What I would like to do in this article is to consider some common approaches to modern torture and consider whether they are reliable guides for thinking about torture. More specifically, I set out to dislodge certain natural and tempting misconceptions about political violence today by simply laying out what we now know about torture. In the case of torture, these temptations are particularly inviting and so a different perspective much more difficult for us to appreciate. Insofar as I have a positive account of torture, its chief purpose is to act as an antidote to mistaken or misleading claims about modern political violence.

I will consider three approaches to torture in this article. These approaches are, what I call, the humanist approach, the developmentalist approach, and the state terrorist approach. What I will do is briefly consider the strengths and weaknesses of these approaches. I will focus first on the weaknesses and then return to their strengths. Finally, I wish to consider in the broadest way, how these approaches fail or at least fall short of serving as a guide to thinking about torture and how, by considering their strengths, we could turn our thoughts in a more promising direction.


Humanism involves the very strong claim that human beings are inherently worthy of respect. For many humanists, this claim cannot be separated from the creation of a vital public sphere, a sphere in which individuals treat and learn to treat one another with dignity. Briefly stated then, the humanist approach is the view that as societies become civilized and establish vital public spheres, barbaric practices like torture will eventually disappear.

Now in the nineteenth century, many humanists believed that such a public life was firmly established in Europe, and the prospects for the rest of the world looked good. Pointing to the disappearance of many ceremonial spectacles of penal torture in the colonies, they could assert optimistically that torture would shortly disappear throughout the world due to the impact of enlightened government.1 Today, this version of events seems wildly implausible even as a description of the nineteenth century. European and American police practices were notoriously brutal and these practices set the standard for colonial interrogations as well, notably in India and the Philippines.2 Since then, many Western societies have resorted to torture or, at the very least, training others in the use of torture including Germany, France, England, and the United States.

Humanists are hard pressed to explain these events. One can say that torture persists because modernizing societies have failed to adopt civilized norms. But this explanation simply sidesteps the main issue, namely, that European societies themselves have practiced torture in the last century. A more interesting explanation was advanced by Hannah Arendt in The Origins of Totalitarianism.3 Although Arendt affirmed the importance of a public sphere, she argued that this sphere was under considerable pressure from the bureaucratic tendencies of modern societies. For Arendt, torture appears whenever bureaucratic life overwhelms the public, democratic life of modern societies. I shall focus on Arendt's work not so much because it is unique, but because it is so representative of radical humanist writers. 4

Arendt's argument turns on a familiar thesis that there is an inevitable tension between bureaucracy and democracy in modern states. On the one hand, democrats needs capable bureaucrats to administer the law impartially and effectively. On the other hand, bureaucrats are specialists and are hostile to amateur politicians interfering with their work. Further, because bureaucratic work can be quite technical, democrats have a hard time regulating bureaucracies or making sure that work is being done properly. Bureaucracies, as a result, can expand in unregulated and sometimes anti-democratic ways. If democrats cannot do without bureaucrats, it is equally clear that democrats empower "a bureaucratically articulated group which in its turn may occupy a quite autocratic position, both in fact and in form." 5

Arendt explored the tension between bureaucracy and democracy further. What would happen if, as in Nazi Germany and Stalinist Russia, the tension was resolved as a victory of bureaucratic life over public life? In such conditions, she observed, men no longer interact as equal subjects deliberating on a common good but rather as objects within a huge chain of command. To put it another way, men and women no longer act according to the rule of law (a substantive understanding of human dignity). Rather, they act according to administrative rules and quotas, what Arendt calls instrumentalist rationality. 6 Arendt illustrated this point forcefully in the case of Eichmann. By learning to submit to bureaucratic regulations, Eichmann learned to disregard any residual notions of dignity he possessed. 7 The disastrous results - repression, torture and genocide - illustrated the workings of administrative terror. In the modern age, violence was no longer necessarily exercised by evil, cruel tyrants as it had always been in the past. Rather, it was exercised by bored yet dutiful bureaucrats, pointing up the fact that, in the modern age, evil was banal. And what was most troublesome was that evil of this sort was imbedded in beings who were indispensable for modern political systems, bureaucrats.

This sort of account is certainly a gripping one, but as an account of modern torture, it is quite misleading. The difficulty is that writers in this genre often simply denounce terror rather than explain it. 8 They do this despite their desire to clarify how torture works today. In fact, they are driven to do so by the very way in which they talk about the world.

The humanist approach turns on a key distinction, the distinction between the public realm and the realm of administration. Actually, this simple distinction hides a remarkable dissymmetry. In the humanist view, the ideal of public life is set forth and everything which is not part of this life is labeled "administration." The notion of rule by bureaucracy is constructed not on the basis of observations, but rather as a hypothetical opposite to rule of law. This fact is reflected in the problems that arise when we try to use the humanist distinction empirically.

Torture is, no doubt, administered by bureaucracies, but it is exercised in remarkably different ways. The manner in which torturers act on victims can vary according to the type of rationality that characterizes an administrative system. However, in the humanist approach, all these modes of governing individuals are thrown together in the same category, although they share little more than the label "administration." And to the extent that humanists ignore these differences, they have a harder time explaining what is going on in torture bureaucracies.

