In the last decades, policies against violence against women (referred to hereafter as AVAW policies) have been similar in all industrial countries. Such measures have mainly been of two types: legal reforms, in order to declare violent actions against women unlawful acts which are punishable; and social services for victims of violence, for instance, refuges for battered women.

In Spain, as in other European Union (EU) member states, AVAW policies have reached only partial achievements, mainly because of two problems. First of all, most policies are directed for victims, that is, women, but not to male perpetrators of violence, who are the cause of the problem. Secondly, there is an implementation deficit in this policy area in all countries. Measures are formulated but weakly implemented. This paper argues that this implementation deficit can be partially explained cross-nationally by the structural characteristics of this policy area. The implementation of most AVAW policies is the responsibility of a high number of political and social actors: judges, prosecutors, police, staff in hospitals, forensic surgeons, and social workers, to name just a few. All these actors deal directly with "clients" (whether the alleged perpetrators of violence or, more often, their victims). These bureaucrats can easily jeopardise the implementation of most AVAW policies due to the fact that they have a high degree of discretion and autonomy to perform their jobs, and are not subjected to strict supervision by the people who occupy higher positions in the organisations for which they work. Therefore, this paper concludes that in order to elaborate more effective AVAW policies it is necessary to direct more efforts to the root of the problem (violent men). It is also mandatory to increase the degree of collaboration of social and political actors involved in this policy area.

The first section of this paper contains general remarks on violence against women and on the characteristics of this policy area in industrial societies. The second and third sections present a national case: Spain. The second section describes the main AVAW policies in Spain since 1975. The third section focuses on the role played by different social and political actors (chiefly feminists, state feminists,[2] and state officials)[3] in the policy-making process in Spain over the last two and a half decades.[4]

General remarks on violence against women and the characteristics of the policy area

A broad definition of the phenomenon of violence against women "includes any act of verbal or physical force, coercion or life-threatening deprivation, directed at an individual woman or girl that causes physical or psychological harm, humiliation or arbitrary deprivation of liberty and that perpetuates female subordination" (Heise et al., 1994:1165). Nevertheless, for reasons of economy of space, time and research resources, this paper focuses on the study of policies directed at the following violent behaviour against adult women: rape and any other form of sexual attack, and domestic violence, that is, violence perpetrated in the family sphere. Other types of violent behaviour, such as forced prostitution, sexual harassment at work, genital mutilation and abuse of female children are not considered in this paper.

Jane Caputi (1992:204-205), among other authors, has conceptualised violent acts against women as a necessary means to maintain patriarchy (the dominance of women by men which exist in all societies). From this premise, it can be deduced that violence against women will be a permanent characteristic in our societies, which would have to be transformed radically in order to eradicate this sort of violence. It can also be deduced that currently, AVAW policies are mainly inefficient because they do not substantially undermine the source of the phenomenon of violence against women: male domination. Amy R. Elman and Maud L. Eduards (1991:420), while examining the assistance that battered women receive in Sweden, conclude that such assistance is basically insufficient, although not negligible, chiefly because in the policy-making process in the area of gender equality, "progress in women's conditions is defined exclusively in terms of gainful employment and shared parenthood rather than in terms of sexuality." This definition of the improvement of women's status systematically deflects policy-makers' attention away from the phenomenon of violence against women.

In contrast with the propositions of Caputi, Elman and Eduards, this paper claims that in most Western countries, AVAW policies have been in part inefficient not because of the presumed necessity of violence as a means to maintain patriarchal domination, or because of the definition of the betterment of women's lives formulated by policy-makers, but partly due to two reasons: (i) the intense attention paid to victims in comparison with the attention paid to perpetrators (who are the cause of the problem); and (ii) the peculiar characteristics of this policy area, concretely, the inevitable high number of social and political actors involved.

(i) As illustrated below in the case of Spain, most AVAW policies are in practice directed towards battered or sexually assaulted women. In contrast, most AVAW measures handle violent males only at the very end of the process of implementation of policies (if at all). Let me use the case of battered women to substantiate this point. Policies against domestic violence are elaborated in a way that assumes that a battered woman has to be the active part in the solution of "her" problem. In all EU member states she is the person who has to fill a complaint, let a doctor examine her to certify injuries, go to court, and leave her home in order to protect herself from the violent batterer. The state has established a full battery of AVAW measures around her. The state has put in motion campaigns directed at her and other victims to encourage them to stop putting up with domestic attacks and dare to fill a complaint; has established procedures to handle complaints; has trained doctors to certify injuries; has arranged legal processes for crimes or misdemeanours to be judged; and has set up refuges. Only at the very end of a legal process (if such a legal process ever takes place) the state deals with the violent partner. This happens when his actions are judged in court, although in most cases male batterers are absolved.

I am not arguing that policies directed to victims are unnecessary and should be abolished. On the contrary, women are still unprotected by the state (as I will show for the Spanish case). The point I am making now is that in order to combat violence a policy with at least two dimensions is needed: First, programmes to protect victims; second, measures to eradicate the violent behaviour of male perpetrators. In the absence of the second policy, the first is clearly insufficient.

The equivalent of fighting against violence against women with measures directed at women would be to combat anti-semitism with policies to protect Jews (rather than preventing Anti-Semites from attacking Jews), or to eradicate robbery with measures protecting the whole population against thieves (rather than making the lives of thieves really hard).

(ii) Regarding the peculiarities of the policy area of AVAW measures, the characteristic that often hinders the effectiveness of state policies is the inevitably high number of actors involved in all stages of the policy making process (but especially in the implementation stage). If many actors are involved, concerted action among them is required in order to formulate a coherent set of measures. But such concerted action is difficult to establish since each political actor has its own priorities which might not coincide with the priorities of the other state units.

This proposition can be clarified with the example of the elaboration of several measures which deal with rape victims in Spain. As shown below, legal reforms have been passed in Spain by parliamentarians since 1975 in order to expand the definition of rape to include not only vaginal, but also anal and oral intercourse, and also to permit rape victims who become pregnant to have an abortion. Nevertheless, policy formulation does not finish with the works of parliamentarians since rape victims need to be examined and treated by health professionals as soon as possible. Therefore, the Ministry of Health should give instructions to personnel employed in health centres to perform such tasks correctly. The Ministry of Health also has to organise the provision of abortions in health centres. In addition, some social services such as therapy or psychological support have to be set up to help rape victims overcome such experiences and rebuild their lives. In Spain, most of the organisations which provide services to victims are non-governmental, non-profit associations, which receive some financial help from the state. Therefore, some state units have to formulate a policy of subsidising social movements. Moreover, generally speaking, the different state units (for instance, Ministries), are not used to working closely with each other but loosely, or worse, independently.

