Tshaka Barrows, deputy director of the Burns Institute, discusses his organization's collaborative and community-centered approach to addressing and eliminating racial and ethnic disparities in the justice system. Barrows spoke with Robert V. Wolf, director of communications at the Center for Court Innovation, after participating in a panel on Race and Procedural Justice at Justice Innovations in Times of Change on Sept. 30, 2016.
The following is a transcript of the podcast:
Tshaka BARROWS:
We call it a system, but it really isn't a system. It's much more of a grouping of semi-autonomous agencies that have very little accountability to each other.
Robert V. WOLF:
Hi, I'm Rob Wolf, Director of Communications at the Center for Court Innovation. And today I'm in North Haven, Connecticut, at the Justice Innovation in Times of Change Conference. Sitting down with me is Tshaka Barrows, who is the Deputy Director of the W. Haywood Burns Institute, which works to address racial and ethnic disparities in the justice system. The Institute is based in Oakland, California. So, you've come a long way to attend the conference and participate, and to sit and talk with me. So, thank you very much.
Tshaka BARROWS:
I'm glad to be here.
Robert V. WOLF:
Let's talk about the work of the Burns Institute, and in particular how you work with jurisdictions to reduce racial and ethnic disparities in the justice system. You have a specific approach you take to looking at this issue and trying to address it. So, maybe you could summarize for me what that approach is.
BARROWS:
So, at the Burns Institute, our approach is to build a collaborative of the different agencies that make up the justice system. And I always tell people, I just told a group, we call it a system, but it really isn't a system. It's much more of a grouping of semi-autonomous agencies that have very little accountability to each other. So, the whole notion of trying to address disparities has to be done with that context in mind, because much of one agency's decision bump into the next, bump into the next, and the impact is felt by the individuals who are going through it. And we see it in the disparity numbers, but to really create a strategy to address it, you have to have all those key players from each of those agencies as a part of your collaboration.
And we also, fundamentally, don't think that just having those traditional stakeholders is enough. So, our process requires that we engage meaningful participation from community stakeholders who've had direct experience with the justice system, who live in the neighborhoods that our data shows are the target neighborhoods, where more people are coming from, so that they can both bring that experience from having traveled through the system, through the various agencies, being passed from one to the other, but also what it's like living in a community that is targeted for higher involvement for various reasons, policies, policing policies, could be that there's a lack of resources, any number of conditional factors.
But this whole notion of creating a more fair notion of procedural justice can't be done without accounting for the fact that certain neighborhoods are much more highly represented in the system. So, our process, we really aim for participation with community stakeholders, which is very different. People are a bit afraid of that, the idea that you're sitting in a meeting, sharing data with people who are upset with your agency, who did not feel that they were treated fairly, who are angry about the realities that folks in their community face, is a threatening notion for most traditional stakeholders who already a conversation of race in this country typically is a bit unnerving for people. It's not like that's a regular practice that we have. So-
WOLF:
And you literally bring everyone together in the same room, that's the process? It's, "Let's all sit down together?" So, what does that look like? How many people are actually sitting around a table or in an auditorium?
BARROWS:
That's a great question. So, we build a collaborative, and it's a process to even build it. So, we don't try to just come in with a cookie-cutter prescription. We want to understand from the local players. Justice happens locally, there's a culture. Who do they think are the key people that need to be there. So, sometimes we may get huge representation from one agency where it's, "You guys are dominating the meeting, and we may need to adjust that." So, there's a need to attend to the actual formation. Typically, it's, I'd say between 10 to 20 stakeholders, depending on the size of jurisdictions. We work in very small, rural places that may not have a huge collaboration.
I've worked in jurisdictions that have had up to 30 people who meet every month, but that becomes to be a challenge in and of itself because if everybody just introduced themselves, that would take time. To have meaningful dialogue about certain issues in a meeting of that size, it can be a challenge. So, you really want to look for a sweet spot that allows for equal representation across the agencies and doesn't leave behind any one particular group.
WOLF:
And what happens then? What's the process? You said, every month. So, are you trying to build a permanent infrastructure for dialogue, or is it a time-limited "let's meet for X number of times to work on this"?
BARROWS:
That's another great question. Our process will be monthly. The hope is we're setting the jurisdiction up to maintain the process without us. We do a whole orientation to really try to help everybody to fully participate. We don't want people to just sitting there and they're, "I don't know what this is. I don't know what's going on," acronyms are flying over their head. So, we spend time doing coach-ups for the community stakeholders. And we also orient the system folks to what the meeting will be like when they have community members there who might be more frustrated or can ask lots of questions.
Rarely are system stakeholders very good at telling the story of their institutions and how they've gotten to this point. So, we also have started to work with them on telling their story. This is, you've got to own this. You didn't do all of this. You don't have to apologize for the history, but you need to own the fact that there were some practices that were not the best that we were doing, that we've been working on trying to address, because that engenders a level of respect for the process and opens up the community to thinking that, "Okay, you really are serious about doing something different."
