Western opinion is surprisingly untroubled by Algeria's anguish, which has lasted more than six years and claimed almost 100,000 lives. The two main reasons for Western indifference are hostility toward the Islamist rebels, who are widely seen as intolerant, and the opacity of the Algerian political system. To penetrate the fog and understand the crisis, one should focus less on the Muslim fundamentalists and more on the key player in Algeria's politics: the army.

Islamist guerrilla warfare broke out in January 1992, after the army canceled elections won by the Islamic Salvation Front (FIS). Blocked from running the state, the FIS was banned in March 1992. A subsequent crackdown drove its moderate wing to rejoin the radicals, who resorted to violence after the elections' annulment. The Islamists targeted military vehicles, barracks, the police, and government buildings. Another Islamist organization, the Armed Islamic Group (GIA), then appeared and went even further, killing intellectuals, journalists, women, and foreigners and massacring villagers in western Algeria. But the lack of information about the GIA murders bred widespread skepticism about the group's identity. Many observers suspect that the GIA is a product of the state's intelligence service, designed to discredit the Islamists. These suspicions have been heightened by the Algerian government's sharp refusal to allow any international inquiry into the massacres.

Violence is erupting in Algeria in an almost total information blackout. This is hardly by chance; the regime has always preferred the clandestine to the transparent. Only the tip of the iceberg, the most insignificant area of decision-making, is visible. Understanding Algeria's crisis and its possible resolution requires looking beyond the Islamist movement, the focus of most Western attention, to the country's political structure and to the real power in a terrorized land: the army. The Islamists are not the only ones resorting to violence.

The state's power, inherited from the war of liberation from France of 1954-62, takes two forms: the army and the government. This dates back to the dichotomy between the general staff of the National Liberation Army (ALN) and the Provisional Government of the Algerian Republic (GPRA). The GPRA was set up in 1958 to represent the National Liberation Front (FLN) abroad, mobilize the funds needed to organize the anti-French underground, and support the refugees who had fled to Morocco and Tunisia. But the general staff of the ALN was actually in charge of the revolution. When the war ended, the army "dismissed" the GPRA and took over the new state. Thirty-six years after independence, the government still implements the policies chosen by the army.

Although the army exercises sovereignty and sees itself as Algeria's supreme authority, with the Council of Ministers merely running the administration, it does not form a homogeneous whole. It is made up of several structures -- including the national police, the military security force, and various military districts -- that are all formally subordinate to the general staff but nonetheless retain some autonomy. In addition, the officers in charge have their own networks of supporters, which make them even more independent of the authorities. The government, of which the army and police are supposed to be the secular branches, is short-circuited by a system outside the official power structure.

This is why repressive measures against the Islamists are both uncoordinated and totally outside the law. Special military units in hoods arrest people, who then disappear without their families ever knowing which section of the army is responsible. This illegal state of affairs has arisen because soldiers are trained to believe that they alone are responsible for keeping civil peace. They need not account for their actions before the courts or the people.

The government, including the president, lacks the authority to punish the men conducting the antiterrorist war. Overzealous officers are accountable to no one, and no judge can autonomously investigate attacks or murders. The situation has slipped out of hand. Reports by international human rights organizations make it clear that the army's war against Islamist terrorism is being waged with no regard whatsoever for the law.

Military die-hards are constantly trying to ensure that there will be no turning back for Algeria. Their response to the rise of the FIS was murder, torture, and collective reprisals. Similarly, after the recent massacres were blamed on GIA fundamentalists, the army's secret services embarked on a strategy of terror aimed at bringing the FIS to its knees so the generals can impose their own conditions on the Islamists if the regime agrees to negotiations.


Algeria's agony has lasted for more than six years because the army is the backbone of the regime. In times of crisis, the top military brass meet in conclave. It was they who decided to cancel the 1992 elections after the first round gave the lead to the FIS and to nominate instead Liamine Zeroual for the presidency in November 1995. The generals' informal meetings are not reported in the press -- understandably, since the constitution does not provide for them.

The criteria for taking part in these conclaves are unknown. The participants are thought to include officers from the general staff and the heads of the central services of the defense ministry, the military districts, the national police, and the domestic security force -- in other words, men who have a certain autonomy in the way they use the forces of law and order. Given the importance of the decisions it makes, this informal assembly is, in fact, installing itself as a sovereign body. Increasingly, Algeria is run by a military caste.

The system works only if the military figure appointed as head of state cooperates. A president who takes literally his constitutional role as supreme head of the armed forces critically upsets the balance of power. Hence Houari Boumedienne's coup d'etat against Ahmed Ben Bella in June 1965, the forced resignation of Chadli Bendjedid in January 1992, and the tragic disappearance of Mohamed Boudiaf in June 1992.

According to the scant information available, the general staff and President Zeroual disagreed sharply in the summer of 1997. The generals suspected the president of wanting to negotiate a civil peace with the FIS, against the wishes of the commanding officers who had decided to cancel the 1992 elections. To frustrate the president, the army chiefs apparently negotiated a cease-fire in October 1997 directly with the Islamic Salvation Army, the armed branch of the FIS. The object of this maneuver was to prevent Zeroual from posing as the architect of peace in order to boost his popularity and acquire greater independent power. The recent revelations by deserters that the security services were implicated in some of the recent massacres and killings could be another turf battle between army factions, some supporting the general staff and others the president.