Let me illustrate this point by noting briefly the variety of different administrative rationalities and how these are related to torture. In some bureaucracies, human beings relate to one another solely as objects to be manipulated, that is, they act according to instrumentalist rationality. Yet there are also bureaucracies in which people are treated as subjects to be transformed, converted, or healed. Examples of such bureaucracies might be church or psychiatric bureaucracies, all of which have had historical roles in the development of torture. Indeed, in the fifties, one of the key ways in which torture changed was with the introduction of psychological warfare and, with it, the participation of a new group of specialists in torture. The purpose of these specialists was not to treat their victims as objects of punishment, but rather as subjects whose perceptions had to be altered. 9

Humanists, of course, are less concerned with these modes of administration than they are with modes of administration that treat people as objects. But it is a mistake to think that because some bureaucrats treat persons as objects, there must be a single rationality that characterizes all their actions. There are at least three different types of instrumentalist rationality, that is, three different ways in which people learn to treat other people as objects. 10 Each of these describes a particular type of administrative structure. One can, for instance, treat someone as a means to an end, as one does in a military organization or a post office. Or one can interact with them as part of a system, as people interact with one another in an information network. Or one can interact with people as opponents to be strategically defeated or won over, as one does within a policy institute.

Torture may be characterized by one or another or some combination of these instrumentalist rationalities. One may torture an individual for confessions, that is, use them as a means to an end. Torturing people in this way seems to be particularly associated not so much with lax judicial systems but, rather, with judicial systems with particularly rigid and severe standards of legal proof. This is because, in such systems, confessions are often the only sure way of obtaining a conviction. 11

Or one may torture individuals simply to set an example to others. Thus, we have the spectacle in many countries today of tortured dissidents appearing on national television to praise the regime's policies. There is a semeiotics of torture here that needs to be deciphered, one in which individuals and their torturers operate as part of a large media spectacle. This sort of media attention can raise the climate of fear and suspicion and, sometimes, governments that torture welcome independent coverage by journalists. Amnesty International reports that in Guatemala, newspapers are allowed to publish pictures of dead torture victims, although the accompanying articles are faithful to the government line. 12

The main point is that in such a system, torture does not cease when individuals confess because, in most cases, individuals have little or no information to give. They are of no use except as parts of a system of media representations. And that is why they are tortured.

Finally, torture may be characterized by strategic rationality, especially when it occurs in the context of counter-insurgency warfare. For example, during the Huk campaigns in the Philippines, torture was part of a carrot-and-stick approach to win the support of the peasantry and to strategically undermine the rural support of Huk insurgents. 13 What is worth emphasizing here is that torture was not used primarily to win confessions or to use individuals as part of a communications system, but, rather, to strategically outmaneuver an opponent. Individuals could be tortured even when torturing them served no intelligence or public relations purpose. The main point was simply to make the environment too hostile for political opposition to operate and one way of doing this, it would seem, was to torture peasants.

So torture can be exercised by different sorts of administration characterized by specific forms of rationality. The torturer may act on the tortured as a priest seeking a conversion, a surgeon operating on a patient, a psychiatrist transforming a subject. He may explain what he does in the same terms as a detective, a publicist, or a counter-insurgency expert. Each of these ways of acting describes a distinctive mode of government and casts an entirely different light on how torture operates in a society.

Can humanists integrate these insights into their explanatory scheme? Humanists can acknowledge that they conceived of administrative torture too unclearly but, in any case, torture does occur when administrative rationality overwhelms public, democratic life. The test of this, of course, would be to show that torture decreases as democrats gain more control of public life. But the record on this score is more than a little mixed. Torture has been practiced by many Western democratic states such as France and continues today in others. 14 Perhaps humanists have in mind an ideal of public life rather than contemporary democracies; but in that case they should not give the impression that they are trying to explain torture.

But even if we accept the humanist approach on these terms, I fear the normative ideal of public life serves as a poor guide as to how to struggle against torture. On the humanist approach, the key problem is to establish rule of law and basic democratic institutions; democrats will then solve the torture problem. 15 I am not persuaded that this argument is applicable to more than a handful of countries today, forgetting as it does so many states which torture and have little prospect of becoming democratic. More than that, I would say this argument is misleading; it asks us to struggle against torture in the wrong places. If torture is shaped by administrative rationalities, then human rights groups must struggle directly with these rationalities of modern administration - as, in fact, they do. 16 Finally, even if democracy is re-established in repressive societies, this does not put politicians in a better position to confront the administrative groups responsible for torture. 17 At best, the struggle for democracy may be related to the struggle against torture; but it is not identical with it for it cannot provide the tools for confronting modern administrative rationalities.


The developmentalist approach is concerned with the dynamics of economic modernization. Briefly, it is animated by the belief that as societies modernize, there is a decrease in the corporal severity of punishments. This decrease does not necessarily occur because people become more enlightened. Rather, it is brought about by the rationalization of economic and political life. As individuals are introduced to civic and labor discipline, they would learn to regulate themselves according to their conscience. 18 External sanctions, at least barbaric ones, are no longer necessary to maintain order.

