It is impossible to find conclusive statistical evidence on the question of whether or not Irish society is becoming more violent. This is because of a lack of research and data on the topic and because of the inherent unreliability of some of the apparently key official statistics, such as those on assaults.

As Victim Support has argued this week, many assaults are not even reported to the police because victims, especially young people, fear reprisal or believe little will be achieved by reporting.

We also learnt this week that 1,400 victims of assault have attended the emergency departments of Dublin hospitals already this year. This is an important statistical indicator of the level of violence on our streets, but it is more than twice the total number of indictable assaults recorded by the Garda for the whole State in 1998; a fact that must raise serious doubts about the usefulness of the official crime figures.

While the discrepancy between hospital figures and indictable crime statistics suggests a considerable level of under-reporting of violent attacks, it is also partly explained by a growing tendency for assaults to be classified as non-indictable by the Garda.

In 1998 the Garda recorded almost 9,000 assaults as non-indictable but only 700 as indictable, whereas in 1990, when there were 1,400 fewer assaults in total, the force classified nearly 1,100 assaults as indictable and only 7,200 as non-indictable.

Analysis of crime trends focuses almost exclusively on the supposedly more serious indictable assaults, so categorising a much greater proportion of assaults as non-indictable can put a positive, but seriously misleading, spin on annual crime figures.

The problem of the "dark figure" of unreported crime and an unknown amount of reported but unrecorded crime, along with the uncertainty of Garda classification methods, justify a high degree of scepticism about the official, optimistic picture of declining violent crime.

Moreover, the statistics for homicide, arson and malicious damage against schools, offences which are almost invariably reported and recorded, have shown a sustained upward trend over the last decade.

Most seriously, the number of homicides has almost doubled from average figures of about eight homicides per million in the 1970s and 1980s to about 14 per million over the last five years. For example, in 1998 there were 51 homicides, compared with only 26 in 1990. These figures are compelling evidence of a deteriorating situation in Ireland.

But there is also powerful experiential and anecdotal evidence about the increasing violence of Irish society. Many people or their family and friends have been personally exposed to street violence.

The media in recent years have also been to the forefront in exposing the hidden violence of Irish society, such as child sex abuse, and in this way as well have increased public awareness of violence.

One of the most credible and insightful witnesses to the Irish crime scene has been Mr Eamon Barnes, until recently director of public prosecutions. Mr Barnes has concluded from his privileged, professional access during 25 years to the books of evidence on Irish crime that the level of violence in crime has changed beyond recognition.

In this period, he has noted an increased frequency, gratuitousness and savagery in the use of violence. The deliberate targeting of highly vulnerable victims, such as women, the elderly and the disabled, and the occasional use of torture also give testimony to a radically worsening climate of violence.

The sort of appalling assaults that we used to read about happening in Britain or the US, including random, unprovoked and apparently unmotivated acts of violence, now happen here, but they are still much less frequent here. But there can be little doubt that increasing street violence reflects the immense social and cultural changes in Ireland.

There are many factors that may actively promote violence or loosen the restraints on it. These include the rampant Tiger culture of laddish competitiveness and crass materialism, and the continuing high levels of inequality in income and opportunity.

Other factors include the cumulative toll of the many stressors in modern life, such as traffic chaos, the relatively common patterns of neglectful and disruptive child-rearing, the desensitising effects of TV and video-game violence, the widespread abuse of alcohol by the young, the role anxiety of the modern male and the loss of the sense of community solidarity.

Even if it is obvious that street violence is on the increase, the overall level of violence in society remains shrouded. The violence within a society can be manifest in many different ways, some of which remain hidden from public view.

Domestic violence, sexual violence, child abuse, intimidation on housing estates and at football games and bullying in the workplace and schools, even national suicide rates, are all important indicators of the level of violence of a society.

While figures for street crime and assault inform us about security in public places, they do not tell the whole story of societal violence.

We are now aware that, in the 1950s, when the level of reported violent crime in Ireland was among the lowest in the world and the streets were very safe, there was widespread and brutal institutional and personal abuse of women and children.

It is a paradox in the study of violence that effective repression of aggressive feelings in the public world can be linked to vicious expression of aggression in the private domain.

Modern societies strive, with considerable success, to contain violence and regulate its use. Physical punishment has been banned from schools and in some countries from the home and aggressive conduct has become a greater taboo than sex.

Safe areas and comfortable lifestyles are increasingly protected by the use of gated communities, CCTV systems and exclusive, privately policed districts for shopping and socialising. These measures keep the external enemy at bay but are useless against the more common perpetrator of violence, who lurks within the intimate circle.

By and large, the sanitising and securitising process has succeeded only in redistributing visible street violence rather than abolishing it. The result is that while the lives of the majority may be more free of violence, it has become more concentrated, more intense and more visible at the anarchic margins and among those with nonconformist lifestyles - the substance abusers, the criminal, the homeless and the teenage gang members.

Dr Paul O'Mahony lectures in psychology in the School of Occupational Therapy, TCD, and is the author of Criminal Chaos