In light of the GOP victories Tuesday, immigration reformers are pessimistic about legislative changes and are turning their attention to President Obama. But even before any executive action by Obama, America’s immigration law has been changing, driven by our immigration court system.

For example, a few weeks ago, just as the media frenzy over the crisis at the southern border was dying down, the highest decision-making body in the American immigration court system made a surprising choice. In a case called Matter of A-R-C-G, the Board of Immigration Appeals (BIA) decided for the first time that Guatemalan women who are trapped in physically abusive marriages could potentially qualify for refugee protection. The question of whether domestic violence could be grounds for receiving refugee status had been unsettled in U.S. law for two decades. Now, it finally seems like the tide has turned.

Especially in light of intense public debate over the summer about whether Central Americans arriving at the U.S.-Mexico border were unauthorized migrants or refugees fleeing persecution, many migrant advocates heralded the A-R-C-G decision as an important victory. However, amid the celebration, it is worth remembering an important fact about the American asylum program: it has rarely taken giant, irreversible leaps. In the absence of major legislative action from Congress, asylum law has changed in steps that are small, limited and incremental. And each of those steps is heavily contested.

In my recent book “Let Me Be a Refugee,” I compare the ways in which various countries process applications for refugee status. It turns out that the United States has a uniquely politicized, legalistic, inefficient and unpredictable way of deciding asylum claims. These features of the American system tend to be exacerbated in cases like domestic violence claims, where the persecution is committed by a private actor, not the state. These cases do not fit neatly within the definition of refugees outlined in international law as people who are persecuted because of their race, religion, national origin, political opinion or membership in a particular social group.

Due to the fragmentation and instability of the American asylum program, we should not expect the A-R-C-G decision to ensure consistency moving forward. A-R-C-G was the result of two decades of highly politicized position-taking across multiple government agencies and presidential administrations. Under pressure from advocacy groups, the Clinton administration created guidelines suggesting that domestic violence claims could fit under the refugee definition. However, they stalled for so long in doing so that the regulations were not finalized before President Bush took office in 2001. The Bush administration refused to finalize them, leaving hundreds of women’s cases in legal limbo for eight years. In December 2008, Attorney General Michael Mukasey called on the BIA to make a precedent-setting decision on the question, but it took six years before the A-R-C-G decision was made. Under the Obama Justice Department, more women have been granted asylum for domestic violence claims but without any official precedent being set.

There is little reason to expect a different story as the ramifications of the A-R-C-G decision are interpreted and tested. Instead, we should expect legal battles, a heavy burden of proof and continued uncertainty.

First, there will be legal battles over whom the decision can be applied to. The A-R-C-G opinion was narrowly written, and it only applies to the very particular social group of “married women in Guatemala who are unable to leave their relationship.” An expansion to include women from other countries, even from Honduras or El Salvador, or unmarried women, will not be automatic. It will have to be extensively litigated case-by-case, and the outcome will depend heavily on the specific evidence.

Second, even for women who clearly fit within the group outlined in A-R-C-G, refugee status is not automatic. In each individual case, the asylum seeker must prove that she was abused, that she sought help from the authorities and that no help was given. Ever since Congress passed the REAL ID Act in 2005, the standards of proof in asylum cases have been high, requiring documentation and evidence that are difficult if not impossible for many women to provide. The precise requirements in domestic violence cases will be litigated over time, but for now, only women who are able to secure a skilled pro bono attorney are likely to succeed.

Third, we know that even when issues in asylum law get settled, they tend to be unevenly applied. Scholars who have unpacked the data on asylum applications have found a staggeringly wide range in the likelihood of decision-makers to grant refugee status, even when factors like the asylum seeker’s country of origin are the same. For example, one Immigration Judge at the Miami court accepts Colombian applicants 5 percent of the time, and another accepts 88 percent of Colombians. In other words, women who fit the framework of A-R-C-G, have proof of their abuse and have an excellent lawyer may not be granted asylum status because they have been randomly assigned to an immigration judge who routinely refuses applicants.

American immigration law moves in heavily contested fits and starts, leaving women fleeing gender-based persecution vulnerable. However, my research on the Canadian asylum system makes clear that it does not have to be this way. Canada has been a pioneer on gender-based asylum claims since the early 1990s. The Canadian Immigration and Refugee Board instructs its decision-makers to assume that gender-based persecution fits within the definition of a refugee despite the fact that large numbers of women experience it. The guidelines are applied consistently, with very little interference by policymakers or the courts. This approach has not opened the floodgates, but it has led thousands of women to find safe havens.

On balance, some may believe that a litigious, fragmented, politicized asylum program is preferable because it allows for vetting of issues and thorough screening. But incremental and uneven reforms take a toll on the people who must wait for change.

Rebecca Hamlin is an assistant professor of political science at Grinnell College. She is the author of Let Me Be a Refugee: Administrative Justice and the Politics of Asylum in the United States, Canada, and Australia.