However, if difficulties exist with regard to the definition of the problem and policy formulation, the chief obstacles for establishing most AVAW policies appear at the implementation level. This feature has been pointed out by other scholars. For example, Jill Radford (1992:255), while examining the treatment of domestic killing by the legal system in England, claims that "discretion dominates every stage of the process - from the decision to act on reports in the first place to the appropriate punishment for the crime at the final stages." The Spanish case confirms Radford's proposition, since the main problem at the implementation stage (which has frequently been an insurmountable obstacle) has been the enormous number of "street-level bureaucrats" who hold disparate views on the same problem. Street-level bureaucrats are state officials who interact directly with citizens and "have considerable discretion in determining the nature, amount, and quality of benefits and sanctions provided by their agencies" (Lipsky, 1980:3). As Michael Lipsky has clearly explained, several reasons account for this high degree of discretion, for example, that these bureaucrats are professionals, in the sense that are expected to use their own knowledge and savoir-faire to solve problems, which are quite complex and not easily remedied by the implementation of routinized procedures. In addition, the population expect from such bureaucrats not only impartiality but also flexibility and responsiveness to unique circumstances. Furthermore, the number of rules which, in theory, street-level bureaucrats have to follow is almost infinite and these change constantly. No wonder, then, that such state officials adhere only to the most basic rules and exercise discretion in the remaining aspects of their work, without being de facto closely supervised by anybody above them. Moreover, street-level bureaucrats' priorities can be different from the goals of the high-rank managers of the institutions and agencies for which they are working. As a result, although the basic guidelines of any policy are set up by legislators, or by policy-makers and high-level bureaucrats, street-level bureaucrats can actually modify and distort such basic guidelines in accordance with their own priorities and views. Finally, civil service provisions, originally established to guarantee non-arbitrary hiring and promotion, actually confer a high degree of autonomy on civil servants, since it is very costly for their superiors to fire them or to reduce their rank (Lipsky, 1980:13-24).

The numerous street-level bureaucrats involved in most AVAW policies can easily obstruct the implementation process, since they enjoy a high degree of discretion. In Spain, the main AVAW policies have been legal reforms, precisely the measures in whose implementation the highest number of actors is involved. An example might be useful at this point. In Spain, repeated domestic violence against women committed by their husbands or cohabiting partners is defined by the penal code since 1989 as a crime, and is punishable accordingly. For this law to be applied, the police and civil guards have to know it, and be willing to receive and adequately handle the complaints. Moreover, personnel who work in health centres and/or forensic surgeons have to be prepared to examine the victims rapidly. If both health centre personnel and/or forensic surgeons perform their tasks with delay, injuries might be partially or totally healed, and the victims of violence might be deprived of some of the most important incriminating evidence. In addition, it could be the case that victims of domestic violence have to leave the family home, where violent acts are repeatedly committed against them, and can therefore be repeated in the near future. If victims do not have enough economic resources to afford other accommodation, and have no relatives who can lodge them, they might have to go to a refuge, if vacancies exist. In some cases refuges are full, and their personnel (or social workers who work for state agencies) have to decide rapidly which potential clients are to be given preference, and which others have to be dismissed (or sent to other social services which provide accommodation to their clients). As for the legal system, judges and prosecutors first investigate the cases, and then convene a trial where the alleged perpetrator of violence is judged. Judges and prosecutors have to know the special characteristics of the offence of repeated domestic violence, among other things, that it normally takes place in the domicile, without witnesses except the victims (or their children), and that proof (injuries in most cases) might already have disappeared. Due to these and other characteristics of domestic assaults, judges and prosecutors have to be guided by the purpose of trying harder than in other (easier) cases to elucidate cases of domestic violence in order to punish the offenders. The co-ordination of the performance of all these and any others involved is required in order to effectively apply the legal reforms in its explicit goal of protecting victims of domestic violence. Such co-ordination is difficult to attain for the reasons above mentioned.

In sum, AVAW policies are inefficient for two reasons. On the one hand, AVAW policies are directed to victims of violence rather than to the cause of the problem (violent males). On the other hand, a high number of political and social actors are involved in this policy area. These actors tend not to be co-ordinated, because they are street-level bureaucrats. These two hypotheses are fully developed in the next two sections of the paper, which are on the Spanish case.

Policies to combat violence against women in Spain[5]

As noted above, the main AVAW policies are in Spain chiefly of two types: (i) legal reforms, and (ii) support services for female victims of violence. AVAW policies have been formulated and implemented with some delay in Spain in comparison with other Western countries. This delay was due in part to the fact that from the mid-1930s to 1975 Spain was governed by a right-wing authoritarian regime, which was notably anti-feminist.

(i) With regard to legal reforms, they are the most important AVAW policies in Spain. The Spanish legal system is a codified system. In common law systems (for instance, those of the UK and the US) judges build case law, and the importance is placed on precedent. In contrast, in code law systems, judges are supposed to apply the principles of the code and laws in each particular case. The source of law is therefore not the precedent but what is written in the code and other pieces of legislation. This is why it was so important in Spain to reform the law, particularly the penal code. It defines the most reprehensible behaviour in a modern society, such as killing, raping or stealing and assigns them punishments).

In the Penal Code, the different violent acts perpetrated against women are defined as either misdemeanours (faltas) or offences (delitos), and each of them is assigned a punishment (pena), which is lower for misdemeanours than for offences.[6] From 1975 to 1989, sexual attacks against women were still listed as before under the title "offences against purity" (delitos contra la honestidad). Specifically, most sexual attacks against adult women different from rape were still called "indecent abuses" (abusos deshonestos). This terminology was not an anecdote but a very telling aspect of the legal definition of such attacks, because it reflected the fact that when it was formulated, it was considered that perpetrators committed them against the purity, decency or chastity of women, and not against women's freedom to decide whether to engage or not in sexual relations.[7] Besides, rape was defined in a very restricted way, because it referred only to heterosexual vaginal coitus, and not to anal or oral coitus, and because it was established that only men could rape women. Furthermore, in all cases of sexual attacks against women (including rape), if the victim "forgave" the perpetrator, no prosecution could take place.

It is important to note that divorce was established in Spain in 1981 (Act Number 30 of July 7).[8] This meant that if the executor of violent acts against a woman was her husband, she could not, until 1981, obtain a divorce, remaining therefore legally married to (although perhaps separated from) the violent husband.

A relevant reform or the Penal Code took place in 1983 (organic[9] Act Number 8 of June 25), which established that when victims of rape (not of other types of sexual attacks) forgave the perpetrators, they should still be punished according to the law.

It should be noted that until 1985 abortion was a crime in Spain in all circumstances, penalised in most cases with a period of imprisonment which ranged from six months to six years, plus the prohibition on health professionals against performing their professions in private and public centres. Therefore, if a woman had been raped and became pregnant, according to the Penal Code, she had to give birth to the baby. Organic Act Number 9 of July 5 1985, however, allows abortion in three circumstances: when the woman has been raped, when pregnancy seriously endangers the life of the mother, and when the foetus is malformed.