This argument had a tremendous influence on social scientists studying the process of modernization in developing countries. Nonetheless, these scholars disagreed among themselves as to the kind of development that should be emphasized. Some believed that economic development itself allowed for a decrease in the severity and extent of violence over time. 19 Others - notably Huntington, Pye and Olson - argued that economic development itself had a destabilizing effect on societies and promoted violence. 20 On the latter view, the real cause of violence was the lack of political, not economic, development. Developing states were simply unable to broaden the scope of political participation fast enough to meet the new demands placed upon them. As a result, disaffected groups turned to violence and governments resorted to harsh sanctions including torture.

As the second view was the one that eventually became orthodoxy, I shall be concerned mainly with the way its adherents deal with torture. For modernization theorists, the question was this: how could governments expand the scope of political participation while in a potentially violent and unstable situation? Huntington and Pye placed particular emphasis on counter-insurgency warfare. 21 Counter-insurgency warfare not only contained the extent of random civil strife but also facilitated the process of political development. This was because, by training police and military to work effectively, one created individuals who regulated themselves internally rather than by threat of external sanction. Such self-regulating individuals were crucial for the development process:

The capacity for coordination and discipline are crucial to both war and politics and historically societies which have been skilled at organizing the one have always been adept at organizing the other. 22

State warfare then plays a role in the formation of disciplinary habits, habits which in turn facilitate political and economic development. "Discipline and development," as Huntington says, "go hand in hand." 23

From this perspective, the violence of the state can be characterized as an unfortunate, necessary, and (in the long run) beneficial response to high levels of civil violence in developing countries. 24 To be sure, counter-insurgency warfare is violent, but it is far more preferable than government torture and civil strife. As Huntington puts it, "the civil violence which development produces is not the violence that produces development." 25 And if counter-insurgent troops did on occasion torture, this was clearly a technical problem of providing adequate resources and training to troops on the field: good soldiers would simply know better.

Now it is reassuring to know that torture is a temporary aberration, a dysfunction produced by the lack of political development. The trouble is that, while economic modernization can lead to increased civil strife, there is no clear relationship between civil strife and the extent to which a government resorts to torture. In certain regions, notably Latin America, civil strife is strongly correlated with other factors. States that do torture are also mainly states that receive the highest per capita foreign aid prior to repression; or military regimes that receive counter-insurgency training; or regimes with weak organized labor and high levels of direct foreign investment. 26 One can hope that military training can reduce random violence in the long run, but this does seem to be a dim prospect indeed.

Furthermore, development theorists suggest that when governments resort to torture, this is primarily reactive. The trouble is that at least in Latin America and the Middle East, the pro-active element is an important feature of government repression. Much of this is due to the impact of counter-insurgency training or, more recently, narcotics control training. For example, in Brazil, the military had always played a role in maintaining internal order. However, counter-insurgency training intensified the military's interest in internal, rather than external, warfare. Once the military was shaped in this way, soldiers set about performing this task in a professional disciplined way. And this meant that rather than wait for a real incident of violence, paramilitary forces sought out social groups that they suspected might protest government actions. 27 Counter-insurgency training does entrench better organizational skills, but this does not mean that it will also reduce government violence against civilians.

And this leads to a third problem with the developmentalist approach. Development theorists suggest that torture occurs only in proportion to the extent of civil opposition whereas, in many cases, torture continues long after there is no organized civil opposition to the state. In Argentina and Iran, for example, the intensity of government violence seems to bear no special relation to the extent of civil opposition. These states tortured systematically and intensely even when the opposition had been decimated. 28

Why is the developmentalist approach so misleading about the world of torture? The reason has to do with the way in which development theorists think about the determinants of political stability. For them, political stability is based on the consensus of the political community. And this consensus grows to the extent that political systems allow for popular participation and effectively meet social demands.

Now while political community may enhance political stability, it is not the only factor that conditions political stability. Take the example of South Africa. The lack of political consensus in this country may explain the high levels of civil strife, but it does not account for the continued stability of the South African regime. Its continued persistence raises an interesting theoretical point, namely, that the lack of political community can be offset by the efforts of a determined minority as long as it is willing and able to repress a population. 29

Most states today do possess large and well-armed standing armies as well as efficient means of transport and communication. Many of these states, notably those in Africa, Asia, and Latin America, are led by rather determined elites who are quite willing to employ force to achieve their ends. The difficulty with the developmentalist approach is that one finds it hard to conceptualize such states within its framework. If such brutal states persist, it must be because they enjoy some measure of political community. 30 But of course this doesn't have to be the case at all. It's just that development theorists draw the conditions for political stability too narrowly. They assume that only the consent of the governed is sufficient to ensure persistence of a regime. In this way, development theorists turn a blind eye towards the commonest experience of the last century, namely, the creation of military-industrial states which have deliberately set about deporting, torturing and exterminating their populations.


Although death squads and terrorist groups may torture victims, only states have the resources to torture systematically. States possess the financial and human resources to sustain a torture complex. Further, they can rely on support from other sectors of society to provide technical support (hospitals and mental asylums) and information (universities, unions, or the criminal underground). If state officials are responsible for most tortures, then state terrorism ought to be the focus of analysis and not civil strife.

