a person to overcome internal and external resistance to act to help someone they do not know. A person or group of people must first notice the situation and determine that it requires intervention. They must feel some responsibility to act, decide how best to do that and have confidence in their ability to be successful. In a perfect situation, where all these criteria are met, interventions still don’t happen as often as they should.

What happens, then, when we add a few more complications into the mix?

Intervention can mean many things, some as simple as saying “Hey, that’s not cool!” or “Please stop, I think you’re hurting him.” It can involve bringing others to the situation: teachers, other bystanders or the authorities. Even a relatively lower-stakes verbal assertion carries within it the implication that if the action is not stopped, the intervener will do something to make the person stop, by physically intervening themselves or calling upon the authorities.

For those who feel as though the authorities, the people in power to assist, are there to protect them and the person being harmed, this recourse makes sense. But what if there is a deep-seated lack of confidence in the police, security staff or any other authority the bystander might call on? Think about communities where calling the police might create more problems than it would solve: communities and families in which people are undocumented, with histories of suffering from police brutality, surveillance and incarceration. Before we hail bystander intervention as key to ending violence against women, we must ask: Can we prevail upon people to intervene as bystanders when the bystander’s community lacks confidence in the authorities it would call upon? Where the historical legacies of state-supported violence and oppression are active in shared memory and are still experienced daily?

Reliance on the criminal justice system as the primary solution to intimate partner violence is a long-standing point of debate in the movement to end it. Thinking through the implications of bystander intervention can help us further articulate the need for community-based solutions to violence that do not rely solely on the criminal justice system as the sole or ultimate recourse for survivors and bystanders alike. A reliance on the criminal justice system is insidious not merely because of its ineffectiveness but largely because this approach tacitly excludes of certain victims of violence. Reporting is very low in communities of color, immigrant communities and queer and trans communities. For fear of additional, unwanted police presence in their lives and communities, many of these folks just refuse to call the police. Think of the impossible choice an undocumented woman faces, for example, if calling the police for help might mean risk deportation and detention of her children, herself or other family members. Native women, until a recent change to the Violence Against Women Act, had no legal recourse at all if they were sexually assaulted by a non-native person on native land.

People who live under threat of persecution by the state cannot be asked to rely solely on the state to protect them. Considering that we’re offering them solutions that do not take seriously the material conditions of their lives, when we ask this of these communities, we send the message that these victims and survivors of violence don’t fully matter to us.

Community-based responses to violence are, by definition, specific to the communities they are intended for. The most commonly available resources—domestic violence or battered women’s centers, shelters, sexual assault or rape crisis centers, legal assistance clinics, counseling centers, victim-witness programs and family justice centers—assume the following: that survivors want to separate from the person perpetrating the violence and that the best way to seek help is to call 911. These resources mandate reporting to child protective services and require dealing with the police for any intervention programs with the perpetrators. These factors accumulate and serve as deterrents for those who have histories of distrust with these institutions. Community-centered solutions often include the following: reliance on friends, family, neighbors, co-workers and other trusted community members to intervene and address the violence instead of the police, as well as options that allow survivors to stay together and sometimes even with the perpetrators. We must also expand our understanding of who an “expert” is and broaden the scope of who can work with survivors of violence to create safety plans. Community-centered solutions also think of perpetrators as people who are capable of change—and it’s important to note here that Charles Ramsey himself has a felony domestic violence charge on his record.

It’s a challenge to find ways to engage perpetrators of violence to prevent further violence and minimize their harms without removing them from their home or community. But advocates of community-based solutions often point out that high recidivism rates, as well as the fact that removing and incarcerating people does not offer them any skills or create new structures of accountability when they return home, as limits of a criminal justice–focused approach. While we must leave open the possibility of a survivor’s need to be separated from the perpetrator, we must also do no further harm by requiring police intervention in every case.

If we are really going to change our culture of violence, we have to take seriously the most vulnerable survivors and we must consider the impact of all strategies on the most ostracized and targeted communities. We must work to change the criminal justice system to work for them, and we must simultaneously offer some community-based models that communities can trust, that are rooted in their lives and that can be utilized in the face of violence. That ought to be our call to action.

Here are some resources for community-based solutions to violence:

• Creative Interventions: They have used data from their Community Based Interventions Project and Storytelling and Organizing Project (STOP) to construct a comprehensive toolkit for anyone interested in carrying out a community-based intervention to violence.

• INCITE!: Women Of Color Against Violence: Please see their Community Accountability Working Document for an overview of community accountability issues, principles and guiding questions.

• Philly Stands Up: a small collective of individuals working in Philadelphia to confront sexual assault in our various communities using a transformative justice framework. Here they explain their Points of Unity.

• Transformative Justice Law Project of Illinois (TJLP): a collective of radical lawyers, social workers, activists, and community organizers who are deeply committed to prison abolition, transformative justice, and gender self-determination.