The first most important reform of the Penal Code regarding violence against women took place in 1989 (organic Act Number 3 of June 21), which instituted changes that had already taken place in other countries. Sexual attacks were no longer called "offences against purity" but "offences against sexual freedom" (delitos contra la libertad sexual). By the same token, some sexual attacks different from rape were no longer called indecent abuse but sexual aggression (agresiones sexuales). Besides this, the concept of rape was expanded, to include not only vaginal, but also anal and oral coitus. Nevertheless, penetration with the penis was required in order to legally define an assault as rape. Two consequences followed immediately from this requirement: a sexual assault with penetration of foreign objects was not considered a rape; and men could rape women and men, but women could only rape men (Bustos, 1991:115; Cabo, 1993:261). From 1989 to the next reform (1995), rape, like homicide, was punished in Spain with a period of imprisonment which ranged from twelve to twenty years, and sexual aggression with a period of imprisonment which ranged from six months to twelve years. In both cases the perpetrator had to compensate the victim financially, although this sum of money was only very rarely paid to the victim (Rodríguez et al., 1988:151). Another point should also be remembered: rape and other sexual aggression were offences defined in law independently from the marital or professional status of victims, for instance, irrespectively of whether the perpetrator was the husband of the victim, or whether she worked as a prostitute (Bustos, 1991:115). Finally, the "forgiveness" of the victims of any offence against sexual freedom (and not only in the case of rape, as established in 1983), did not cancel the punishment of such behaviour.

The 1989 reformed Article 425 of the Penal Code classified repeated physical domestic violence against women perpetrated by husbands or cohabiting partners as offence, and not as a misdemeanour, as it had been legally defined in the past. "Repeated" (habitual) here meant violence which has been perpetrated at least three times (Bustos, 1991:65; Cobo, 1993:229). The offence of repeated physical domestic violence was punished with a period of imprisonment which ranged from one to six months.

Finally, since the 1989 reform, state officials (for instance, prison guards) who take advantage of the power and influence over their clients that their jobs conferred, to ask sexual favours of their clients or their relatives, are punished more severely than before (López, 1992:317-323).

The second most important legal reform regarding violence against women took place in 1995 (Act 10 of 23 November), with the establishment of the new penal code (the existing code was a modified version of that instituted in 1848). The definition of rape was again expanded, to include penetration with objects. Rape is now punished with a lower maximum number of years of prison (twelve instead of twenty). The punishment of the offence of repeated physical domestic violence was increased (from a period of imprisonment which ranged from 1 to 6 months to a period which ranges from 6 months to 3 years). In addition, the 1995 penal code also establishes that a legal process regarding sexual aggressions, sexual abuses or sexual harassment could be initiated with an action of the prosecutor (before 1995, the complaint by the victim was required).

On 30 April 1998, the Council of Ministers approved an Action Plan Against Domestic Violence (Instituto de la Mujer, 1998), which was formulated under the direction of the main feminist institution of the central state, the Women's Institute (Instituto de la Mujer). It contains proposed of measures to combat violence against women regarding prevention, education, support services for victims, health, legal reforms, and research. Legal reforms followed this Action Plan. On June 9, 1999 (Organic Act 14) the 1995 Penal Code and the Act of criminal indictment (Ley de enjuiciamiento criminal) were modified regarding domestic violence (Malos tratos). The offence of repeated psychological violence in the domestic place was defined (up till then, the Penal Code only defined physical violence). New punishments for aggressors were established: the prohibition to approach the victim, to communicate with her, or to live close to her, in order to avoid a repetition of violent behaviour. This is one of the rare instances in which the state attempts to prevent the perpetration of violence, rather than to intervene after violent attacks have already taken place.

On June 1998, it was also stipulated that judges should not impose fines on violent males if this economic punishment also hurts economically the victim or her family. It should be borne in mind that the most common marital property regime in Spain is community property (gananciales). Under this regime, each spouse is the owner of half of common properties, that is, of all properties and income obtained by any of the two spouses since they marry. When in this situation a violent husband has a fine imposed on him, he normally pays it with common properties, half of which belong to his wife. Therefore, this fine damages the financial position of his spouse, who might have been herself the victim of violence. Finally, since 1999, in some cases of potential misdemeanours (for instance, threats), prosecutors do not need the complaint filled by the victim in order to initiate a case (before, they needed the complaint).

A paramount policy established in parallel with legal reforms has been the collection of statistics of reported cases of violent attacks on women, statistics which, for instance, in the case of domestic violence, hardly existed in Spain until 1983. Feminists and state feminists have urged the police and civil guard (police who work chiefly in rural areas) to collect data of reported cases of aggression in which victims have been women (Gutiérrez, 1990:129). However, it is necessary to bear in mind that the Spanish statistics on this issue, as is the case with the statistics of many other countries, only deal with reported cases.[10] In Spain, as in many other countries (Kornblit, 1994:1181), under-reporting is a common phenomenon, in such a way that estimations of the real number of cases are only tentative. Nevertheless, the judiciary was urged to collect data about court decisions regarding cases of violence against women (Gutiérrez, 1989:9), as were the personnel who work in social services, such as refuges for battered women (Spanish Senate, 1989:12185-12187). It should be noted that still in the late 1990s all these statistics are usually incomplete and hardly comparable (Defensor del Pueblo, 1998). Statistics mainly reflect women suffering from violence (but not men perpetrating violence). For instance, regarding domestic violence, statistics count the number of complaints made by battered women, the number of years on average that women withstand violent attacks before daring to file a complaint, the number of women who seek refuge in a shelter, and the average number of days that they stay in refuges. Statistics usually contain much less information on violent males. For example, statistics do not tell us basic things such as how many violent males there are, or when they started behaving violently towards women).

(ii) As for services for female victims of violence, these consist mainly of the diffusion of information about women's rights and measures to protect victims. Diffusion of information is important because only when women are aware of their legal rights (including that nobody has the prerogative to treat them violently) can women efficiently defend themselves against assaults. It is also useful that women know which social services and other resources are available for them in case of being victims of violence. In this regard, the main state feminist institution of the central state, the Women's Institute created in 1983, has set up and administered women's rights information centres in some cities, where citizens can obtain information about women's rights, through an enquiry made in person, by phone or by mail.[11] Also, a free-of-charge women's rights information phone line was set up in 1991, with the main purpose of reaching women who do not live in cities. In addition to these general information services, the Women's Institute has organised several information campaigns related to the specific issue of violence against women (Gutiérrez, 1990:125; Threlfall, 1985:63). Two of the most recent information and awareness-raising campaigns were put into practice in Spring and Autumn 1998 respectively. The former contained dramatic pictures of women who had been assaulted. The latter was promoted by the Spanish Confederation of Neighbours (Confederación Española de Vecinos del Estado Español), the Women's Institute, the Ministry of the Interior, and the Ministry of Labour and Social Affairs. The main slogan was "If he beats you, he does not love you. Love yourself! File a complaint against him! (Si te pega no te quiere. Quiérete tú. Denúnciale) (El País 14 October 1998:31).