Analysts who adopt this perspective work within what I call the state terrorist approach. Rather than survey this literature, I shall focus on the work of Noam Chomsky and Edward Herman, analysts who have done more than many others to provide a coherent picture of state terror in the late twentieth century. 31 Of course, Chomsky and Herman do not attempt to explain state terrorism everywhere, but only states closely allied with the US. But what is especially distinctive about their approach is that they put the focus firmly on the "economy of violence"32 in these countries and not, as many others do, on the economy of exploitation.

Chomsky and Herman draw attention away from torture to what might be called a torture complex. Torture, they point out, is "a mode of governance"33 characterized by "standard operating procedures in multiple detention centers, applicable to hundreds of detainees and used with the approval and intent of the highest authorities." 34 There is some dramatic evidence to support this claim. In 1978, the Vietnamese routed the Khmer Rouge from Kampuchea. The retreat was so sudden that the Khmer Rouge left behind an intact torture complex, the Tuol Sleng prison. Amnesty International researchers have uncovered remarkable material at Tuol Sleng including torture manuals; biographies of torturers; very thorough prison records; detailed accounts of interrogations, confessions, and medical examinations; and elaborate diagnostic flow charts showing the interrelationship of enemy networks based on the forced confessions. 35 Similarly, in 1975, human rights investigators visited the Second Army Headquarters in Sao Paolo, a site of numerous human rights abuses. They found the headquarters to be "a huge torture complex which has at its disposal the most modern and sophisticated equipment, and which requires an increasing number of staff--jailers, drivers, executioners, typists, public relations officers, doctors and others--to run." 36

Chomsky and Herman also point to the international side of torture, including the arms suppliers, the various foreign governments that provide training, and the way torture complexes interact on a global scale. Again researchers have gathered evidence that gives qualified support to this thesis. 37

Finally, Chomsky and Herman link torture complexes to the international economic system. On their account, torture appears due to the creation of national security states in developing countries. The task of these states is to crush class protests while maintaining economic growth on behalf of multinational interests. In short, torture exists in order to maintain labor discipline and to keep the cost of labor within a range acceptable to capitalist interests. Thus, Chomsky and Herman have no difficulty explaining why government violence is pro-active or, for that matter, why this violence is disproportionate to the incidents of civil opposition.

Nevertheless, Chomsky and Herman do describe the operation of torture complexes in the language of economics, and it is here where difficulties arise. No doubt repression may be useful in maintaining exploitative economic relationships, but this does not clarify the character of repression. Why, for example, is torture employed over a more intensive disciplinary system? And what accounts for changes and variations in a torture complex, for example, from prison-based torture to the psychoprison?

Moreover, while economic rationality explains why businessmen support regimes that torture, it does not explain the behaviour of those who torture. A torture complex is in fact a very costly and inefficient system by economic standards. It is costly because it involves large expenditures of money, equipment, and training for torturers as well as the maintenance of extensive detention facilities. Yet despite these expenditures, torture complexes are remarkably inefficient. Within them, discipline becomes lax. Bureaucrats and guards stop performing their duties and become more involved in the pursuit of personal profit and pleasure. These activities may range from forms of sadism and blackmail to large scale prostitution or drug smuggling operations. The Tuol Sleng "Interrogator's Manual" illustrates some of the problems that administrators confront in running a torture complex. At one point, the instructor tries to explain the ethics of torture to his students:

The purpose of torturing is to get their responses. It's not something we do for fun. We must hurt them so that they respond quickly. Another purpose is to break them and make them lose their will. It's not something that's done out of individual anger, or for self-satisfaction. So we beat them to make them afraid, but absolutely not to kill them. When torturing it is necessary to examine their state of health first, and then whip Don't be so bloodthirsty that you cause their death quickly. You won't get the needed information. 38

Here the instructor lays out a basic paradox of torture. Torture involves the destruction of human beings, but a torturer fails if he allows the victim to die. Torturers must have a particular discipline to keep their victim in pain and useful for political purposes. Yet torture just encourages the loss of this self-control. As the instructor explains, "our experience in the past has been that our interrogators for the most part tend to fall on the torture side. They emphasized torture over propaganda. This is the wrong way of doing it. We must teach interrogators how to do it." 39

So torture complexes encourage indiscipline and this can pose tremendous difficulties for governments that torture. The Brazilian military government, for example, phased out torture in part because it needed to bring "those undisciplined military and police personnel under at least a degree of central government control even though their was little effort to punish them for their crimes." 40 But this is not all. It is not entirely clear whether torture produces any documents that are particularly useful. Torturers do not seem to be more successful than regular policemen in securing confessions or reliable logistical information and have greater difficulty than police in keeping their victims alive long enough to provide such information. If Tuol Sleng is any indication, torturers use the confessions they gather simply to invent fanciful anti-government plots, plots which help them justify their continued operation.

