"WE ARE ABOUT TO drive a stake into the heart of the Taliban," the U.S. military official in Kabul confidently declared. It was late January, three months after Afghanistan had successfully held its first democratic presidential election, and the mood in the capital--at least among American policymakers--was buoyant. Despite Taliban threats, voting had taken place in October 2004 without incident, as had President Hamid Karzai's inauguration in December. The security situation was calm.

Looking ahead, officials could cite further cause for optimism. As the weather warmed, formerly snowbound U.S. troops would be able to launch a renewed series of offensives against the Taliban, working in partnership with their increasingly capable counterparts in the Afghan National Army. An impending reconciliation program, meanwhile, would offer amnesty to fighters grown weary of spending their nights in cold, dark caves, and thereby draw them into the country's peaceful political process. Although countless problems still loomed on the horizon, there was a sense that the counterinsurgency campaign against the Taliban might at last be reaching its tipping point. Was victory in sight?

Eight months later, the answer would appear to be a resounding no. After a swell of extremist violence this spring and summer, the story from Afghanistan has become one of a revived, entrenched insurgency and flagging stability. With four months still to go in 2005, this year has already witnessed the greatest number of American military fatalities since the 2001 invasion to topple the Taliban, with over 70 U.S. servicemen lost since January. Even as the country prepares for its first parliamentary and provincial elections on September 18, peace and stability seem as distant as ever.

This whipsaw from hope to despair and back has become a standard feature of the Afghan war. How many times has the Taliban been declared on the verge of defeat, only to spring back to life? How many times, for that matter, have U.S. efforts in Afghanistan been declared on the brink of collapse, only to rebound? It's a safe bet that, should parliamentary elections go smoothly this month, a note of triumphalism will return to the rhetoric coming from Kabul.

This schizophrenia of perceptions has a host of causes, but much of it can be traced to the inherent difficulty of measuring progress in a guerrilla war. Tactically, there's no question Islamist insurgents have proven that they are stronger, better equipped, and more resilient than the U.S. military and its Afghan allies were predicting just a few months ago. Strategically, however, their impact is much harder to gauge.

One sure sign of the Taliban's resurgence is the growing sophistication of their attacks, especially the use of improvised explosive devices. The focus of insurgent operations has also diversified, with more emphasis this year on soft targets like schools, medical clinics, and government offices. It is estimated that over 1,100 Afghans have been killed in the last six months, including a half-dozen parliamentary candidates and clerics.

The Taliban's tactical successes stem, in part, from the increased freedom of movement they enjoy across the sieve-like border with Pakistan, where they have evidently deepened and expanded their safe haven this year. Captured Pakistani militants in Afghanistan, Taliban defectors, and even Pakistan's own Islamist opposition all allege that insurgent camps have reopened, facilitating recruitment and training of new jihadists.

Even without Pakistani sanctuary, however, it seems unlikely the Taliban would have remained a static force. "Why are the insurgents getting better? One U.S. officer told me, 'We've killed the stupid ones,'" says Kalev Sepp, a professor at the Naval Postgraduate School and coauthor of the official Army study on Special Operations in the Afghan war. "It's normal, and predictable, that the insurgents would become more proficient with time, just like our own soldiers."

Ironically, the most conspicuous evidence of the renewed insurgency--the American casualty count--is also the least revealing of its nuances. Despite the grim headlines this year's losses have caused, nearly half of U.S. fatalities in 2005 were the result of just two incidents. The first was the April 6 crash of a CH-47 helicopter in Ghazni in which 15 servicemen perished--a crash due to bad weather. The second took place on June 28, when a helicopter in Kunar province--attempting to come to the aid of a four-man Navy SEAL team--was downed by insurgent fire; three of the four SEALs were killed, along with all 16 on board the helicopter.

It's natural for such large-scale tragedies to capture public attention, but they have limited value in revealing underlying tactical trends. It is more esoteric, less accessible metrics--such as the quality and quantity of actionable intelligence flowing to troops, and the balance between attacks initiated by coalition forces and those initiated by enemy forces--that offer a better guide to progress on the ground.