Generally speaking, support services for female victims of violence have been set up later and are currently less comprehensive in Spain than in other countries, as happens with social services in general. If legal reforms are (on paper) fairly complete in Spain, support services for victims are still clearly insufficient (Defensor del Pueblo, 1998). The state does not always provide directly all these services for victims of violence, but in many cases it subsidises non-governmental non-profit women's organisations which provide them.

The best-known services of this type are battered women's refuges (Instituto de la Mujer, 1986:22; Scanlon, 1990:99). The first refuges were set up in 1984, and in 1997, 129 refuges existed in Spain (Instituto de la Mujer, 1997:117).[12] There is a refuge for every 302,000 inhabitants in Spain. This proportion is still lower than the proportion recommended by a Resolution of the European Parliament in 1997: a shelter for every 100,000 inhabitants. The supply of Spanish refuges is geographically uneven, and only one region (Castilla y León) has the appropriate number of shelters according to its population (Defensor del Pueblo, 1998). Refuges are administered by the central state, local and regional governments, and organisations of civil society. As in other countries, refuges are mainly temporary safe accommodations for female victims of violence and their children. In addition, women receive there other services which range from legal advice to psychological support and vocational training, with the aim of helping them to initiate a new type of life away from perpetrators of violence.

In Spain, very few programs for male perpetrators of violence against women exist. Some treatments for rapists and executors of sexual aggressions (but hardly any for batterers) are available in some prisons. These are pilot experiments, and have not been generalised to all prisons. These programmes are voluntary. The inadequacy of the design and supply of programmes for violent males was made public in 1998 by the mass media with the case of the so-called "Rapist of the urban expansion area" (El violador del Ensanche), after the name of the area in the city of Barcelona where he committed the 140 sexual attacks to which he confessed. According to a legally correct decision, he was released in 1998 after 15 years of prison. The judge thought that he was still dangerous. While in prison, he had systematically refused to undertake any psychological programme for violent males (El País 18 October 1998:17).

In conclusions, in Spain since 1975 the main policies to combat violence against women have consisted of legal reforms and, to a lesser extent, of support services for women. The former defined attacks against women as misdemeanours or crimes and assigned them the corresponding punishment. The latter try to protect victims. Generally speaking, and with some exceptions, the state does not intend to prevent the perpetration of violent acts against women, but to intervene only after violence has taken place. This is why state actions usually centre around the victim, encouraging her to file a complaint and initiate legal proceedings. If the state intended to prevent the perpetration of violence, it would establish more measures directed at men, who after all are the aggressors in most cases, and are the potential aggressors.

For instance, the state would set up extensive programmes in schools to teach minors to solve conflicts through negotiation rather than imposition and violence. The state would put into practice rehabilitation programmes for violent males and would do this in various settings and not only in prisons. The state would also be much more active in the area of mass media communications, developing actions jointly with the media regarding the prevention of violence. There are only pilot research projects of this type of measures directed at men and all of these have a voluntary and exploratory character. For instance, in 1998 the government started to talk about prevention of violence with mass media, especially with television (El País 15 October 1998:28).

It is argued in this paper that the fact that AVAW measures are mainly directed at women impedes the effectiveness of policies to combat violence against women, since men (not women) are the root of the problem. This proposition is further illustrated in the next section of the paper, which is on the role played by social and political actors. It is also shown below that the high number of actors involved in this policy area (a structural and inevitable characteristic of the policy area) jeopardised the effectiveness of AVAW policies.

Social and political agents in the policy area of measures to combat violence against women in Spain since 1975

Generally speaking, in Spain the issue of violence against women had not been a priority for activists in the women's movement up to the late 1970s or early 1980s, when certain feminists "discovered" the problem of violence against women, in some cases accidentally (Threlfall, 1985:62-63). For instance, feminists from the Separated and Divorced Women's Association (Asociación de Mujeres Separadas y Divorciadas) who provided counselling and legal advice to women who wanted to initiate separation and/or divorce proceedings, found that the main purpose of many of their clients was to escape from a situation of high levels of domestic violence. By the same token, some activists who worked in health centres as physicians or psychologists were shocked by the high number of female victims of violence who turn to these centres for help. Such professionals started to suspect that the number of victims (from all social and economic backgrounds) who did not ask for help, due to their lack of awareness of their rights, or simply to their fear of being more violently treated by their aggressors if they dared to report their cases, was very high.

The next step taken by feminists was to "open the eyes" of state officials and of the population in general, in order to develop zero tolerance towards violence against women. Women's advocates also posed demands to policy-makers to intervene in this area, organising services for the relief and help of the victims, and reforming the legal system in order to protect women effectively against aggressors. It should be borne in mind that in those years only very few statistics existed which counted the number of violent assaults on women (no reliable statistics existed in the case of domestic violence). Therefore, it was extremely hard for feminists to argue convincingly before policy-makers that violence against women was a serious social problem (in many cases, a life-or-death matter for victims) which deserved state attention and solutions.

One of the first means used by feminists to call attention to the magnitude of the phenomenon of violence against women and to demand the first state measures was the organisation of sporadic but highly visible mass mobilisations against sexual attacks on women, mobilisations which also dated from the late 1970s and early 1980s. Demonstrations generally took place after violent assaults against female citizens (generally rapes) had taken place in a city or town. These first mass mobilisations were organised not only by feminists or members of other social movements (such as neighbours' associations), but also by political parties and trade unions. Not only the organisers participated in these demonstrations, but also many other people who were not normally very active politically, but who felt concerned about the violence perpetrated on women who lived in their towns or cities.

In 1982, a group of women who provided direct assistance to victims of violence constituted the Commission to Investigate the Ill-treatment of Women (Comisión para la Investigación de los Malos Tratos a las Mujeres). This commission was composed mainly of social workers, psychologists and lawyers and they started to pressurise policy-makers with regard to the formulation of more AVAW measures, and to the implementation of those which already existed. For instance, they requested the Ministry of the Interior, on which the police and the civil guard depended, to remind police officers of their obligation to receive complaints of all cases of violent acts against women and to report them to the judicial institutions. Members of this commission also urged this Ministry to order agents to prepare statistics (Gutiérrez, 1990:124; Threlfall, 1985:62).

Activists in the women's movement rapidly found an important ally in state feminists. Since its establishment, one of the priorities of the Women's Institute has been the issue of violence against women (Gutiérrez, 1990:124). Approximately 10-15 per cent of the budget of the Women's Institute has been devoted to subsidising women's organisations. The emblematic services provided by feminists to female victims of violence --refuges-- have been financed with the subsidies of the Women's Institute (and of other state units). The times of feminist activism based exclusively on goodwill, altruism and absolute lack of financial resources were over; high-cost services such as refuges could be instituted partly because of the existence of subsidies.[13] In addition, the Women's Institute, like the majority of state feminist institutions in the Western world, has neither the competences nor the budget to formulate and implement most gender equality policies, but does have the explicit task of trying to convince other state units to set themselves such policies. The Women's Institute has constantly pressurised policy-makers to develop effective measures to combat violence against women.

