clock menu more-arrow no yes mobile

Filed under:

Towards a new approach to MLB’s domestic violence policy

I’ve written often about the problems with MLB’s domestic violence policy. Now let’s propose a solution.

If you buy something from an SB Nation link, Vox Media may earn a commission. See our ethics statement.

League Championship Series - New York Yankees v Houston Astros - Game Six Photo by Tim Warner/Getty Images

UPDATE: This afternoon, MLB Executive Vice President of Communications Michael Teevan sent the following email to me about the piece below.


I hope you are well. I saw your piece today and wanted to point out the following:

Respectfully, many of the suggestions within the piece are things that we already do. Among them:

- Our mandatory trainings already “cover all baseball personnel rather than only MLB players” on at least an annualized basis;

- These trainings already do “focus on prevention” and not “punitive justice”;

- The trainings always “prioritize the needs and goals of survivors.”

We have three policies in place – one covers Major League players and is a joint policy with the MLBPA; one covers Minor League players; and the other covers employees. We work with many different organizations in the space for our various trainings at the Major League and Minor League levels.

If we can ever be of assistance in clarifying our many efforts, please do not hesitate to reach out, because our staff takes this issue very seriously.


Michael Teevan

Vice President, Communications

Office of the Commissioner, Major League Baseball


This week, as the baseball world prepared for what should be its proudest moment—the World Series— the ugly specter of MLB’s failed attempts to deal with the increasingly prevalent issue of domestic violence reared its head yet again.

The Astros responded with this statement:

It’s not a secret that MLB’s attempts to deal with domestic violence have had less than stellar results, to put it mildly, and I’ve been extremely critical of the policy both here and for FanGraphs. Less than a month ago, I wrote in these very pages that MLB’s domestic violence policy had degenerated into little more than an attempt at public relations.

This, then, encapsulates the real impact that the Domestic Violence policy has had - and the lack thereof. Yes, the policy suspends - some - players for domestic abuse, and yes, offenders should be punished. But now, after a few years, we can fairly definitively state that to the extent the goal of the policy is to reduce the number and severity of incidents of sexual and domestic abuse, it has rather spectacularly failed. Because although it is sold to the public for that purpose, it is really just public relations: a way for the league to keep control over the narrative and keep its players out of prison when they do commit domestic abuse, with the limited axe that does fall targeting people of color. As an actual system of discipline for domestic violence, that really doesn’t work. As a public relations strategy, however, it’s working perfectly.

But simply pointing out flaws in a policy only goes so far. Yes, the MLB-MLBPA joint domestic violence policy seemingly has had no deterrent effect, and has had little noticeable effect in aiding survivors.

Yes, the policy seems to have caused some teams, like the Yankees and Astros, to view players suspended for domestic abuse as some sort of inefficiency, using the suspension as leverage to acquire those players at a lower cost. If not for domestic abuse suspensions, neither of the teams in the American League Championship Series would have their closers, and the Yankees wouldn’t have a 22-year-old budding star in Gleyber Torres. All of this is true.

And yet, at a certain point, it becomes less about identifying flaws in the current scheme and more about identifying solutions.

That’s what I’ve attempted to do here. I don’t pretend to have all the answers. There is no such thing as a perfect domestic abuse policy; we don’t live in a world where such a policy is even possible, and in such a world there would be no domestic abuse at all. Instead, what I’ve attempted to do is, as a lawyer, draft a set of proposed policies which would hopefully improve upon the current structure.

Let me start with what we’re not going to do. This will not be a punitive, zero-tolerance policy. Why? For a few reasons, the most important of which is that they don’t work, as University of Colorado Law Professor Aya Gruber wrote for the New York Times.