In this vein, it's worth noting that one reason American and Afghan soldiers have clashed with insurgents so often and so heavily this year is that they have been seeking out these engagements, pushing into the mountain valleys in southern and eastern Afghanistan that previously had largely been left to the Taliban.

In Ghazni province, for instance, U.S. forces rarely ventured into the remote wastes of the Nawa district prior to this summer, despite the known presence of insurgents there. "We ignored them [last year]," explained one soldier in January. "It didn't matter. They were hiding. Was it worth our effort to send a company-plus there?"

That calculus has evidently shifted, however, with the dispatch of troops from the 82nd Airborne into Nawa and neighboring areas this August. The soldiers conducted raids and cordon-and-search operations over several days, attempting to flush enemy fighters from their redoubts.

The underlying rationale for these and related maneuvers is not to kill or capture insurgents, per se, but to keep them on the defensive in the run-up to the September 18 elections. It is the latter event, much more than the former, that U.S. military strategists maintain is the center of gravity for the counterinsurgency campaign. As one U.S. officer with recent experience in Afghanistan argued to me: "Are [the insurgents] going to be able to continue to mount isolated periodic attacks on Western interests and central government interests? Yes. Is it going to have a significant impact on the continuation of the democratic process and the state-building process and the building of Afghan institutions? Less and less so."

According to this line of thinking, the perception of political and economic progress in Afghanistan is ultimately the key to isolating extremist groups and fracturing their ranks. In the midst of this summer's violence, for instance, approximately 200 Taliban accepted the Karzai government's amnesty offer--driven to defect, in some cases, by revulsion at the insurgents' new, more extreme tactics. "I came back from Pakistan because people in Pakistan had commanded me to kill schoolteachers and parliamentary candidates and burn schools. I did not believe in this," one returnee told Agence France-Presse in August.

Encouraging words--but a measure of caution is in order. Even if the ebbs and flows of the Taliban insurgency are incapable of derailing Afghanistan's long-term development, plenty of other things still can. For instance, while considerable effort has been focused on carrying off the provincial and parliamentary elections as an "event," there has been comparatively little investment in the new structures they are ostensibly establishing. It is symbolically appropriate that the groundbreaking ceremony for the new parliament building in Kabul took place only in late August. Quite literally, we are supporting the creation of institutions without bothering to lay their foundations.

Just as the U.S. military has come to accept that the development of a functional Afghan army depends on its sustained, comprehensive nurturing--with embedded trainers and advisers, a massive web of logistical support, and delicate diplomacy--there is tremendous need for focused, long-term technical and fiscal assistance to the key institutions of the still-inchoate Afghan state. More than dispensing aid to humanitarian-minded NGOs, the United States and its allies need to dig deeper into the workings of Afghanistan's police, courts, parliament, and civil service bureaucracy.

Absent this commitment, the democracy-building and institution-building projects in Afghanistan are likely to begin to diverge, following the example of other states in the developing world where democratic frameworks coexist uneasily with weak institutional cores. As political philosopher Stephen Holmes noted nearly a decade ago of post-Soviet Russia: "Russian elections do not produce anything even vaguely resembling accountable or responsive government largely because of institutional weakness. . . . Elections in Russia, in fact, do not create power. For the most part, they mirror the power that already exists."

Holmes's observation that "liberal values are threatened just as thoroughly by state incapacity as by despotic power" is the perfect epigram for present-day Afghanistan, where victory ultimately depends not just on the dispatch of bullets and ballots, but on the emergence of state institutions that are perceived as more or less legitimate, just, and honest.

This is achievable--the Afghan National Army stands as proof--but it is also difficult, costly, and time-consuming. Is it, then, quixotic to suggest the United States can help the people of Afghanistan in building not just a democratic government, but an effective, accountable one, too?

Vance Serchuk is a research fellow at the American Enterprise Institute.

Next Page