Feminist organisations which "specialised" in the issue of violence against women have been created mainly since the mid-1980s, including the Association of Assistance to Raped Women (Asociacion de Asistencia a Mujeres Violadas) or the Anti-Aggression Commission (Comisión Anti-Agresiones), but also others. In the mid-1980s, two actors (state feminists and feminists) were already in motion, defining the different dimensions of the problem of violence against women and pressurising policy-makers to do something (or a number of things) to deal with such issues. They were not the only ones in the years to come. For instance, on November 5, 1986 a unit (ponencia del Senado de investigación de malos tratos a mujeres) of the Senate liaison with the Ombudsman and human rights committee (Comisión de relaciones con el Defensor del Pueblo y de los derechos humanos del Senado) was formed with the aim of collecting information about domestic violence against women in Spain, analysing it and giving advice to the Ministries of the Interior, Justice and Education about potential measures to be taken (Gutiérrez, 1990:124). The unit was headed by a female senator from the social-democratic party (Partido Socialista Obrero Español, PSOE), and its other members were two male PSOE senators, one male senator from the conservative party (Partido Popular), and one male senator from the mixed group (Grupo Mixto). The unit finished its works in Spring 1989. Its analyses and recommendations were very similar to those elaborated by many feminist groups and state feminists (Spanish Senate, 1989).

 The concerted efforts of feminists and state feminists have been at the origin of the formulation of many AVAW policies described in the second section of the paper. Nevertheless, state feminists and members of the women's movement found with dismay that, if it was hard to formulate AVAW measures (for instance, legal reforms), their implementation was by no means automatic. In fact, feminists and state feminists discovered that pressure had to be exercised constantly on state officials for programmes to become more than rhetorical declarations. In 1999, the Women's Institute can show a positive record in having implicated other state units in the formulation of equality policies (AVAW measures, among them), which would probably not have been set up in the absence of the Women's Institute. In contrast, the Institute has hardly intervened in the implementation of such policies, mainly due to the lack of personnel and material resources. As a consequence, activists in the women's movement have been those who have insistently put pressure on street-level bureaucrats for programmes to be adequately implemented (or just put in practice). In the following paragraphs I describe the three main professional groups on which feminists have concentrated their lobbying efforts: the police and civil guard; personnel of health centres; and members of the judicial system. Needless to say, the prominence of these groups with regard to the implementation of AVAW policies is not specific to the Spanish case. On the contrary, this is a common phenomenon in many other Western countries.

With regard to the police and civil guard, an increasing number of agents have performed their duties impeccably while dealing with female victims of violence, but others have not. In fact, feminists have continuously complained about the insufficient protection given by the police and civil guard on some occasions to female victims, especially in the case of domestic violence. Sometimes they have come too late when called, precisely because violent acts being perpetrated were against women (Cova and Arozena, 1985:36). Moreover, they have not given proper protection to women who have repeatedly been victims of violent attacks. Many of these women have finally died or become severely injured. According to the 1998 report on domestic violence by the Ombudsman (Defensor del Pueblo, 1998), 89 out of the 91 women killed by their partners in 1997 had filed complaints of domestic violence against their aggressors. The Spanish state failed to protect them from the violent males who killed them. According to the Ombudsman, the state must protect women from perpetrators of violence, and must be held responsible before the law if it fails to fulfil this task. An increase in the level of protection of victims and potential victims is definitely a pending point in a programme to combat violence perpetrated against women (Asociación de Mujeres Juristas Themis).

Feminists have also denounced the fact that, especially in the case of domestic violence, when women have reported violent attacks against them, the police and civil guards have sometimes tried hard to convince victims not to sign a complaint "for their own good", and have acted as mediators in an effort to reconcile the two partners. In other cases, these agents have not informed female victims that they had the right to be attended by police-women or female civil guards if they have been victims of violent assaults. Still in other cases, the police and civil guards have left complaint forms incomplete (they have not inspected the place where violent acts took place, have not taken pictures to be enclosed with the report, or have not tried to find witnesses) (Asociación de Mujeres Juristas Themis). Finally, when female victims of violence are prostitutes, police and civil guards have not always performed their duties as diligently as in the case of other women.

Activists in the women's movement have thought that one of the reasons for the lack of professionalism and low interest with which some police agents and civil guards can treat cases of violent assaults of women (especially if they are committed in the domestic sphere), lies in the fact that a significant number of them still think that they should not be involved in such cases, because they belong to the sphere of the privacy of the individual, where the state should not intervene, or should intervene but has no adequate means to do it (Asociación de Mujeres Juristas Themis, 1999). Other agents simply do not believe the account given by some women while reporting episodes of domestic violence, and still others think that some women "have asked for it." Without denying the centrality of these sort of convictions, it is necessary to emphasise that police and civil guards can easily translate attitudes of indifference towards the phenomenon of violence against women into inaction or omission, since they are street-level bureaucrats with a high degree of discretion. In fact, Lipsky (1980:13) identifies the police as one of the emblematic examples of such bureaucrats, who daily "decide who to arrest and whose behaviour to overlook."

To overcome attitudes of indifference and weak performance of some police and civil guards, feminists have pressurised those authorities on which the police and civil guard depend to supervise their subordinates in order to make agents more active in matters of violent attacks against women. For instance, in part as a result of such pressures, several circulars have been sent to police stations explaining the laws regarding the punishment of violent acts against women, ordering the police to be diligent in the performance of their duties in this matter and to collect statistics of reported cases (Ministerio del Interior and Instituto de la Mujer, 1991:110). In addition, since 1988 a police station dedicated exclusively to cases of violence against women, staffed only by policewomen, exists in Barcelona. Units specialised in such cases, where some policewomen work, but which are not police stations but departments within them, have also been set up. These units are called Services to Attend Women (Servicios de Atención a la Mujer). In 1998 these services existed in 16 cities, and it was expected that at the end of that year new services would be set up in 9 additional cities. Similar services have also existed in Civil Guard Stations in 15 provinces since 1998, and are called Teams for Women and Minors (Equipos de Mujeres y Menores) (El País 16 March 1998:31). Furthermore, since the mid-1980s, courses and seminars about violence against women have been delivered to the police and civil guard (Gutiérrez, 1990:129; Instituto de la Mujer, 1986:15-16). Nowadays then, after more than a decade of feminist mobilisation, some female victims of violence have been attended in stations by policewomen (or by policemen knowledgeable about the issue), an extremely low number of victims have been truly protected from their aggressors, and some statistics have been collected. Nevertheless, these cases are still not at all the majority. Therefore, there is still a lot to be done with regard to the police and civil guard in this policy area.