Finally, it is not clear how torture induces economic productivity. There are two separate issues here. First, does torture induce greater productivity in labor? Well, if an economy relies primarily on unskilled manual labor, then one might agree with Marx that torture is rather directly related to economic productivity: human beings might work harder in order to avoid torture and so produce more. 41 But torture seems out of place in an economy that relies on skilled labor and industrial discipline. 42 The problems associated with low labor productivity in Third World countries (lack of any real skill training and long work hours) do not seem to be problems that torture can address. In Iran, torture was so unhelpful that SAVAK not only tortured workers less but also tried to lobby to improve salaries and managed welfare programs so that workers would produce more. 43

But perhaps states employ torture to destroy labor unions and thus keep wages in a range acceptable to industrialists, multinational corporations or the IMF. 44 There is a great deal to be said for this account of why states repress labor unions; but it does not explain why states resort to torture. It is worth recalling that the most famous repression of labor agitation, the repression of the French Revolution of 1848, was conducted without torture by an extremely vengeful and powerful police force and this was certainly not because torture was unknown to French police and military officers. 45

States may perceive the need to repress the labor unions, but this in itself is insufficient to explain why states resort to torture. Torture is not essential to such repression; more intensive discipline could be just as effective. Indeed, disciplinary punishments would seem to conform well with the demands of an industrial system, more so at any rate than torture.

In short, torture is difficult to explain solely in economic terms although the temptation to do so is simply overwhelming since Marx. We can use economic motives to explain why businessmen and state officials support torture, but we are at a loss to use economic rationality to explain what goes on in torture complexes. To claim an economic rationality for all this is not simply misleading. It also seems to provide a rationale for the behaviour of torturers, a rationale which makes their behaviour a little more acceptable. No doubt, many torturers also prefer to claim that they are fulfilling an essential function in the process of economic and political modernization.


None of these criticisms are meant to detract from the stronger features of these three approaches to modern torture. In fact, I think researchers have grasped key features of torture and this despite their constant urge to misunderstand violence today. For none of the problems we encountered using these approaches were empirical ones; rather, these problems were solved by looking to the workings of language and showing how the vocabulary of each approach tugged our attention away from the world in which we live. And I think that if we consider their strengths from the right perspective, we can gain a tentative way of understanding modern torture.

The strength of the state terrorist approach is that it is concerned with torture as a mode of government. It draws attention away from incidents of torture to the systematic and international nature of torture today. But its chief difficulty remains moving beyond statistics and personal narratives of detainees to descriptions of this mode of government. And this difficulty arises because it is hampered by an unduly narrow conception of torture as a form of economically rational behaviour.

In this regard, the developmentalist approach is much stronger. It focuses on counter-insurgency practices. These practices describe some of the military, police and bureaucratic conventions employed by those involved in torture. However, because developmentalist theorists regard torture as a pre-modern or accidental feature of politics, they remain peculiarly blind to the relationship between counter-insurgency practices and torture. All they can see is a technical failure in discipline, not the socio-political problem of the emergence of torture.

This socio-political problem is not lost on the humanists or the state terrorist theorists. Although they have a hard time explicating what goes on in torture complexes in detail, they do make the point that torture complexes are part and parcel of modern political systems, that they are constituted out of the same sorts of rationality that also characterize bureaucratic systems. They do not arise accidentally in the twentieth century. They are not aberrations of the modernization process or remnants of traditional society.

To put these insights more constructively, torture is a part of modern life in the sense that torture complexes can be best elucidated by means of ordinary features of modern life. Torture complexes are closely related to bureaucracies, prisons, hospitals, and, to a certain extent, factories. And sometimes we can understand what people are doing when they torture by comparing their activities to these related institutions.

Indeed, sometimes the similarity between other practices and torture is so close that it poses tremendous difficulty for action against torture. Medicine, for example, has a particularly close relation to torture because interrogators use medical practices, hospital facilities, and doctors in their work. 46 Consequently, other doctors have to overcome major obstacles in order to help victims recover from the trauma of torture. Doctors use too many techniques that are similar to the kinds of tortures experienced by victims. This applies not only to actual physical techniques, but also to the ways in which doctors examine and interrogate their patients. Doctors have to exercise tremendous care in the way they speak and treat their patients for the patients themselves can no longer effectively distinguish between what is torture and what is medicine. 47

Of course, we can evade the deep interrelations between torture and modernity by invoking the term 'developing societies' (logically, something belonging to the past of the 'developed' world and, consequently, saying little to the present of the developed world), but we should be aware that, in this brave new language, "becoming modern" no longer functions as an empirical category but, rather, as a moral one. The humanists never tire of insisting on this point. But this means that we are left with a vocabulary through which we can re-affirm our confidence in the celebrated aspects of modernity, but only at the price of being unable to talk about torture except in the vaguest way.

If there is one difficulty that runs through all these approaches, it is that none of the approaches provide satisfactory explanations for modern torture. The causes they identify are too general to account for the details of torture complexes. Nevertheless, they have a tremendous hold on our imagination for they articulate visions of the kind of political community in which we aspire to live. And it is the power of these approaches to grasp our moral imagination that accounts for their persistence today, not their supposed ability to provide a detailed explanation of what's going on in the world of violence or how we got here. Our concern with the ideal turns our attention away from the present and so defers a pressing question: in what ways and at what cost can we get from the world of torture to the ideal of political community? In particular, we should pay less attention to the question "why is there torture today?" - a tale which may be too tangled and deeply enshrouded in state secrecy - and much more attention to "how does torture work today?" For it is by increasing our scrutiny of how torture works that we can understand how we benefit (or don't) from torture as well as to what extent we can (and can't) challenge this practice.

