WHEN TANICIA Goodwin was accused of the unthinkable — of slashing her 8- and 3-year-old children’s throats, dousing them with lighter fluid, and setting fires beside them — the natural impulse was to wonder how this could have happened. If you can’t blame a woman who is deeply mentally ill, you have to blame somebody, right? The school officials who didn’t recognize the breadth of the danger at home? The judge who returned Goodwin’s son to her custody? The lawmakers who didn’t pour more resources into child welfare?
But there’s another potential answer, much harder to accept: What if the system works, the rules are followed, the risks are correctly assessed, and something still goes terribly, tragically wrong?
This is a cold reality of the child welfare system: It is a matter of calculated risk, built around risk-laden lives. And while we don’t yet know all the details of Goodwin’s story, we know it was saddled with peril. She had been on welfare, suffered from mental illness. She was connected to the Department of Children and Families as a child. She once asked the agency to take guardianship of her son, and years later, petitioned successfully to get him back. Last May, after someone reported that she had physically abused the boy, she entered the DCF sphere again, accepting child care, parent training, and contact from social workers.
Should someone have noticed that the help she was getting wasn’t enough? Seth Kalichman, a behavioral psychologist at the University of Connecticut, has found that mandatory reporters, such as teachers and psychologists, are sometimes reluctant to act, uncertain of how to define abuse, aware of how overburdened state agencies can be.
But the Goodwin case isn’t strictly a question of resources. While the overall budget for the Department of Children and Families has fallen over the past five years, funding for social workers, management, and support staff has actually increased. The department has a lower caseload for social workers - 15.7 to 1 - than its stated 18-to-1 goal.
And in this case, the mandatory reporting rules probably didn’t apply, said Harry Spence, who headed the DCF - then called the Department of Social Services - for five and a half years, until 2007. The laws require reporting of actual abuse, not subtle behavioral changes, Spence said.
Spence has no direct knowledge of the Goodwin case, but said it reminds him of a case he faced shortly after he started at the agency. A woman with severe mental illness - and two kids - had been in the system for years, monitored, medicated, vastly improved. She lived with her mother and sister, whom social workers believed could provide support. Then, she went off her pills. Her mother and sister tried, and failed, to convince her to take her meds. Within a few days, she snapped, and killed her children.
Clearly, there were failed assumptions along the way. But as Spence said, “the real issue is a much harder one . . . which is the question about predicting human behavior.’’ A social worker could conclude, correctly, that a placement is 96 percent likely to succeed. But in four cases out of 100, something will go wrong.
This might sound like letting social services off the hook, and we shouldn’t, for the sake of the victims or future kids. But it’s also worth remembering that the cases that generate national headlines - this man in Washington, this woman in Texas - are horrifying, but statistically rare. A department like DCF oversees tens of thousands of kids every year. When things go wrong, there isn’t always somebody to blame.
Sometimes, tragic circumstances do prompt change. After a tragedy in Florida last year, the state child welfare department did a sober self-assessment, hired 100 new investigators, and improved training for counselors and investigators. There should be a similar reckoning here, in the wake of the Goodwin case. Maybe a mistake will be found, a systemic problem corrected. Maybe there will be some overbroad response, some effort to limit custody for parents with mental illness, some bid to eliminate risk completely.
But we should also accept the likelihood that — sadly, coldly, correctly — nothing will change at all.