In relation to personnel in health services, feminists have persistently denounced these street-level bureaucrats for not always examining the victims of violence as soon as they had to. Moreover, such examinations have not always been as exhaustive as they are supposed to be (Gutiérrez, 1989:26-27). In addition, the privacy and intimacy of the victim has not always been sufficiently protected while being examined. Feminists and state feminists have then pressurised the appropriate authorities to order such professionals to be more diligent in their assistance to the victims. As a result, some professionals are currently more responsive to the needs of female victims of violence, but again, they are not the majority. As has been noted (Heise et al., 1994:1172), people who work in the health system are crucial actors, because "as one of the few institutions that see women throughout their lives, the health sector is particularly well placed to identify and refer victims to available services." The last AVAW measures regarding health professionals attempt to homogenise the performance of these professionals and to increase the co-ordination between them and legal professionals. For instance, in October 1998 all public hospitals of the regions where health is still administered by the central state adopted a protocol for doctors. They must fill in a form certifying injuries produced by domestic violence and send a copy to the Court (El País 14 October 1998:31).

The third area of feminists' concern related to the implementation of AVAW policies has been the functioning of the judiciary. In fact, some have identified it as the biggest obstacle for the successful implementation of measures against perpetrators of violent attacks against women (Asociación de Mujeres Juristas Themis, 1999; Gutiérrez, 1989:42-43; Threlfall, 1985:61). Women's advocates have insistently complained about the slowness and superficiality which has characterised the examination of victims by some forensic surgeons (Asociación Española de Mujeres Separadas y Divorciadas, 1985:23; Gutiérrez, 1989:25-26). In addition, feminists have objected to the number of complaints of violent attacks against women (especially in the case of domestic violence) which are classified by judges of the lower courts as misdemeanours rather than offences, and have therefore been punished accordingly in the subsequent trial. In addition, feminists have insistently complained about the proportion of cases in which violent males are not punished: for instance, 82% of men who had been denounced for domestic violence in the region of Madrid between 1992 and 1996 (Asociación de Mujeres Juristas Themis, 1999).

Feminists have also denounced several practices which regularly occur in trials, practices which hinder the explicit aim of the laws to protect the victims effectively and to punish the perpetrators of violence. First of all, as it has been explained by Allison and Wrightsman (1993:171-194) for rape trials in the USA context, and by Sue Lees (1992) for murder trials in Great Britain, on many occasions a trial of violent acts against women becomes a trial of the victims. Victims frequently have to answer questions related to their style of living or to their past sexual activities, under the suspicion that some women (for instance, those who go out alone at night, or who frequent bars, or who do not have a permanent address or a stable partner, or wear certain types of clothes, or are conceptualised as promiscuous) put themselves in danger of being treated violently, since they might indirectly induce men to behave in such way. Feminists have demanded with vehemence that judges, prosecutors and lawyers do not investigate the private life of victims unless strictly necessary, since what it is judged in a trial of this type is the violent behaviour of the presumed perpetrator, and not the intimate life of the victim (Instituto de la Mujer, 1985:71-72). A decision of the Supreme Court of Justice (Tribunal Supremo), that is, the highest judicial unit in all matters except those related to constitutional guarantees, in 1990, corroborated feminists' arguments, declaring that the sexual life of a victim of rape before rape takes place is irrelevant in the trial (El País, November 5, 1990:29). Nevertheless, even now some judges, prosecutors and lawyers make investigations into the previous sexual life of the victim, investigations which are not at all necessary for the elucidation of the cases. Once again, state actions to combat violence against women focus on women, even when violent actions of male perpetrators are being judged in court.

The feminist movement has also complained that prosecutors are very often not active enough in the investigation of violent acts against women before the trial takes place, and do not subsequently maintain charges against presumed perpetrators of violence (Asociación de Mujeres Juristas Themis, 1999; Baiges, 1985:11; Cova and Arozena, 1985:36). This alleged deficit in maintaining charges by prosecutors is very important, because when judges write court decisions they punish perpetrators of violence with a punishment equal or lower than the punishment demanded by prosecutors. This lack of action by prosecutors has occurred in spite of repeated instructions from the attorney general of the state (Fiscal General del Estado) to be active in the prosecution of violent crimes against women --for instance, in October 1998 (El País 17 October 1998).

Another battlefront of feminists' struggles has been the investigation during the trial of the reactions of female victims of violence, especially in cases of rape. The penal code does not say anything about this matter, but in Spain, as in many other countries, it has been a widespread requirement in trials that rape victims prove that they had very actively resisted their aggressors. This de facto requirement is paradoxical, since victims of other offenses or misdemeanours, for instance, robbery, do not have to prove that they had resisted the thieves. After numerous decisions by the Supreme Court of Justice which made reference to the high degree of resistance of rape victims, the Supreme Court of Justice affirmed in a decision of 1987 that rape victims do not have to prove that they have "heroically" resisted rapists, and that it was enough to show that they have been intimidated or threatened, for instance, with a knife (El País October 8, 1987:29). While the matter seemed theoretically to have been settled, feminists have complained that in many trials victims are still asked to prove that their degree of resistance to the rapist was high. As a consequence of this and the investigation of the degree of resistance, victims have had to answer humiliating and embarrassing questions. For instance, in 1989, in a rape trial, the presumed victim was asked if on the day of the rape she was wearing underpants. According to the president of the court, the question was necessary in order to assess how the alleged rapist had acted (and the victim resisted), bearing in mind that he had a knife in one hand, and with the other he had to take some clothes off the victim, a task which would have been harder or easier depending on her resistance (El País June 27, 1989:24; June 28, 1989:31).

Finally, the unfairness of several punishments given to perpetrators of violent attacks on women committed in the domestic sphere has been denounced, especially house arrest (Baiges, 1985:10). It is in many cases a counterproductive punishment for the obvious reason that in the common situation in which the perpetrator and the victim are husband and wife and live together, it is the latter who finally might suffer the consequences of the punishment, since the perpetrator might behave more violently while under arrest.

Feminists have found several allies in their demands on the judiciary. For example, a minority sector of judges and prosecutors is sensitive to women's demands and has supported them while pressing charges and writing court decisions under criteria that have sometimes gone far in the interpretation of the laws. For instance, in one case the judge considered that a woman had been raped in a case of anal coitus, at a time when, according to the law, rape could only take place in cases of vaginal coitus (El País January 19, 1989:32). Feminists and their allies have pressurised policy-makers in high positions to set up guidelines to make the judicial system an efficient mechanism to punish violent attacks. Several superior organs of the judiciary, like the attorney-general of the state (Fiscal General del Estado), that is, the supreme organ of the prosecutors, or the Ministry of Justice, have sent circulars asking prosecutors to be more active in the defence of female victims of violent acts and in the investigation of cases, and to collect data (Gutiérrez, 1989:10-13). As a result of these and other efforts, some trials now take place without irregularities, but not all of them. It should be stressed that judges and prosecutors are also street-level bureaucrats with a high degree of autonomy. The former have also been identified by Lipsky (1980:13) as another emblematic example of such bureaucrats, who "decide who shall received a suspended sentence and who shall receive maximum punishment." In fact, the attorney-general of the state can dictate mandatory norms about how prosecutors have to perform their duties. By contrast, judges are independent, this independence meaning that no state unit exists with capacity to give mandatory instructions to them, and that they write court decisions only in accordance with the constitution and the Spanish laws, but not to any other type of instruction (Gutiérrez, 1989:9-11).