It is as if the army is above civil law. In fact, in terms of prerogatives and authority, the judiciary ranks below both the army and the civil service. Reduced to settling disputes between ordinary citizens, a judge plays no part in resolving conflicts between important figures. He is merely an official in the government administration, subject to its hierarchy, and does not apply the law as he would in any other constitutional state.

Yet the government still has some powers because it organizes the allocation of Algeria's financial resources. The army lays down the major economic guidelines, but the government divides oil revenues between various ministerial departments, determines the structure of investment, and chooses trading partners in Algeria and abroad. The government thereby legalizes the transfer of some oil revenue to its network of clients, both military and civilian -- an especially important dynamic now that the economy is moving toward privatization and joint ventures. Still, the government's composition reflects the political line of the army, whose various factions nominate their proteges as ministers. Those appointed have two briefs: to stand up for the general interests of the regime and to show their loyalty to the faction that appointed them.


The government is not directly responsible for security policy, which comes under the army's purview, but it lends a semblance of legality to all acts of repression. The anti-Islamist war is being conducted on three fronts: underground, in the economy, and in the media. By controlling the national media and information to the outside world, the authorities like to believe they have won the battle on the other two fronts -- as if smashing the thermometer was the way to lower the temperature.

Looking through Algerian newspapers, a reader might feel that he was living in a country where social and economic life was more or less normal, apart from occasional massacres by desperate bandits who are about to be apprehended. That is because even the "independent" newspapers are censored, and everything in them must be filtered through official channels. Journalists must depict the Islamists as criminals, and no one can report on the army's illegal and arbitrary methods. Editors and reporters have been imprisoned for publishing information "prejudicial to security and to the forces of law and order."

Countless questions about the killings of journalists and artists and the massacres of villagers remain unanswered. Given the reluctance of the authorities to throw any real light on these tragedies, rumors are rife about the identity of the terrorists. It is said -- and only a committee of inquiry could confirm or refute this -- that the regime is pursuing a strategy aimed at discrediting the Islamists, presenting them as criminals who rape and kill young women, strangle children, burn down schools, and murder intellectuals. These charges imply government involvement in the massacres.

What is the truth? Official information on the atrocities remains scarce. The perpetrators are never taken alive or brought before the courts. Since there is no freedom of the press, the media confirm the version put forward by the authorities. The army has no intention of letting an international inquest try to uncover the truth.


For a long time, the Algerian regime did not need a political party to perpetuate itself. The only function of the lone party, the FLN, was to manage the symbolic gains from the liberation war. But after the outbreak of riots in October 1988 over corruption and miserable living conditions, the army accepted the establishment of a multiparty system. The generals saw the reforms as a way to reinvigorate the FLN by holding elections that its rivals would not be allowed to win.

The regime tolerates opposition, however biting, provided it does not undermine the government's authority. The regime sees its legitimacy as rock solid since it resides in the army, which in turn derives its authority from Algeria's revolutionary struggle and from the need to guarantee the nation's unity and defend its borders.

For its part, the army believes that the composition of the government should simply reflect the forces that have emerged since independence, as well as the influence of powerful patrons. Elections are purely about formal powers, meaning government offices. The loyal opposition can compete as much as it wants, provided it does not question the preeminence of the army. Hence the discreet negotiations to win over various opposition figures by offering them government posts, which does no harm since they will wield no real power.

The opposition can oppose the government but not the overall system. The FIS was disbanded in 1992 precisely because its victory in the elections threatened the regime. The elections would also have been canceled had another party won by a similar margin. In the eyes of the powers that be, the purpose of a multiparty system and elections is to strengthen the regime by giving it a democratic gloss, not to replace it.

The main parties, then, can be classified by their attitude toward the regime. Parties supporting the regime include, first and foremost, President Zeroual's National Democratic Rally (RND). Set up in February 1997, it enjoys the logistical support of the administration, which ensured that it won the last election. Next comes the FLN, whose previous secretary-general, Abdelhamid Mehri, signed the Rome platform -- a 1995 national reconciliation pact between the Islamists, socialists, and others, endorsing free speech, electoral politics, and an end to violence. But Mehri was overthrown, and his successor, Boualem Benhamouda, denounced the pact, as did the army.

Other parties oppose the government but support the army, such as the moderate Islamists of Ennahda; the former communists of Ettahaddi; the Algerian Renewal Party, led by Nouredine Boukroh; the Social Peace Movement of Mahfoud Nahnah; and the Rally for Culture and Democracy, led by Said Saadi. The last two parties are heaven-sent for the army: the first provides it with an Islamist alibi, the second with a modernist one.

Finally, four parties oppose the regime outright: the FIS; the Socialist Forces Front, led by Hocine Ait Ahmed; the Movement for Democracy in Algeria of Ahmed Ben Bella; and the Workers' Party, headed by Louisa Hannoun.