WOLF:
And when everyone sits down, have they already accepted the premise that there are racial and ethnic disparities-
BARROWS:
Yeah.
WOLF:
... Or do you also need to establish that as the facts on the ground?
BARROWS:
We will likely revisit ... A lot of times people will say, "Oh yeah, no, we've all, we've, we understand we have a problem." And it's, "Well, let's talk about it." Then, we started asking, "What do you think is contributing to the problem?" So, it's one thing to say, "Yeah, our jail or our juvenile hall is full of people of color." It's another thing to say, "And we think we have a responsibility for that. We think we're contributing to that." When we asked the question, "And what do you think drives this," everything but them usually is the response we get, which lets us know, you're probably don't realize what this is going to be like. You're going to feel, well, why are you guys asking us about what our decisions are? It's because you have control over that.
You don't have control over external factors like Hollywood violence in movies, culture of violent music, or just the fact that there is this history of segregation in the country. You can't just undo that in your collaborative. We don't quite have the power to say, "You know what? Let's just change the zoning and all the way the neighborhoods are set up, and let's go ahead and make it so that job discrimination doesn't happen anymore. It's, those things aren't really in the purview of that particular collaborative, but their decision-making practices are. So, you can control who you violate for probation.
Do you send out bench warrants before actually reaching out to people in their native language? So, do you know if your court letters are landing on folks who couldn't understand it in the first place, and now you're putting a warrant out for someone who never fully engaged in the information in the beginning. So, we then analyze each decision point by race, ethnicity, gender, geography, and offense, so that it's a way for us understand, at each decision point, what are we doing, what is the impact of our decisions, where are people going, what's happening?
WOLF:
How do you know what the impact on all those factors ... You just said at each decision point, meaning, at arrest or a decision to charge or a decision to carry a case forward or a decision sentencing or to plea, all those things are decision points, right? Do you just ask people, what do you do, or you're looking at actual hard data and numbers?
BARROWS:
We first go to hard data and numbers, if they have it. Oftentimes they don't, and so that is a huge problem. So, we're also not researchers. This isn't a research project. We're not trying to prove that our data that we've got is super accurate. Basically, we use what you have to try to figure out a way forward, understanding your data might not be perfect. One issue we see all the time is the issue of ethnicity around Latinos. So, very few jurisdictions have a really great practice for capturing Latinos within their justice system. Typically though, they get captured as White. So, it skews the white population up and skews their Latino population down, and it throws all of the comparisons that we want to make off.
And there's a set of conditions that attribute to it because it's an ethnic group, people speak Spanish, maybe they don't. There's a lot of factors. There can be very light-skinned Latinos. So, one of the things we ask, "Well, who decides? Is it your staff? Do you ask the person directly? What's the process for the collection of the data?" Because typically, once we start to analyze it and show it back to them in meetings, we'll start to get some pushback. "Well, where did you get these numbers? What is this? This is wrong." And it's, "Well, these are your numbers. We got it from you. Now, they may not be as accurate as they could be, but this is what we have right now. So let's get started."
So, we don't want to be in a process of never-ending cleaning of data, reviewing the data, and then getting into this just adoration of the question and just adoring this, "Well, what else do we need to think about it," and, "What else is right," versus, "Okay, I think we know enough that there's a tribe on our, we have a reservation in our county, and 30% of the young people in our justice system, or 30% of the adults are coming from that reservation. I think we can start there." Now, maybe we want to tribal affiliation, we need to go a bit deeper, and those things are helpful, but that's where we like to begin.
So, once we orient folks to the process, we'll do a history, talk about how this country started, how these justice systems were started, to give everybody equal understanding of the playing field. Then, what we like to do is start, actually, looking at data, looking at what they have. Like I said, we're not a bunch of researchers. So, we take people's dirty data and use it, and we're not going to just say, "Well, we can't go forward until this is pristine." It's, "Well, no, this is what you have right now." There's tiers. So, the first tier is, if you think it's not clean enough, what do you need to do to address it, and we can try to help with some of that, but really that needed to be owned by the jurisdiction.
How do you analyze it? Is this a new practice? So, if it's new, they might become defensive when all of a sudden you're sitting in a meeting with your peers, other agency heads, looking at data that really shines a light on your staff's practice in terms of making decisions, and feel, "Well, wait a minute. I, you know, why is everyone looking at us?" So, there's a first group to get data scrutiny. Usually it's a bit raw, because this is a whole new practice. They may not even look at this data regularly, internally, and so there's not a defense in place to explain away what's happening. There's this nervousness. So, that's a process in and of itself.