Conclusion

This paper has described the main policies concerning violence against women established in Spain after 1975. It has also been argued that if many difficulties arise when AVAW policies are formulated, the stage which is full of (sometimes insurmountable) obstacles is the implementation phase. This is due not to the patriarchal necessity of violence as a means to maintain the domination of women by men, nor to the definition that in some countries policy-makers (and feminists) make about the betterment of women's conditions (mainly in terms of access to waged labour outside the home), definition which turns attention away from the phenomenon of violence against women, as several scholars have argued. The difficulties in the implementation stage of AVAW policies arise chiefly because policies are mainly directed at women rather than at men, and because of the high number of street-level bureaucrats involved in the implementation of most AVAW policies, including: judges, prosecutors, lawyers, police, civil guards, forensic surgeons and personnel in health centres, among others. Characteristic of these street-level bureaucrats is direct treatment with "clients", and the high degree of discretion and autonomy that they have while performing their jobs. Therefore, each of these bureaucrats is in a key position to jeopardise the implementation of any AVAW programme. For a policy to be implemented, the concerted action of all these bureaucrats is necessary. Such concerted action is difficult to organise, because, among other reasons, each bureaucrat has priorities and ways of working, which might probably be different from those of other bureaucrats.

Obvious as it might be, since violent acts against women are perpetrated everywhere, the implementation of AVAW policies is a very decentralised process. As a consequence, for successful implementation in any single village, town or city, all these street-level bureaucrats have to be willing and prepared to assist victims of violence. Furthermore, when so many actors are involved, the question of "who does what" remains in some cases permanently unsettled. The interviews conducted for the preparation of this paper show the existence of some disagreements among several actors in this regard, for instance, in some cases, between personnel who work in health centres and forensic surgeons.

If the diagnoses of the factors that explain the limited effectiveness of AVAW policies identified in this paper are correct, the recommended future direction of policies is clear. State action in this policy area will be more efficient only when measures try to prevent violence (and therefore are directed to men) and increase the degree of co-ordination of social and political actors combating violence against women.

Two or three decades ago feminists realised in all countries that women live in a violent world. An increasing number of activists have subsequently thought that effective resistance to the phenomenon of violence against women across countries should involve not only the women's movement but also the state (Heise et al., 1994:1174). Nevertheless, the findings of this paper remind us that it is important for feminists to concentrate their pressure on the state not only with regard to the formulation of new AVAW measures but also to the implementation of those which already exist. Besides, due to the high degree of decentralisation of the implementation of AVAW policies and the high degree of autonomy of street-level bureaucrats, feminists have to put pressure not only on the superiors of these bureaucrats, but also on each single bureaucrat of this type. As a consequence, activists in the women's movement accurately sense every now and then that they have to fight on numerous fronts, and that battles are fought but never completely won.

If the description of the characteristics and dynamics of this policy area made in this paper is accurate, it might be concluded that feminists' provision of direct help to the victims of violence is still going to be useful and irreplaceable in the future years. Feminists' direct assistance to victims is sometimes an easier task than lobbying activities to state units, since such assistance can in some cases be organised by feminists alone, or by feminists and state feminists without resorting to a high number of street-level bureaucrats. This has been in fact the strategy pursued by women's advocates in Spain in the last two decades. It is not only that the supply of direct assistance to victims (and programmes for the population in general) has never been abandoned, but also that such programmes have flourished in past years. It is also encouraging to bear in mind that the issue of violence against women, together with abortion, are still unifying motives for the different branches of the Spanish feminist movement to engage in joint activities, even though the movement, in relation to other issues, has usually been very fragmented.

References

Addis, Elisabetta 1989. "What Women Should Ask of the Law: Italian Feminist Debate on the Legal System and Sexual Violence." Harvard University Centre for European Studies Working Paper Series Number 18.

Allison, Julie A. and Lawrence S. Wrightsman 1993. Rape: the Misunderstood Crime. Newbury Park (California): Sage.

Asociación Española de Mujeres Separadas y Divorciadas 1985. "Pago de emplazamientos a funcionarios de justicia." In Primeras jornadas: aplicación del Derecho y la mujer Ed. by Instituto de la Mujer, 21-23. Madrid: Instituto de la Mujer.

Asociación de Mujeres Juristas Themis 1999. Respuesta penal a la violencia familiar. Madrid: Asociación de Mujeres Juristas Themis.

Baiges, Mayte 1985. "Introducción." In Primeras jornadas: aplicación del Derecho y la mujer Ed. by Instituto de la Mujer, 9-12. Madrid: Instituto de la Mujer.

Bustos, Juan 1991. Manual de Derecho Penal: parte especial. Barcelona: Ariel.

Cabo, Manuel (ed.) 1993. Manual de Derecho Penal (parte especial) I. Madrid: Editoriales de Derecho Reunidas.

Caputi, Jane 1992. "Advertising Femicide: Lethal Violence Against Women in Pornography and Gorenography." In Femicide: the Politics of Woman Killing Ed. by Jill Radford and Diana E.H. Russell, 203-221. New York: Twayne.

Connors, Jane Frances 1989. Violence against Women in the Family. New York: United Nations Office at Vienna. Center for Social Development and Humanitarian Affairs.

Cova, Luz M. and Soledad Arozena, 1985. "Aplicación del Derecho." In Primeras jornadas: aplicación del Derecho y la mujer Ed. by Instituto de la Mujer, 35-37. Madrid: Instituto de la Mujer.

Defensor del Pueblo 1998. Informe sobre la violencia doméstica contra las mujeres (unpublished report).

El País 1978-March 1995.

Elman, R. Amy and Maud L. Eduards 1991. "Unprotected by the Swedish Welfare State: A Survey of Battered Women and the Assistance They Received." Women's Studies International Forum Vol.14, Number 5:413-421.

Gutiérrez, Purificación 1989. "La administración de la justicia ante el problema de los 'malos tratos' en el ámbito doméstico" (unpublished paper).

Gutiérrez, Purificación 1990. "Violencia doméstica: respuesta legal e institucional." In Violencia y sociedad patriarcal Ed. by Virginia Maquieira and Cristina Sánchez, 123-136. Madrid: Pablo Iglesias.

Heise, Lori L., and Alanagh Raikes, and Charlotte H. Watts, and Anthony B. Zwi 1994. "Violence Against Women: A Neglected Public Health Issue in Less Developed Countries." Social Science Medicine Vol.39, Number 9:1165-1179.