Of course, I've been arguing that torture does not work in any of the conventional ways theorists suppose. Torture does not increase labor productivity, produce any better intelligence results than ordinary police work, and has a detrimental effect of social and bureaucratic discipline. And I want to take a moment and spell out the implications of this thesis.

One implication is that if torture does not work, then there is little justification for the use of torture. Torture's apologists always assume that torture works; for them, it is just a matter of moral justification. 48 Now it appears that even this assumption can be questioned. However, if torture does not work, then what needs to be explained is why the practice of torture persists today. The answer to this is clearly beyond the scope of this article, but I want to point out some ways in which this question could be answered. Perhaps it is because torturers are just protecting their jobs. 49 Or perhaps intense social fear mobilizes social elites to engage in "permanent counterrevolution"50 in which torture may play a part. But I think a more promising explanation is suggested in Michel Foucault's work on the prison. 51 Foucault asks what purpose is served by the persistent failure of the prison. He replies that although the prison does not reform offenders, its failure serves to extend disciplinary power and re-inforce its legitimacy.

If this is a credible explanation of why torture persists, then we need to pay more attention to the relationship between torture and the process of rationalization. In part, this involves examining the relationship between torture and other social institutions. In his book on torture, Edward Peters argues that, historically, torture was not a primitive practice that survived into the medieval period. Rather, torture was introduced in Europe in opposition to tribal punishments and its practice served to rationalize state power. 52 My own research on torture in modern Iran seems to suggest a similar point, 53 but it is worth investigating how the practice of torture today is related to different processes of rationalization.

At the same time, we need to pay careful attention to the process of linguistic rationalization. This means that we need to be careful how the use of increasingly specialized ways of talking can serve to mislead us about what is actually happening when torture occurs. Torture seems to thrive not so much on this or that ideology, but rather on gossip, rumor, media sensationalism, as well as bureaucratic, social scientific, and legal jargon. 54

Torture needs all the publicity it can get, but we have to be more careful about how we speak about torture. This does not mean that we should abandon the traditional ways we discuss torture, only that we should critically evaluate them at every opportunity.