Like many crimes, domestic violence stems from myriad social, emotional and psychological circumstances that are exacerbated by incarceration, stigmatization and unemployment. Those subject to zero tolerance policies are disproportionately poor men of color. Many women who experience violence want to maintain (but improve) the relationship and do not want their partner deported, their children’s father jailed or their breadwinner to lose his job (just ask Janay Rice). Punitive sanctions may cause victims to avoid getting help because reporting activates the prosecutorial machine. Strikingly, Dr. Lawrence Sherman, the author of a famed 1984 study that found arrest to be more effective at reducing future violence than nonpunitive interventions, recently published a 30-year follow up revealing that the women whose partners were arrested had far worse life outcomes than those whose partners were not, prompting Sherman to question whether our fervor for punishing bad men may have prevented us “from testing what may be better ideas.”

Some data suggest that women of color are unlikely to report domestic abuse because they are far more likely to be themselves arrested. And as Cindy Southworth, the executive vice president of the National Network to End Domestic Violence told USA Today, zero-tolerance is particularly ill-suited to sports leagues as the basis for a domestic-violence policy.

“Counter-intuitively, we don’t want sports leagues to have a zero tolerance policy,” she said. “And the reason for that is if we would say that the first time your partner calls 911 your career is over, her risk of homicide shoots through the roof. Because he has nothing to lose and everything to lose at the same time. We’ve actually been advising the sports league to take a very swift, very robust approach but not to say that first-time and you’re out of it, your career is over because the pressure then on the victim not to call for help is massive. And we want them to be able to call 911. We need them to reach out for help.”

Southworth is correct: the Washington Post has reported, for example, that survivors of abuse by NFL players are less likely to report the incidents where punishments are more severe. And Diana Moskovitz concluded for Deadspin, after examining the relevant data, that “[t]hese women are trying to find what will make them safe again, but they know what won’t help. Sending their partners to the unemployment line won’t help them; creating a disincentive for victims to reach out for support won’t help them; and using victims as props in public-relations campaigns won’t help them. What they need is for their opinions to matter, and to be treated like human beings.”

That’s a good segue into the other half of the equation: not all survivors of domestic abuse seek help in the same way. The problem is that zero-tolerance policies create a one-size-fits-all approach to aiding domestic violence survivors that is inherently disempowering, as Leigh Goodmark wrote brilliantly for the Washington University Journal of Law and Policy:

Domestic violence does not transform every woman who experiences it into a stereotypical victim, nor should this victim stereotype shape domestic violence law and policy. Instead, anti-essentialist feminism compels us to delve into the complexities of the lives of individual women and consider the totality of who they are, rather than reducing them to their lowest common denominator—their common experience with domestic violence. Anti-essentialist feminism reminds us that women who experience domestic violence are more than the experience of that violence. They are rich, poor, middle class, African-American, Latina, Asian, white, Native American, immigrant, disabled, ablebodied, gay, straight, transgendered, rural, urban, self-defensive, aggressive, frightened, and angry. They have different goals, aspirations, concerns, and priorities. The solutions we develop need to be attentive to those complexities.

On the other hand, not punishing domestic violence is a similarly odious approach; research has revealed a strong linkage between abuse and employment factors. For example, in 1992, in an article entitled “CRIME, PUNISHMENT, AND STAKE IN CONFORMITY: LEGAL AND INFORMAL CONTROL OF DOMESTIC VIOLENCE,” authors LAWRENCE W. SHERMAN, Douglas A. SMITH, JANELL D. SCHMIDT, and DENNIS P. ROGAN investigated whether arrest specifically, or police involvement generally, had an impact on domestic abuse recidivism. They started by summarizing the three proposals for the basis of deterrence: conditional, replacement, and additive.

The conditional hypothesis. The conditional hypothesis claims that legal threats only deter potential offenders who are sufficiently tied to conventional society to suffer from its stigmatization of arrest.

The replacement hypothesis. This hypothesis assumes that the threat of legal control is effective only when informal control is absent.

The additive hypothesis. The additive hypothesis, derived from Wrong (1961) by Grasmick and McLaughlin (1978), claims that both infor- mal and legal controls deter potential offenders - the more of either type of control, the greater the deterrence.

Their results were surprising.

