On Sunday 18 September, Afghanistan holds its second general election since the overthrow of the Taliban in November 2001. These polls, electing members to a wolesi jirga (lower house) and to provincial councils, were originally envisioned to coincide with the October 2004 presidential election; they are central planks of both the “Bonn process” for post-conflict reconstruction and the constitution that emerged from the January 2004 constitutional loya jirga.

If successful, these elections may signal the establishment of key institutions of the new Afghan state, and an important move towards the end of the transitional situation defined by Bonn’s provisions. At such a juncture it is natural to assess Afghanistan’s state-building progress. However, such assessments, particularly from outside Afghanistan, have been hampered by two narratives which need to be examined before exploring the meaning of these elections for Afghanistan and for democracy’s future there.

Two narratives: Iraq, and the unconquered land

The first story might be called “Iraq-coloured glasses”, the post-2003 tendency to view Afghanistan as a generalised problem of occupation. For supporters of the Iraq adventure, progress in Afghanistan (often reduced to the holding of an election) is read as evidence that such occupations can – if fitfully – bring stability and democracy to troubled or troublesome states in the fictional “greater middle east”.

For the Iraq war’s critics, Afghanistan’s problems of insurgency, narcotics, and warlordism provide welcome backing for their opposition to assertive militarism by the US. The currencies of this narrative are a stream of figures about mounting insurgent attacks or increasing opium production on the one hand, and Taliban surrenders or numbers of police and soldiers trained on the other. Neither of these perspectives – essentially aimed at assessing United States policy – help clarify a complicated reality.

The second narrative describes a place ruled nominally by a weak central government and a president whose writ barely reaches beyond Kabul, where local commanders, narco-traffickers, and warlords hold sway virtually unrestrained. This view finds convenient journalistic resonance with the outside world’s persistent and orientalist vision of Afghanistan as the unconquered land, where proud tribal warriors have jealously guarded their autonomy, their arms, and their women for centuries against all-comers.

Neither of these stories brings us closer to the context in which these elections will take place, or their impact on the country. However, we may use their claims as a jumping-off point to the more nuanced reality of Afghanistan today.

A complex picture

These narratives draw attention particularly to security, opium and governance as the indicators of Afghanistan’s progress. It is true that violence has worsened in Afghanistan during 2005. The June suicide-killing of a pro-government cleric in Kandahar, bloody firefights between insurgents and coalition forces in Zabul province, and the downing of a Chinook with seventeen US Seals in July punctuate a rising death toll of Afghan police, army, election workers and their opponents that now tops a thousand in the last six months.

At the same time, this violence has been accompanied by the surrender of ex-Taliban leaders, and insecurity related to insurgency is mainly prevalent in a few provinces in the south and east of the country. Interestingly, according to national surveys, the places that a US soldier or United Nations worker are the most insecure are not necessarily experienced as more insecure by ordinary Afghans living there. In general, simple assumptions based on the scale of violence may be misleading.

Similarly, figures on opium production are cited to bolster the case that Afghanistan is making progress or collapsing into a narco-state. After dramatic increases in production following the 2001 conflict to 4,200 tonnes last year, little reduction is expected this year despite intensive eradication efforts and slower attempts to introduce alternative crops.

But there are two good reasons why aggregate figures on opium production do not tell the story adequately. The first is that opium production is only one factor in its influence. Large hoards of opium existed under the Taliban, and also exist today. Reduced production increases prices and thus the income and influence of big traders and the warlords and officials who back them, at the expense of ordinary growers.

The second reason is that the production of opium has been moving to the centre and north of the country, with related shifts in insecurity. So, while insurgency and banditry become more acute but less dispersed in the south, criminal and narcotics-related insecurity may begin to spread.

The greatest problems are where these forces combine – the joint international and Afghan disarmament body estimates that twenty five of the 100 dangerous armed bands in the country combine insurgency, criminality and drug-trafficking – but for now these are largely located in a few provinces such as Kandahar and Zabul.

The issue of state authority in Afghanistan is also complex, and conceals two coexisting and contradictory dynamics. The first, which exists on the internationally visible surface of Afghan politics, is a multilaterally-driven process emphasising the reform and strengthening of state administration and the security sector, the introduction of elected bodies at provincial and eventually lower levels, and the elimination of small and medium-scale corruption.

The values of this process are rationalisation, democratisation, responsiveness, delivery and eventually decentralisation. Its torchbearers are leading international agencies – the United Nations Assistance Mission in Afghanistan (Unama), the United Nations Development Programme (UNDP) and the World Bank – and it is bringing some results. For example, administrative reform has improved the size and timeliness of civil-servant salaries and is introducing gradually more merit-based appointments. While these results have primarily occurred in Kabul, they are increasingly visible in provincial centres such as Herat and Mazar-e Sharif.

Behind this state-building agenda lies a second dynamic, involving a much messier political game, in which the government – with American support where needed through figures like former US ambassador and sophisticated power-broker Zalmay Khalilzad (now ambassador in Iraq) – delicately manages a web of relationships extending down to the country’s 330-odd districts. Very few people know the true dynamics or the sensitivity of this constantly shifting landscape of provincial governors (four were transferred to new seats in July), commanders, former Taliban, and even municipal officials.

Two big changes in this arena have been the entry of Ismail Khan, the former self-styled emir of western Afghanistan, into the cabinet, and the appointment of the northern warlord Rashid Dostum to a central army staff post. These moves helped bring previously unreachable revenues from customs duties and gas to Kabul without major confrontation.

This game, unlike state-building, is based on the retention of centralised power by President Karzai to provide the clout to shift the pieces around the board, and reward potential spoilers for behaving. Without knowing how close these pieces are to upsetting the board, one cannot judge the wisdom of bringing these men towards the centre, sometimes against the desires of ordinary Afghans.

The meaning of elections

With respect to security, the elections will test whether the evolving situation heralds the emergence of two Afghanistans – one which is gradually and fitfully balancing local power with state capacity to provide services and eventually raise revenues, and another in which banditry, drugs and fundamentalist insurgency combine to prevent the establishment of an effective or legitimate state. Even with progress in some places, such a divide may become a threat in itself. An important clue to the future will lie in both the uncertain security of the polls, and the reaction to the results in different regions.

In the parliamentary race itself, the “apolitical” agenda of democratic statebuilding and the machinations of local power-holders are mixed in a volatile brew as candidates vie for position. Recent exclusions of candidates linked to armed groups have both removed some commanders while passing over others with such links.

The impact of the elections on Afghanistan’s future will depend on who the voters choose, and the assertiveness and ability of the new parliament in both balancing executive power and avoiding a descent into corruption and patronage itself. The parliament will therefore be the nexus of the two centralising and democratising tendencies that coexist in the country, as well as a new arena for power struggles. It will no doubt face obstacles, from within and without, in managing these tensions, but its success will be crucial.

The importance of these elections is thus not in their symbolic role as a milestone in the Bonn process, but rather in what they will tell us about security in the country, and in what the resulting parliament can do to reconcile conflicting dynamics in Afghanistan towards a coherent vision shared by electors, their representatives, and with some luck, the international community as a whole.