1. See, for example, Henry Charles Lea, Superstition and Force (Philadelphia, 1870; reprint ed. New York: Haskell, 1971) and W.E.H. Lecky, History of the Rise and Influence of the Spirit of Rationalism in Europe Rev. ed. (New York: D. Appleton, 1872). back
2. The point of reference for Americans was to the 'Third Degree' treatment of prisoners by American police and for the British, the paramilitary tactics of the Royal Irish Constabulary. An excellent account of European police practice can be found in Edward Peters, Torture (New York: Basil Blackwell, 1985), pp. 103-114. For the use of these tactics in torture in India and the Philippines, see Edmund Cox, Police and Crime in India (London: Hazell, Watson, and Viney, n.d.), pp. 170, 180-183; William Thaddeus Sexton, Soldiers in the Sun: An Adventure in Imperialism (Freeport, NY: Libraries Press, 1939), pp. 238-242; Daniel B. Schirmer, Republic or Empire: American Resistance to the Philippine War (Cambridge, MA; Schenkman, 1972), pp. 225-240; and Richard E. Welch Jr., Response to Imperialism: The United States and the Philippine-American War 1899-1902 (Chapel Hill, NC: University of North Carolina Press, 1979), pp. 133-149. back
3. Hannah Arendt, The Origins of Totalitarianism, New ed. (New York: Harcourt, Brace & World, 1966), pp. 443-446. back
4. Henri Alleg, The Question, with an introduction by Jean Paul Sartre, trans. John Calder (New York: George Braziller, 1958); George Orwell, Nineteen Eighty Four (Harmondsworth, England: Penguin Books, 1954); and Arthur Koestler, Arrival and Departure (London: Hutchinson, 1966), pp. 104-113. back
5. Max Weber, From Max Weber: Essays in Sociology, trans and edited by H.H. Gerth and C. Wright Mills (New York: Oxford University Press, 1958), p. 226. back
6. Hannah Arendt, On Violence (New York: Harcourt, Brace & World, p. 1969), p. 46. back
7. Hannah Arendt, Eichmann in Jerusalem: A Report on the Banality of Evil (Harmondsworth, England: Penguin Books, 1963, revised and enlarged edition, 1965), pp. 135-150. back
8. For some recent examples, see A.J. Polan, Lenin and the End of Politics (Berkeley, CA: University of California Press, 1984); Edward Peters, Torture (New York: Basil Blackwell, 1985); Michael Ignatieff, "Torture's Dead Simplicity," New Statesman (20 September 1985): 24-26; and Richard Rubinstein, "The Bureaucratization of Torture," Journal of Social Philosophy 13 (1982): 31-51. back
9. See Frantz Fanon, The Wretched of the Earth (New York: Grove Press, 1968), pp. 283-288. Michel Foucault has characterized this exercise of power as "pastoral power" as opposed to the more familiar way in which persons are trained through "disciplinary power." See Michel Foucault, "Omnes et Singulatim: Towards a Criticism of Political Reason," in The Tanner Lectures on Human Values ed. Sterling McMurrin (Salt Lake City: University of Utah Press, 1981). back
10. I adopt and adapt these distinctions from Jurgen Habermas' work, particularly The Theory of Communicative Action, Vol. I: Reason and the Rationalization of Society, Volume II: Lifeworld and System: A Critique of Functionalist Reason, trans. T. McCarthy (Boston: Beacon Press, 1981, 1987). back
11. See John H. Langbein, Torture and the Law of Proof: Europe and England in the Ancien Regime (Chicago: University of Chicago Press, 1976). See also Gavan McCormack, "Crime, Confession, and Control in Contemporary Japan" and Igarashi Futaba, "Forced to Confess," both in Democracy in Contemporary Japan, ed. Gavan McCormack and Yoshio Sugimoto (Armonk, NY: M.E. Sharpe, 1986). back
12. Amnesty International Briefing, Guatemala (December 1976), pp. 5-6; 11-16. back
13. See Franklin Mark Osanka, ed. Modern Guerilla Warfare: Fighting Communist Guerilla Movements, 1941-1961 (New York, Free Press, 1962), pp. 175-212. back
14. See Futaba as well as Hylah M. Jacques, "Spain: Systematic Torture in a Democratic State" Monthly Review 37 (November 1985):57-62. back
15. For the full argument, see James David Barber, "Rationalizing Torture: The Dance of the Intellectual Apologists," The Washington Monthly 17:17-18. back
16. See Marjorie Agosin, "Notes on the Poetics of the Acevedo Movement Against Torture," Human Rights Quarterly 10 (1988)339-443 and Amnesty International, Torture in the Eighties, pp. 71-72. back
17. In 1989, the Uruguayans absolved military officers responsible for human rights violations by a 60% majority. This outcome reflects not a sense of forgiveness, but the fact that more democracy has not put Uruguayans in a better position to confront torturers. back
18. Max Weber, The Protestant Ethic and the Spirit of Capitalism trans. Talcott Parsons (New York: Charles Scribner's Sons, 1958), pp. 97, 128, 197. See also p. 167. back
19. Seymour Martin Lipset, "Some Social Requisites of Democracy: Economic Development and Political Legitimacy," American Political Science Review 53 (March 1959): 91. back
20. Mancur Olson, "Rapid Growth as a Destablizing Force," Journal of Economic History 23 (1983):529-52; Samuel P. Huntington, Political Order in Changing Societies (New Haven, CT: Yale University Press, 1968); and Lucian W. Pye, Aspects of Political Development (Boston: Little, Brown, 1966). back
21. See Pye, p. 127 and Samuel Huntington, "Civil Violence and the Process of Development," Adelphi Papers 83 (December 1971):6. back
22. Huntington, p. 23. back
23. Huntington, p. 24. back
24. Lucian Pye, "The Roots of Insurgency and the Commencement of Rebellions," in Internal War, ed. H. Eckstein (New York: Free Press, 1964), pp. 157-179, and Samuel Huntington, "Guerilla Warfare in Theory and Policy," in Osanka, pp. xv-xxii. See also Barber, pp. 12-13. back
25. Huntington, "Civil Violence and the Process of Development," p. 15. back
26. See Miles Wolpin, Militarization, Internal Repression and Social Welfare in the Third World (London: Croon Helm, 1986). back
27. See Alfred Stepan, "The New Professionalism of Internal Warfare and Military Role Expansion," in Armies and Politics in Latin America (New York: Holmes and Meier, 1976), pp. 244-260. back
28. David Pion-Berlin, "Political Repression and Economic Doctrines: The Case of Argentina," Comparative Political Studies 16:1 (April 1983):37-66; Barry Rubin, Paved with Good Intentions: The American Experience in Iran (Harmondsworth, England: Penguin Books, 1981), pp. 177-178; and Fred Halliday, Iran: Dictatorship and Development (Harmondsworth, England: Penguin Books, 1979), p. 50. back
