In recent decades, the reliability of recorded crime statistics in assessing crime levels and comparing crime in international terms has increasingly been called into doubt. Court and police statistics are too much affected by the national features of law enforcement authorities and of the crime recording practice to fulfil that purpose. By now, it has become symptomatic of this that general crime indicators are interpreted above all as characterising of the activity of the criminal justice system1. More adequate ways for assessing the crime situation have been sought, including searching for types of crimes that are most immune to the above-mentioned shortcomings.
Intentional taking of another person's life is a type of criminal offence that is characterised by relatively little dependence on the will of the particular regulator involved or on national statistical manipulations; by low latency; and by a high rate of detection. That is why intentional homicides rather than general numbers of recorded crimes are often used for comparing the crime in different countries. The attention to the homicide rate has spread from criminology to more general sociological approaches, where it is used as an indicator in the assessment and comparison of the crime situation and social security of different countries2.
This article looks at violent crime and its level in Estonia as compared to other countries. The most significant type of criminal homicides in Estonia-intentional homicide-is used as empirical material3. The objective is to assess the level of intentional homicide in Estonia in comparison to other countries and to point out the most important short- and long-term trends. Based on homicide rates, an attempt is made to determine the general vector of Estonia's social development.
Up to now only a few in-depth analyses of violent crime in Estonia that have been published in the domestic scientific literature4. Local homicides have been analysed by M. Lehti, who has also gathered material on this subject in Estonia. His research on this subject has been published in Finnish 5 and also in English6. Homicides in Estonia as indicators of the local crime situation were mostly beyond the scope set for that study, and the time period covered ended with the mid-1990s.
Homicide statistics in Estonia since the restoration of independence must be viewed in the context of the criminal law in force. Firstly, the period from 1991 to 2002-i.e., the time when the Criminal Code of the Republic of Estonia (hereinafter 'CrC') was in force 7 -must be considered and, secondly, the period that started with the entry into force of the Penal Code 8 (hereinafter 'PenC') on 1 September 2002.
Until 2002, criminal homicides in Estonia were accounted for in two statistics: the recorded criminal offences (crime) statistics compiled by the Police Board and the court statistics of the Ministry of Justice. Criminal homicides were also accounted for in the health statistics of fatalities caused by external forces-i.e., violent deaths-kept by the Statistical Office. Police and court statistics are normally maintained in terms of criminal offences and types of crimes, whereas violent death statistics focus on crime victims.
In the Estonian crime statistics of the first, earlier, time period, the compilation principles of which remained relatively unchanged from those of the Soviet era, intentional homicides were considered separately9. They were taken to include the CrC § 100 (homicide without aggravated circumstances), CrC § 101 (murder), CrC § 102 (infanticide), and § 103 (homicide by one in a provoked state, so-called affect homicide). In presentation of the intentional homicide statistics, one type of intentional homicide was omitted-CrC § 104 (homicide through exceeding the limits of self-defence).
Another important aspect of Estonian homicide statistics up to the year 2002 was the fact that completed and attempted homicides were viewed together on account of legislative logic according to which attempted and completed crime resulted in prosecution under the same section of the law. The percentage of attempts from the total numbers of homicides varied greatly by types of homicide. In the earlier Estonian crime statistics, causing of fatal bodily injury (CrC § 107 (2) 1)) was not regarded as intentional homicide, which was quite problematic in view of the similarity between this type of criminal offence and homicide with regard to motivation and result10. As demonstrated by international comparisons of judicial practice, some countries prefer to classify violent crimes whose victim survives as attempted homicides, whereas others classify them as causing of serious bodily injury. As a result, the percentage of completed and attempted homicides from the total number of intentional homicides varies greatly by country11. At the same time-as demonstrated by Estonian crime statistics in the Soviet period, among others-the numbers for attempted homicide and fatal bodily injury were almost equal. Given that by the addition of attempted homicides to homicides the total number seemed to rise, while the omission of fatal bodily injury caused the total number of homicides to fall, the total number of intentional homicides in Estonian statistics from that time period can be regarded as quite reliable for the purpose of international comparison.
In 2002, when the PenC entered into force in Estonia, the corpus delicti of intentional homicide have been changed and some alterations were made to the principles for compilation of homicide statistics. The Penal Code now has four corpus delicti'sthat refer to intentional taking of another person's life and are regarded as intentional homicide in crime statistics. The first of these is manslaughter (PenC § 113), the second is murder as an aggravated form of intentional homicide (PenC § 114), third is manslaughter in a provoked state (PenC § 115), and fourth is infanticide (PenC § 116).
It must be pointed out as an organisational characteristic that it is now the Ministry of Justice that compiles and presents national crime statistics, not the police. The statistics differentiate between completed and attempted homicides. At the same time, violent crimes that result in a human fatality but in whose case the death of the victim is not intentional but caused by negligence are still not classified as homicides in Estonian statistics. In that case, the offender's behaviour is qualified as an aggregate of corpus delicti in such a way that PenC § 117 (negligent homicide) is combined with one of the followingcrime events: PenC § 118 (causing serious damage to health), PenC § 121 (physical abuse), or PenC § 122 (torture)12.
The data on intentional homicides are more comparable on an international basis than are the data on other types of crime. At the same time, it is, of course, impossible to know and consider all country-specific nuances arising from under- or over-recording and differences in classification when one is utilising these data13. Comparative data on intentional homicides can be found from several, quite different sources that have been compiled by international organisations. As a general rule, the number of intentional homicides per year per 100,000 inhabitants is used as an indicator.
The UN (through the UNODC, United Nations Office on Drugs and Crime) has been gathering information on crime, including homicides, in member states since 1946. In 1974, a series of special studies on crime trends and the activity of criminal justice systems were commenced. The first of these covered the period from 1970 to 1975 and included 56 member states. The fifth study covered 1990 to 1994 and, among other things, presents the data on intentional homicides in 65 member states. This study employs the concept of intentional homicide ('death intentionally caused by another person, including infanticides by mothers and attempted homicides')14. The newest publication in the series covers the years 1995-200415.
INTERPOL (the International Criminal Police Organization) used to collect and publish crime data obtained from national police organisations since 1950. Criminal homicide was defined as 'any act performed with the intention of taking another person's life regardless of the circumstances of the performance of said act'. These statistics included infanticide but not always attempted homicide. The member states of INTERPOL did not submit crime data on a regular basis, and the number of countries that submitted data varied greatly from one year to the next16. Since 2006, INTERPOL is no longer involved in collecting and publishing the data on intentional homicides in member states17.
WHO (World Health Organization) data on causes of death, collected since 1948, are regarded as one of the best statistical sources. The WHO has defined homicide as 'death resulting from injuries intentionally caused by other people'. In contrast to the two above-mentioned statistics, this proceeds not from consideration of criminal offence but from information on crime victims, and it includes all homicide victims, regardless of how the case has been described in the criminal code of the country in question18. The newest data from this series cover the situation in 200619.
The Council of Europe has compiled international crime statistics for European countries. In 1996, a committee of the Council of Europe-the European Committee on Crime Problems-formed a specialist group that first collected data on the period from 1990 to 1996. In...