5 PTSD sufferers showing positive interference for Vietnam-related words (BODYBAG), but not for other intrusive words (GERMS) (McNally et al.1990). Thus, some of the key features of PTSD, sleep disturbance, intrusive cognitions, psychological reactivity and physiological distress, seem to be supported empirically, and there is support for the notion of a specific disorder produced by trauma, and capable of being delineated from other disorders. It seems clear that exposure to violence has severe, persistent and delayed sequelae, with a dose-response effect, but there still remain some difficulties, and some critics. The major critics come from amongst those working with torture survivors, who are critical of many aspects of the PTSD definition, and suggest that there may still be such a thing as a "torture syndrome" apart from PTSD. TORTURE: Torture clearly represents an extreme form of exposure to violence, in that the effects are premeditated and designed, the process usually involves attacks of both a physical and psychological nature, and, most importantly, torture has an explicitly political purpose in a clear socio-political context. One estimate sees "government-sanctioned torture" as being present in 78 countries in the world (Jacobsen & Vesti.1992), whilst another estimate reckons that between 5% and 35% of the worlds refugees have suffered at least one torture experience (Baker.1993). So it is well to have an understanding of the scale of the problem, and to see that it has a particular socio-political value. In general, those who work with torture survivors argue that PTSD is an insufficient definition of the consequences of torture. This argument requires some brief consideration. At the outset, it is worth noting several important features of modern torture, for it is clear that torture, as a socio-cultural phenomenon, may well have had different effects through history. Contrasting ancient and modern torture, Rasmussen notes that torture was an accepted practice in previous times, that it was practised publicly, and that it was usually carried out after legal proceedings, whereas today torture is clearly not acceptable, is invariably carried out in secret, and is mostly arbitrary in its infliction (Rasmussen.1990). This last point is not trivial, for it is well-established that torture is specifically used a political weapon in order to achieve political ends: the use of terror and torture as an arbitrarily applied means of political coercion is an increasingly common feature of modern life. Thus, the meaning of torture has altered over time, and it seems pertinent to remember that torture may differ from other trauma because of its meaning alone. It may seem to be hair-splitting to raise the socio-political and meaning in a consideration of psychopathology, but it is obvious that it is just these aspects of torture that set it aside from disasters, catastrophes, wars, accidents and abuse. It is the specific purpose of torture that sets it aside from most other trauma. Torture and repressive violence are specifically targeted at individuals and groups with the specific intention of causing harm, forcing compliance, and destroying political will, frequently in the absence of war, but always in a situation of civil conflict (Somnier & Genefke.1986). Thus, there is considerable debate over whether torture should be conceptualized in a narrow medical framework, or should be seen in some broader framework including the political. The deliberate and systematic attack on people, and the attempt to destroy personality and political will, are felt to be such intrinsic features of torture that a narrow definition, such as PTSD, may miss this. For this reason, many workers in the field prefer the concept of "Psychosocial Trauma" to PTSD, for it seems to specifically allow for links to be made between the causes of trauma and trauma itself (Pagaduan-Lopez.1994). As Basoglu has pointed out, this means that there are problems involved in the classification of torture, and three main arguments may be identified (Basoglu.1993). Firstly, torture is a political phenomenon, and thus is not easily captured within a psychiatric diagnosis: this refers specifically to some criterion of meaning. Secondly, PTSD does not apply to torture since it does not reflect the understanding that torture is only one of a series of ongoing trauma affecting a survivor, and, thirdly, psychiatric labels are stigmatizing and should be avoided. The rationale behind each of these views can be given quite shortly. The first point relates to the validity of psychiatric diagnosis, and, in particular, the validity of PTSD. As was pointed out above, it is not in question that the diagnosis of PTSD can be made reliably. Studies of the prevalence of PTSD in torture survivors clearly demonstrate high rates of PTSD in torture survivors. For example, In a study of Turkish prisoners, it was shown that 85% of the sample had been tortured (Paker et al.1993). Of the tortured group, 39% showed PTSD, whilst none of the non-tortured group had the disorder, and, of those who showed physical sequelae of torture, 71% had PTSD. A study from Gaza, showed that more than 70% of political prisoners had received more than one form of torture, with 30% showing PTSD (El-Sarraj & Salim.1993). So, it is not in dispute that PTSD can be found in torture survivors, nor that torture is not a cause of PTSD, but it can be argued that the meaning of torture is not well reflected in the current classification of PTSD. The narrow classification of torture as PTSD also does not reflect the reality for torture survivors, and, in particular, the finding that torture is merely one of a series of stressful life events for the survivors. Survivors do not merely suffer psychic and physical injury, but they also lose families, jobs, educational opportunities, and suffer alienation, displacement from their communities, and frequently end up as refugees (Baker.1993). In fact, torture survivors suffer a wide range of adverse consequences, and this frequently means that