Leaving aside the parties nominally in power, the RND and the FLN, which would probably not exist without the support of the administration, the position of the other political parties reflects their opposition either to the army or to the Islamists, and on this basis they have formed alliances that go beyond their ideological differences. For instance, the "Rome alliance," bringing together the FIS and other anti-regime parties, tried to cement the opposition in order to force the army to abandon the political role it has played since 1962. But a group of parties more favorably disposed to using the army against the Islamists strongly condemned this entente as an attempt to legitimize the dissolved FIS.

Another difference between the non-Islamist parties lies in their conceptions of democratization. Some believe the first step should be to neutralize the Islamists, even if that means temporarily supporting the current regime; in their view, democracy means social consensus on a set of values. Others see the main task as creating the machinery for a multiparty system, even at the risk of an initial victory by the Islamists. By January 1992, the debate was moot: the army had canceled the elections and the Islamists had opted for armed resistance.

The rigging of the elections since 1991 has masked the actual support for each group. The secularist parties are divided over their positions toward the Islamists, but the lack of free speech prevents an accurate assessment of the depths of these rifts. The inability of the secular parties to agree on a strategy for ending the crisis reflects the divisions between the social groups that compose the non-Islamist electorate -- civil servants, engineers, doctors, lawyers, and other liberal professionals. They all want an end to the single-party regime that the army imposed and that the FIS would like to emulate. They are regarded as democrats, even if some of their representatives try to justify the most extreme authoritarianism. Indeed, the adjective "democratic" has undergone a semantic shift in the media. It now refers to any individual or opinion differing with the Islamists. The RND, the FLN, and even the army are now described as "democratic."

The fact that the democratic debate makes no reference at all to religion limits its impact on the Algerian people as a whole. They see this omission as a sure sign that the non-Islamist parties have not broken with the regime. Public distrust of the secular parties is heightened by their recruitment from an elite that is not merely secularized but also French-speaking.

Yet the FIS' supporters did not vote against democracy. True, they disliked the existing regime, but they also wanted more democracy. Algeria's secular democrats did not understand the extent to which support for the Islamists reflected a desire to break with the army-dominated system and participate in political life. Conversely, the man in the street did not understand how the democrats could still call themselves democratic while supporting an army that had just canceled the elections won by an opposition party, especially since some democrats had discredited themselves by remaining silent about human rights violations suffered by the Islamists.

This attitude raises both a political and a moral problem. To become a rallying point, a party must establish its political identity in positive terms. Any movement that defines itself only in terms of what it opposes will find it difficult to mobilize support. Algeria's putative democrats therefore cannot confine themselves to denouncing the Islamists; they must articulate an ideology based on principles such as respect for human rights, a multiparty system, freedom of the press, and free and fair elections. In Algeria's current agony, a supposedly democratic party that does not condemn human rights violations against the Islamists has abandoned one of the principal values of its supposed political identity and thereby discredited itself.


In January 1992, the pro-democracy political groups were not powerful enough to prevent the army from embarking on repression or the Islamists from resorting to terrorism. Both sides were bent on their own physical survival. But another reason why the outbreak of violence was unavoidable was the absence of any national agreement guaranteeing the rights of the individual and those of the opposition in the event of an Islamist victory. The resultant conflict crippled and divided the democrats. Their abstract references to freedom of expression and moral condemnation of terrorism sounded like the pious hopes of impotent spectators -- or, worse, of accomplices in the descent into bloodshed.

By definition, a democrat rejects aggressive violence. An Islamist, however, may be prepared to kill in the name of a religious utopia that regards the individual as a terrestrial means to a celestial end. Similarly, the nomenklatura uses the means of the state and hides behind the law to kill in order to ensure its survival.

The democrats are simply less ruthless. As such, the non-Islamist opposition is in a painfully weak position: caught between an intransigent regime that tries to exploit the democrats' weakness and an Islamist movement tempted to perpetuate the one-party system. Not only can democrats not rally to either camp, but the victory of either would weaken the democrats even more. So it is in the interests of the secular opposition to return to the ballot box. All political movements -- Islamists very much included -- should negotiate a minimum consensus to consolidate electoral legitimacy, ensure freedom of expression, and establish an independent judiciary. For such a compromise to be credible, the army would also have to guarantee it. If the party that won the elections breached the new contract, that would justify military intervention, provided the army had meanwhile regained the trust of the people. The army would first have to abandon its opposition to a multiparty system and its belief that it alone is the true source of power.

In the final analysis, however, Algeria's crisis can be resolved only by ending the dichotomy that lies at its roots: between an informal real power and an impotent formal power. It is time to identify the real powers that be. The army must stop interfering in affairs of state. But that means that all the political parties, including the FIS, must first agree on a national contract that sets out the rules of the game and enshrines a multiparty system, civil liberties, and electoral choice. What is at stake is not just the honor of the officers but, above all, the political future of Algeria.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Lahouari Addi, a Visiting Professor of Sociology at the Institut d'Etudes Politiques de Lyon, is the author of L'Algerie et la Democratie. An earlier version of this article appeared in Le Monde Diplomatique.
  • More By Lahouari Addi