I arrived in Kabul in September 2004 as the advocacy coordinator for CARE Afghanistan. I assumed, out of naïve arrogance, that my efforts would contribute to the development of the country. It didn’t matter that I had never been there before, didn’t speak the language, and didn’t understand the culture.

In the decade since, I’ve gained a little humility but at the price of nagging doubt—doubt about the role of the international development community in creating change in countries where we work.

We in the aid community have influence and power, especially when it comes to funding and implementing social programs. Resources flow primarily north to south, and not the other way around. This insulates us from the costs of our mistakes; in fact, it leaves us free to keep making the same mistakes.

Einstein described insanity as doing the same thing over and over again and expecting different results. We assume that our funding and technical expertise can transform the countries where we work, whether we’re fostering democracy, or improving social indicators around education and health. Yet in countries ranging from Afghanistan to Zimbabwe, sustainable change remains elusive. For every success story like Liberia, where a nascent democracy is hopefully taking root, there is a countervailing example like South Sudan, where independence was soon followed by civil war.

Creating systemic change is fundamentally a political process insofar as it involves power—who has power, how those in power act, and especially how those in power allocate resources. You can't build a functioning health system, much less a functioning democracy, unless the people and institutions in charge act to make it so. Or, more accurately, are convinced to do so.

Yet we in the international development community often do our best to write the messy world of politics out of the equation. Instead, we take political issues over the allocation of power and, by rhetorical slight of hand, turn them into technical problems. We don't talk about politics; instead, we focus on democratization, or governance, or anti-corruption, or civil society capacity building. And then we present ourselves as the necessary technical experts to address these issues.

This places us center stage vis-à-vis both the people we are trying to help and the political systems we are trying to change.

If we accept that international development issues are fundamentally political, we must accept that we are not the experts. Politics are inherently local, based on context-specific histories and networks. Outsiders will almost always know less about the relevant power dynamics than people who live inside those systems. Therefore, while we can claim credibility as technical experts, however tenuously, we cannot claim credibility as political experts.

There’s a similar arrogance lurking within our field’s dry terminology. For instance, we speak of empowerment as though power is something that we bestow or at least abet. Again, we cast ourselves in a central role.

As Gloria Steinem explained 30 years ago: “Power can be taken, but not given. The process of the taking is empowerment itself.” By washing our hands of politics, we also wash our hands of the fact that while political change is not necessarily violent, it is always predicated on pressure and an implicit threat.

To be clear, international aid in its current form is not useless. Development and humanitarian programs save tens of millions of lives each year—a woman survives childbirth, an infant doesn't suffer malnutrition, a child learns to read.

But to move forward, we need to view our work through a political lens. First, we cannot take a denatured, ahistorical view of development, where change results from technical interventions alone. We also can’t assume that change will occur in a 10- or 20-year timeframe—especially knowing the long, tortured processes that led to existing liberal democracies.

Second, understanding that systemic change flows from political change means accepting that we must be, at times, peripheral players. It also means accepting that technical expertise is necessary but never sufficient; it only truly succeeds when the political stars align. Long-lasting change happens only with the support of those in power, and no amount of technical advice will change their basic political calculations.

Finally, we need to be realistic about outcomes and impact. We need to live with ambiguity; we will struggle to measure the change we seek to catalyze—not only because of the timeframes involved, but because these systems are so complex as to make attribution for any specific development difficult, if not impossible. Just because we wish something to happen doesn’t mean that it will. And even if a long-sought change does occur, it might not have resulted from our efforts. We should embrace new methods of measuring impact, while understanding that no system can provide all the answers we seek.

I struggle with doubt about what we can accomplish, but I try not to give into cynicism. The international development community can help—we just have to be realistic about what we can do.