Instituto de la Mujer 1985. "Conclusiones de las primeras jornadas de aplicación del Derecho en relación a la mujer." In Primeras jornadas: aplicación del Derecho y la mujer Ed. by Instituto de la Mujer, 69-75. Madrid: Instituto de la Mujer.

Instituto de la Mujer 1986. El Instituto de la Mujer 1983-1986. Madrid: Instituto de la Mujer.

Instituto de la Mujer 1994. La mujer en cifras: una década, 1982-1992. Madrid: Instituto de la Mujer.

Instituto de la Mujer 1997. Las mujeres en cifras 1997. Madrid: Instituto de la Mujer.

Instituto de la Mujer 1998. Plan de Acción contra la violencia doméstica. Madrid: Instituto de la Mujer.

Kornblit, Ana Lia 1994. "Domestic Violence -- An Emerging Health Issue." Social Science Medicine Vol.39, No.9:1181-1188.

Lees, Sue 1992. "Naggers, Whores, and Libbers: Provoking Men to Kill." In Femicide: the Politics of Woman Killing Ed. by Jill Radford and Diana E.H. Russell, 267-288. New York: Twayne.

Lipsky, Michael 1980. Street-Level Bureaucrats. New York: Russell Sage Foundation.

López, Jacobo 1992. Manual de Derecho Penal: parte especial III. Madrid: Akal.

Ministerio del Interior and Instituto de la Mujer 1991. Violencia contra la mujer. Madrid: Ministerio del Interior.

Newton, Michael T.; Donaghy, Peter J. 1997. Institutions of Modern Spain: A Political and Economic Guide. Cambridge: Cambridge University Press.

Radford, Jill 1992. "Womanslaughter: A License to Kill? The Killing of Jane Asher." In Femicide: the Politics of Woman Killing Ed. by Jill Radford and Diana E.H. Russell, 253-266. New York: Twayne.

Rodríguez, Luis; Francisco J. Álvarez; and Pilar Gómez 1988. La justicia ante la libertad sexual de las mujeres. Madrid: Instituto de la Mujer.

Scanlon, Geraldine M. 1990. "El movimiento feminista en España, 1900-1985: logros y dificultades." In Participación política de las mujeres Ed. by Judith Astelarra, 83-100. Madrid: Centro de Investigaciones Sociológicas and Siglo XXI.

Spanish Senate 1989. "Informe de la Comisión de Relaciones con el Defensor del Pueblo y de los Derechos Humanos encargada del estudio de la mujer maltratada." Boletín de las Cortes Generales. Senado May 12, Number 313:12182-12211.

Stetson, Dorothy McBride 1987. Women's Rights in France. Westport (Connecticut): Greenwood Press.

Stout, Karen 1992. "'Intimate Femicide': Effect on Legislation and Social Services." In Femicide: the Politics of Woman Killing Ed. by Jill Radford and Diana E.H. Russell, 133-140. New York: Twayne.

Threlfall, Monica 1985. "The Women's Movement in Spain" New Left Review 151:44-73.

Valiente, Celia 1995. "The Power of Persuasion: The Instituto de la Mujer in Spain." In Comparative State Feminism Ed. by Dorothy McBride Stetson and Amy G. Mazur. Newbury Park (CA): Sage (forthcoming).

Valiente, Celia 1996. "Partial Achievements of Central-State Public Policies Against Violence Against Women in Post-Authoritarian Spain (1975-1995)" In Women in a Violent World: Feminist Analyses and Resistance Across 'Europe' Ed. by Chris Corrin, 166-185. Edinburgh: Edinburgh University Press.

    [1] This paper draws heavily upon an earlier version, which was published in Valiente (1996). Two additions make this paper different from the aforementioned version: the study of men in this policy area, and the analysis of policies elaborated between 1995 and 1999. This paper will be substantially revised for publication after the Council of Europe Seminar on "Men and Violence Against Women", Strasbourg, 7-8 October 1999. I would like to thank Ángel J. Sánchez for his help provided in the collection of secondary sources and Roberto Garvía for his invaluable comments on an earlier version of this paper.

    [2] Since the 1960s, institutions with the concrete purpose of promoting gender equality have been set up, developed (and sometimes even dismantled) in most industrial countries. In social science literature such institutions have been called "state feminist" institutions or bureaucracies. The people who work in them are described as "state feminists".

    [3] A good guide to political institutions in modern Spain is: Newton and Donaghy (1997).

    [4] This paper is largely based on an analysis of secondary literature, legislation, press files, published and unpublished political documents and seventeen in-depth personal interviews with four members of feminist associations, one judge, one police officer, three civil guards (police officers mainly for the rural areas), one social worker, two members of the personnel who work in a battered women's refuge, one female victim of violence (rape), one forensic surgeon, one physician specialising in the examination of female victims of violence, and two lawyers specialist in AVAW legal measures, conducted in Madrid in March 1995. In order to maintain the anonymity of those interviewed, their names do not appear in this paper.

    [5] I will concentrate on the central-state policies considered most relevant, that is, those which affect a large number of women, and/or are financed with a significant amount of public resources, and/or are specially innovative. The description of the programmes made here is by no means exhaustive.

    [6] "Offences" and "crimes" are used in this paper as synonymous.

    [7] This type of terminology also enjoyed certain currency in other Roman law countries. For instance, in Italy, sexual violence was listed in the penal code under title "crimes against public morality and right living" (Addis, 1989:2). In France, sexual assaults were prosecuted according to an article of the penal code which dealt with "assaults on morals" (Stetson, 1987:163).

    [8] Divorce had also existed during the II Republic (1931-1936), but it was abolished by the subsequent dictatorship.

    [9] According to article 81 of the 1978 Constitution, an organic Act (Ley orgánica) regulates, among other matters, fundamental rights and public liberties. An absolute majority of the Low Chamber, in a final vote on the whole project, is necessary for the approval, modification or derogation of an organic Act. For an ordinary -not organic- Act, only a simple majority is required.

    [10] Number of reported rapes in Spain: 1,723 in 1989; 1,789 in 1990; 1,1936 in 1991; and 1,599 in 1992. Number of reported sexual aggressions: 2,502 in 1989; 2,277 in 1990; 2,282 in 1991; and 2,335 in 1992. Finally, the number of reported cases of domestic violence against women was: 13,705 in 1984; 15,681 in 1986; 15,230 in 1987; 13,644 in 1988; 17,738 in 1989; 15,654 in 1990; 15,462 in 1991; 15,184 in 1992; 15.908 in 1993; 16.284 in 1994; 16.062 in 1995; and 16.378 in 1996 (Instituto de la Mujer, 1994:92-93; 1997).

    [11] These information centres were not an original creation of the Women's Institute, because the former Subdirección General de la Mujer, dependent on the Ministry of Culture, had already set up three centres, which the Institute inherited. The number of centres increased steadily. Whereas in 1984 there were only three, since 1987 there are eleven.

    [12] The first battered women's refuge was set up in 1971 in Great Britain (Connors, 1989:34) and in 1974 in the USA (Stout, 1992:134).