If we ask whether arrest influences the subsequent violence of those arrested, the answer is that, in general, it depends on the arrested person’s stake in conformity. Although arrest had a significant positive association with recidivism in the main effects equation, subsequent analyses that allowed the influence of arrest to vary with employment status and marital status revealed a different picture. Arrested persons who lacked a stake in conformity were significantly more likely to have a repeat offense than their counterparts who were not arrested. Conversely, among those who were married and employed, arrest deterred subsequent violence. Collectively, these findings are consistent with the conditional hypothesis and provide little support for the replacement or additive hypotheses.

Other studies found that mandatory arrest actually increased domestic abuse recidivism. In other words, punishing an abuser only works, generally speaking, if the abuser will feel a type of constructive shame as a result of the punishment.

Cornell Law School has proposed this model policy based on available research. That’s helpful, but it isn’t enough; professional sports leagues add a spotlight most employers don’t have. So our goal in drafting this policy is to create a system with these three goals in mind:

  1. Create a process by which survivors can seek assistance on their own terms;
  2. End the ability of teams to treat domestic abuse as an inefficiency to be exploited;
  3. End the “sliding scale” for intimate partner violence and establish equal, across-the-board treatment and replace it with a system of accountability for offenders built on constructive shame.

Let’s address each of these issues.

I. Pathway for Assistance

Section VIII of the Domestic Violence Policy (it’s on page 324 of the Collective Bargaining Agreement, if you’re inclined to look for yourself) provides for confidential assistance to be provided to survivors of domestic abuse.

A. Confidential Assistance Program. The Parties shall contract with a mutually-agreed upon domestic violence services provider to offer support services to Players, Players’ families, and victims on a confidential basis. The vendor shall maintain a 24-hour helpline (both in English and Spanish), staffed by Masters-level or highly experienced counselors with the capacity to refer callers to a health care professional in their local area with expertise in domestic violence and family counseling. With the exception of general usage statistics, the Parties shall not be provided with any confidential information regarding usage of the service by Players or their families.

B. Family Resources. The Parties, through the Policy Committee, shall develop a plan for the publication of referral information, websites, and resources (including hotlines, shelters and outreach facilities) for spouses, partners and families of Players in every Major League city, Spring Training site and in the home countries of all Players.

This is, on its face, not nearly enough. The policy doesn’t require a referral to these services until after the league has opened an investigation, and by then it may already be too late. Studies suggest that prevention of domestic violence and abuse is far more effective than punishment.

According to Audra Bowlus and Shannon Seitz of Queen’s University in Ontario, Canada, the two most important factors in preventing domestic abuse are (1) whether the would-be victim is employed, and (2) whether the would-be offender has received pre-abuse resocialization. Both of these factors can and should be addressed by a new joint domestic violence policy.

In other words, our new domestic violence policy should include mandatory one-on-one and/or group training of all major- and minor-league players, coaches, managers, front office executives, and agents, in Spring Training, at the All-Star Break, and after the season concludes, focusing on providing all attendees with tools they need to avoid behaviors which may be abusive in the future. The key here is that everyone who works in baseball, in any capacity, must attend, thereby avoiding the negative consequences of limiting this training to those already accused of domestic abuse.

These programs should be universal, thereby mitigating retribution risks. The policy should also provide for job training and placement for all intimate partners of attendees, helping them find financial independence. This should be in addition to the confidential assistance program, which should be empowered to provide financial and other support to survivors on a confidential basis, but expanded to include all minor league players, front office executives, coaches, front office executives, and agents. These programs should be funded with the monies forfeited by suspended players.

II. End the Use of Domestic Abuse As An Inefficiency

Whereas punishment and deterrence may not be the primary goals of our new policy, we must end the current loophole whereby players can actually be rewarded for their odious behavior. For example, both Roberto Osuna and Aroldis Chapman can be considered to have benefited from domestic abuse, going from struggling teams to contenders. By the same token, teams should no longer be able to trade for players at a discount as a result of domestic abuse.