29. In the case of South Africa, this point is made cogently in D. Russell, Rebellion, Revolution and Armed Force (New York: Academic, 1974). For a more general criticism along the same lines, see Charles Tilly, "Does Modernization Breed Revolution?" Comparative Politics 5:3 (April 1973):425-48. back
30. Pye, "The Roots of Insurgency and the Commencement of Rebellion," pp. 159, 170, 178-9. See also Barber, pp. 13-14. Huntington distinguishes between Western Revolutions (in which the state has lost popular consensus) and Eastern Revolutions (in which the state possesses some measure of popular consensus). But Huntington's examples of the latter sort of state (e.g. Chiang Kai-Shek's China or South Vietnam) could be understood better as highly repressive military regimes led by determined minorities. back
31. Noam Chomsky and Edward S. Herman, The Political Economy of Human Rights, Volume I: The Washington Connection and Third World Fascism (Montreal: Black Rose Books, 1979); and Edward S. Herman, The Real Terror Network (Boston: South End Press, 1982). See also Michael Stohl and George A. Lopez, eds, The State as Terrorist (Westport, CT: Greenwood Press, 1984) and Amnesty International, Report on Torture (London: Gerald Duckworth, 1973), pp. 18, 27. . back
32. I adapt this phrase from Sheldon Wolin's discussion of Machiavelli and the role of violence in politics. See Sheldon Wolin, Politics and Vision: Continuity and Innovation in Western Political Thought (Boston: Little, brown, 1960), pp. 220-224. . back
33. Edward S. Herman, The Real Terror Network, p. 113. . back
34. Ibid., pp. 112-113. . back
35. David Hawk, "Tuol Sleng Extermination Centre," Index on Censorship 15:1 (January 1986):25-31. . back
36. Amnesty International, Torture in the Eighties (London: Amnesty International, 1984), p. 66. . back
37. See Wolpin as well as Lars Schoultz, Human Rights and United State Policy towards Latin America (Princeton, NJ: Princeton University Press, 1981); and Michael Klare and Cynthia Arnson, Supplying Repression, with Delia Miller and Daniel Volman (Washington, D.C.: Institute for Policy Studies, 1977); Carol Ackroyd et al, The Technology of Political Control (Wolfboro, NH: Longwood, 1980); Suzanne Franks and Ivor Gaber, "The UK's Torture Trade Has Quietly Resumed," New Statesman 108:4 (September 21, 1984):4; and Amnesty International, Torture in the Eighties. . back
38. Hawk, p. 27. . back
39. Ibid. . back
40. Amnesty Internationa, Torture in the Eighties, p. 68. back
41. Karl Marx, Capital: A Critique of Political Economy (Penguin Books: Harmondsworth, England, 1976), I:344-352. back
42. See Georg Rusche and Otto Kirchheimer, Punishment and Social Structure (New York: Columbia University Press, 1939). back
43. Halliday, pp. 193-197; 202-206. See also William H. Bartsch, "The Industrial Labor Force of Iran: Problems of Recruitment, Training, and Productivity," in The Population of Iran: A Selection of Readings, ed. Jamshid Momeni (Honolulu: The East-West Center, 1977), pp. 322-326. back
44. See David Pion Berlin, "Political Repression and Economic Doctrines: The Case of Argentina." back
45. See Thomas R. Forstenzer, French Provincial Police and the Fall of the Second Republic: Social Fear and Counterrevolution (Princeton, NJ: Princeton University Press, 1981); Jean Gottman, "Bugeaud, Gallieni, Lyautey: The Development of French Colonial Warfare," in Makers of Modern Strategy: Military Thought from Machiavelli to Hitler, edited by Edward Mead Earle, with the collaboration of Gordon Craig and Felix Gilbert (New York: Atheneum, 1966), pp. 234-256; Anthony Thrall Sullivan, Thomas-Robert Bugeaud, France, and Algeria, 1784-1849: Politics, Power and the Good Society (Hamden, CT: Archon Books, 1983); and Melvin Richter, "Tocqueville in Algeria," The Review of Politics 25:3 (July 1963):362-398. back
46. See Eric Stover and Elena O. Nightingale, The Breaking of Bodies and Minds: Torture, Psychiatric Abuse, and the Health Professions (New York: W.H. Freeman, 1985) and Richard H. Goldstein and Patrick Breslin, "Technicians of Torture: How Physicians Became Agents of State Terror," The Sciences (March/April 1986):14-19.. back
47. See Peters, pp. 174-176 and Kevin Krajick, "Healing Broken Minds," Psychology Today (November 1986): 66-69. back
48. Theorists justifying torture draw on a variety of penal philosophies. Here I am concerned especially with the utilitarian justifications of torture. See Michael Levin, "The Case for Torture," Newsweek (7 June 1982): 13 and the more cautious and thorough arguments in Gary E. Jones, "On the Permissibility of Torture," Journal of Medical Ethics 6 (March 1980):11-13. For retributivist justifications, see Edward G. Rozycki, "Pain and Anguish: The Need for Corporal Punishments," Proceedings of the Philosophical Education Society of Australasia 34 (1978): 380-392; Graeme Newman, Just and Painful: A Case for the Corporal Punishment of the Criminal (London: MacMillan, 1983); Ernest Van Den Haag, "Refuting Reiman and Nathanson," Philosophy and Public Affairs 14:2 (Spring 1985):171. One of the earliest defenses of torture in the utilitarian vein was provided in Roger Trinquier, Modern Warfare: A French View of Counter-Insurgency (New York: Praeger, 1964). back
49. Rubinstein, pp. 37-38. back
50. Thomas R. Forstenzer, French Provincial Police and the Fall of the Second Republic: Social Fear and Counterrevolution. back
51. Michel Foucault, Discipline and Punish: The Birth of the Prison (New York: Random House, 1977). back
52. Peters, pp. 40-73. back
53. Darius Rejali, "Discipline and Torture, or How Iranians Became Moderns," Ph.D. Dissertation, McGill University, 1987. back
54. See Barber, 12-13; Agosin, p. 339; Peters, pp. 6-7; Michael Taussig, "Terror as Usual," paper delivered at the Conference "Talking Terrorism: Ideologies and Paradigms in a Postmodern World," Stanford University, Humanities Center, February 4-6, 1988 and idem, "Culture of Terror--Space of Death. Roger Casement's Putumayo Report and the Explanation of Torture," Contemporary Studies in Society and History 26 (1984):466-497. back

Copyright (c) 2004 by Darius Rejali, all rights reserved. This text may be viewed, downloaded, redistributed or republished in any medium for the express and non-commercial purposes of criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, provided such use is in accord with Section 107 of the United States Copyright Act. Archiving, redistribution, or republication of this text on other terms, in any medium, requires the consent of the publisher and/or author at