So, all this takes time. None of this is fast. Our main goal is to get to the point where we can have the group establish a target population for racial and ethnic disparities that they want to move safely out of their system. So, we keep looking at the decision points, not going to pick the most politically challenging, so we're not going to look at armed robbery, if you will. A lot of times people are not ready to say, "Yeah, let's move those folks out of the system safely." But looking at bench warrants, violations of probation, offenses that really aren't about public safety at all, but much more about administration of services, but totally contribute to disparities in real ways that we can ...
WOLF:
Then, you get consensus, and you say, "We're gonna target-
BARROWS:
We're going to target the population-
WOLF:
... People who have violated probation or young people,"` or something that-
BARROWS:
Try to show it as a number per month. So, what I don't want to do is say, "Yeah, each year you have 500 violations of probation, and 50% of those are Black men and these two neighborhoods. Well, over the course of the whole year, how do you understand what your work is?" What I'd like to say is, "Okay, of that per year, how many is that per month? What are we actually talking about on a monthly basis? And can we dig deeper to understand how these live?" Now, we're looking at, each month it may be 25, 30 people who were violated. Well, let's understand the nature of that. What are their probation officers' perspectives on this? What programs were they in?
So, you want to then, what we call "peeling back the onion", so you get down to this target. Now, you want a focus group, you want to bring line staff, you want to talk to people directly who have been in that experience. And you're looking for, not just a policy change, but you're trying to understand, well, what innovation or intervention can we come up with to move this?
WOLF:
So, it sounds though like it's on a very, I don't want to say small scale, but you have to target this group and that group in terms of making a difference. It's not, "Here's the solution, oh," and it ripples throughout the whole system and disparities?
BARROWS:
No. It's just to monitor and track it. So, it's everything you said, and you have to monitor and track it. Literally, we've come in and people said, "Yeah, we have, we have disparities, you know. 81% of our RMAs are African-Americans." And it's, "Okay, well, what else do you know about them?" "Nothing." So, what could you do? You just going to say, "Oh, let's just release 81% of the inmates and reduce the disparity." Nobody's going to do that. So, they get their hands tied. So, we have all these big picture data, annual shots. None of it helps people to know what to do to move forward.
So, what we've developed is a strategy and approach that really breaks it down into workable pieces. And we even have a slide that we go through that really shows people, if it's a state law, that is the reason why this person's locked up, you can't change that. But if it's a policy that you just detain people for this because you feel strongly, well, you can stop that tomorrow. That's just an internal office policy. That's not a state law. So, understanding how these things play out is really crucial, but it takes time, it takes that investigative work. You have to actually include the people who do the work on the day-to-day, the line staff, not just the supervisors and the managers, because you these are people that can try make it work.
WOLF:
I want to ask one more question, but I think it's probably a complicated one that has a long answer, but how do you deal with the issue of implicit bias? Everything you've described to me is something you could see on paper and go, "Oh, look at this number, look at this policy. You put these together and that equals a disproportionate disparity." But what about these things that are more intangible, yet that we know impact at these decision points, why they decide to charge someone, give someone a higher charge and someone not a higher charge. If there's bias involved and it's happening in the back of their heads and they don't even know it, how do you address that?
BARROWS:
Well, because we can do case-level analysis. So, we could show two pretty similar situations and say, "Let's talk about how did you make this decision here," and then, "Why did you make this decision here," and not try to label someone and say, "We've caught you," but we'd rather show them what they're doing and see if they themselves can see these patterns. But we also bring in community people in the meeting who are going to naturally see those patterns because that's their experience, and they'll ask the question very directly to say, "Well, I don't think that that makes sense." So, you need that person who's not going to play so much by the rules to say, "Well, and why did we do that? That doesn't seem to make sense," or, "Why is that in this neighborhood?"
I'll give you an example. In one city we worked in, in a particular area of town, any Latino kid with a marker was considered in a gang and was writing gang messages on the walls and creating potential shootings. And there was the narrative that basically turned into an automatic hold for any kid, Latino kid with a sharpie. So, somewhere there's bias loaded into that, but if you just came in the door and said, "You guys are racist and you're picking on Latino males," you're going to run into a lot of opposition. It's another thing to start peeling it back to say, "Okay, well, this is some of what we're hearing from your own staff. Public defenders, certain judges see these kids with markers and they think, 'gang membership'. And everybody follows suit. But when we've actually looked at it, that's not the case," and try to come at it in a way where people are going to be able to listen in hear.
WOLF:
Absolutely fascinating. Sounds like you're doing amazing work-
BARROWS:
Trying, trying.
WOLF:
... On a very difficult and complicated issue.
BARROWS:
Yeah.
WOLF:
I've been speaking with Tshaka Barrows, who is the Deputy Director at the Burns Institute in Oakland, California, which is working to address and diminish and eradicate racial and ethnic disparities in the criminal justice system. Thank you so much for taking the time to talk with me.
BARROWS:
Thank you.
WOLF:
I'm Rob Wolf, Director of Communications at the Center for Court Innovation here at the Quinnipiac University School of Law for our Justice Conference. And thank you very much for listening.