There are two ways we can address this problem. First, any person suspended for domestic abuse under the policy should be immediately ineligible for postseason play, just as with a performance-enhancing drug suspension. It may also seem logical to bar that player from being traded entirely, but again, we’re trying to avoid draconian punishments which have the effect of increasing recidivism risks. As such, instead of barring a player from being traded entirely, we’ll state that trading for a player under a domestic violence suspension will void that player’s contract after the current year. In other words, if a team wants to trade for a player who is subject to a domestic violence suspension, that player cannot be used in the playoffs and will be a free agent after the season. That would essentially negate the incentive for the trades of Roberto Osuna and Aroldis Chapman moving forward. This also dovetails nicely with the reintegrative scheme below: reintegrate the offender into the life he had, rather than improving his situation as a result of wrongful conduct.

III. Create A Uniform Punishment Scheme

How do we punish domestic abuse when punishment itself can be an aggravating factor? Some studies have found that punishment is generally only an aggravating factor when unaccompanied by other measures.

As to the “good risks,” arrest may be beneficial regardless of the action or inaction of other parts of the law enforcement system. As to “bad risks,” it appears that arrest alone might be an aggravating circumstance. Even among the likely recidivists, the data from Milwaukee suggest that arrest does appear to have a short-term (one month) deterrent effect.31 But to achieve a longer-term deterrent effect, it actually may be necessary to enforce the law in other more lasting ways.

On the other hand, Cheryl Hanna is just one of many researchers to find that treatment of domestic abusers is generally ineffective in limiting or eliminating recidivism. That’s led to some proposing a middle ground, as Georgetown Law Professor Rachel Camp writes:

In 2017, a judge in Guilford County, North Carolina convicted three men of domestic violence misdemeanors. Their sentence? To spend hours over the course of several days standing outside the courthouse holding signs that read “This is the face of domestic abuse.” Though fervently criticized by some, formal shaming sanctions like this one from North Carolina are steadily used. Judges across the United States impose shaming sentences as an alternative to more traditional sentences in a variety of cases, including cases involving Intimate Partner Violence (“IPV”). While shaming wrongdoers may be identified as empowering for victims and may have utilitarian benefits on maladaptive behavior in particular contexts, within the context of IPV nearly all of the benefits are outweighed by shame’s tendency to undermine the goals of violence reduction and survivor safety. Shaming perpetrators risks making survivors more vulnerable, not less.

But not all shame is created equal, as Camp later writes.

When a person is denigrated or embarrassed—when he is deprived of dignity—that person is more likely to perceive the source imposing the directive as less legitimate and therefore, the issued order as unfairly imposed. One result from “sanctions imposed in a manner that harms a person’s dignity” is an increase in future offending. By contrast, a commitment by judges to defend the dignity of litigants and embrace procedural justice can lead to increased legitimacy of courts through a reduction in statesponsored denigration of wrongdoers and, for IPV specifically, increased survivor safety through increased perpetrator compliance with the orders intended to protect them.

It’s important to understand what we’re proposing here. Preserving the dignity of the offender does not mean failing to impose consequences. Camp proposes a “reintegrative shame model”:

Reintegrative shame is foundational to many restorative justice models. Indeed, many restorative justice models are built on materializing the utilitarian benefits of shame by using public- and community-driven conferences or circles to enforce standards of behavior. Restorative justice can provide survivors and perpetrators opportunities for interventions outside of the “standard procedures of criminal and civil law,” resulting in “other avenues of support for healing and accountability, rather than a short period of incarceration of the offender.” Indeed, many survivors identify community condemnation of the offense as more important than punishment, retribution, or rehabilitation. Restorative justice models allow victims to describe the harm they have experienced and ask the wrongdoer to publicly accept responsibility for that harm. Such interventions can help achieve vindication from the community “as a rebuke to the offenders’ display of contempt for their rights and dignity” and can help craft creative and targeted interventions designed to address the impact of, and impetus for, the caused harm. From the perspective of the wrongdoer, restorative justice can provide a supportive space for gaining empathic understanding of the victim’s experience, admitting responsibility, and being accepted back into his community.

On the one hand, an approach based in reintegrative shame has been demonstrated to lead to lower recidivism rates and higher levels of accountability by offenders. That’s because reintegrative shame prioritizes shaming the action without stigmatizing the person.

On the other hand, there are real concerns about voluntariness with reintegrative shaming, and the risk that it overly privatizes what some argue, quite convincingly, is and should be a public issue. Overly privatizing domestic abuse leads to a number of negative outcomes, as Donna Coker writes for the University of Miami School of Law.

In essence, restorative justice and reintegrative shaming require an alliance between the state and ‘communities of care’. Given the way in which crime policy is used to control poor people and people of colour, given the racist and classist practices of criminal justice officers, and given the way in which significant numbers of poor people and people of colour are locked out of electoral politics, establishing an alliance between the norms of an offender’s community of care and those of the criminal justice system asks a great deal of subordinated communities.

In other words, a purely reintegrative approach risks handing off dealing with domestic abuse and intimate partner violence to the very underprivileged communities most harmed by the neglect of these systems. It would be inequitable in the extreme merely to pass the buck back to the impacted communities and tell them to figure it out on their own.

There’s one additional wrinkle here. Requiring a public apology and acknowledgement of culpability may be the most effective methodology presently available, but it also poses real legal problems. No criminal defense attorney with more than one functioning brain cell would ever allow a client to make a public declaration that their client committed a domestic battery.

So what is the solution here? Unfortunately, there’s no universal approach that will address these issues. At the same time, the current “sliding scale” for intimate partner violence just isn’t tenable. The Commissioner shouldn’t have plenary authority to determine what is and isn’t a severe offense, and thereby substitute his judgment for that of survivors.

As such, I propose here a hybrid system. First, the new suspension system will mimic that of the performance-enhancing drug disciplinary system. Where an investigation reveals that a violation of the policy has occurred by a preponderance of the evidence (a “more likely than not” evidentiary standard), a first offense will receive an 80-game suspension; a second offense will receive a full-season suspension; and a third offense will receive a lifetime ban; players will not accrue service time whilst serving their suspensions, and they will be unpaid, with their salaries dedicated to the resources we discussed in Section I. This universal policy will end the sliding scale and demonstrate that domestic abuse will not be tolerated.

However— and this is key— offenders will be permitted to apply for reinstatement after serving half of their suspension (or one year, in the case of a lifetime ban) by showing that they have, to the satisfaction of the survivor in question, accepted responsibility for their actions. That private determination by the survivor protects the offender from concerns about criminal prosecution, whilst still prioritizing the goals and aims of the survivor.

The offender will also have to show compliance with all aspects of the prescribed treatment plan in the opinion of his providers, including regular attendance at group and individual counseling sessions, if ordered. Where such sessions are ordered, if the offender ceases attending after reinstatement, the suspension will go back into effect. In short, the offender will have a choice: make a reasonable, good-faith effort to accept accountability and improve, or accept an unpaid suspension for at least 80 games. Chapman, for example, would have served an 80-game suspension under this system unless he had made a real, good faith effort. This system incentivizes accountability through reintegrative justice. It may not work in every case - it probably won’t work in every case, in fact - but the data suggest it will be more significantly more effective than the system in place now.

* * *

This is, by no means, a perfect proposal. I would myself freely admit that it might not even be good. In the context of domestic abuse, “good” covers very few things indeed. What this proposal is, however, I believe, superior to the present system in almost every way: covering all baseball personnel rather than only MLB players; focusing on prevention and reintegrative rather than punitive justice; designed to reduce recidivism rather than public relations; disincentivizing the use of domestic abuse as an inefficiency; and trying as much as is possible to prioritize the needs and goals of survivors.

I also recognize this is not a finished product. I’m hoping with this proposal to start a conversation about how the domestic abuse policies of the future can and should look. I welcome your thoughts and comments.