SELF-MURDER AND SELF-MURDERERS IN SOCIAL REPRESENTATIONS OF YOUNG RUSSIANS : AN EXPLORATORY STUDY

There are no doubts that the current situation is extremely serious and complex; as a result, the importance of the further development and realization of the preventive measures is obvious. Each preventive program in the field of public health should be based on the results of social psychological studies concerning the problem. In the presented paper, we discuss the results of the exploratory study based on the ideas of the social representations theory. The objective of the exploratory study was to analyze the speciBerezina Elizaveta

ficity of the social representations of self-murder and self-murderers in two groups of young Russians.A total of 106 subjects (67 females and 39 males) aged 18 to 35 participated in the study.It was supposed that the social representation of self-murder and of self-murderers would be less shared in the group of subjects who have friends or acquaintances among people with self-murder experience (committed self-murder or attempted it); it means that the structure of the social representations would be more complex (the central system would be composed of more elements that correspond to different themes on self-murder), whereas, the social representation of self-murder and of self-murderers in the group of subjects who have no friends or acquaintances with self-murder experience (committed self-murder or attempted it) would be more shared (the central system would consist of less themes) and less complex.
The different themes that form the social representations of self-murder and of self-murderers in two groups were revealed.The suppositions got empirical support.
Key words: self-murder, self-murderers, social representations theory, social representation structure, young Russians publications concerned with different aspects of self-murder.For example, analysis of the publications in the database PsycAR-TICLES by using the key word "suicide" reveals an increase of publications during the last 30 years: 127 publications from 1981 till 1991; 596 publications from 1992 till 2002; finally, 830 articles were published between 2003 and 2013.The growth of the scientific interest is obvious.On the other hand, the growth of self-murder during the last 45 years was also observed in people aged 15-44 years (WHO, 2012).
If one takes a closer look at the situation in Russia, the growth of the self-murder rate would be also obvious.The statistics (Valiakhmetov, Mukhamadieva, & Khilazheva, 2012) say that in Russia in 1876 the death rate from self-murder was one on the least among European countries.Indicators calculated for 100,000 people say that the highest rate

Introduction
Self-murder is not a new phenomenon.People took their own lives in different societies and at different times (Zdanow & Wright, 2012).However, the important point is that this cause of mortality becomes one of the leading causes.The premature death caused by self-murder1 is one of the most serious problems of public health in the whole world.About one million people die for this reason each year.The further predictions are rather pessimistic, because this number can reach one and a half million of people by the year 2020 (Westerlund, Hadlaczky, & Wasserman, 2012).These predictions could be based on the fact that the World Health Organization recorded the growth in the number of self-murders in the last four decades regardless of age group (in fact, the age is an important "marker" of self-murder's risk) (Kholmogorova & Volikova, 2012).
The current situation concerning research on self-murder is characterized by a certain paradox.On the one hand, there are many 2011), Russia is still one of the leading countries in terms of mortality from self-murder (Valiakhmetov et al., 2012).
Despite the numerous works on selfmurder in general and on self-murder prevention in particular, the number of self-murders is still growing.Therefore, this problem needs more researchers' interest and innovative preventive measures.The actual preventive programs are based on a variety of different theoretical models that appeal to a wide range of interacting factors (from biological to psychosocial and sociocultural ones) (Conner, Duberstein, & Eric, 2003;Henden, 2008;Westerlund et al., 2012).
Leaving aside the discussion on the polyphony of the scientific explanations that can be found in the literature on self-murder itself and the personality and mental health of a self-murderer (Bae, Ye, Chen, Rivers, & Singh, 2003;Conner et al., 2003;Henden, 2008;McAuliffe, 2002), we would like to highlight one important conclusion from it: self-murder is a polymorphous phenomenon that provokes numerous scientific discussions concerning the reasons of self-murder (for example, psychological, social, cultural, genetic, biological) and concerning the personality of a self-murderer.Similarly to scientists, ordinary people in their everyday life try to answer the same questions about self-murder and self-murderers.Contrary to scientists, ordinary people work out their own theories without any test of theirs hypotheses; they use these theories in everyday life.
Therefore, the objective of the reported study was to reveal the lay explanations of self-murder and self-murderers among young Russians.
The importance of this perspective of analysis is explained by the fact that in everyday life people use these lay explanations facing the danger (self-murder and self-murderers -in case of our study) in order to understand the phenomenon, to protect themselves, and to regulate their social behavior towards people from the so-called "risk groups".The preventive programs need to take into consideration these lay explanations of people in order to be effective.
Among the possible theoretical frames that could be productive in achieving our objective, there are at least two: 1) attribution theories proposed and developed by Heiden, Kelley and others (Kelley & Michela, 1980), 2) social representations theory proposed and developed by Moscovici, Abric, Doise and others (Abric, 2001;Doise, 2001;Moscovici, 1961).In our study, we preferred to follow the ideas of the social representations theory as the most appropriate framework for our objective.
According to one of the many definitions of this concept worked out in the theory of "social representations could be defined as commonsense knowledge about general topics (e.g., AIDS, computers, gender, health, intelligence, psychoanalysis, work) that are the focus of everyday conversation" (Lorenzi-Cioldi & Clémence, 2003, p. 311).Self-murder, definitely, belongs to this list of everyday conversation topics in different social groups; it is an extremely important question of life and death.For example, only one Search Engine "Yandex.ru" reveals 3 millions answers to the term self-murder.In 2012, the law concerning the security of information targeted upon under-age persons was adopted in Russia2 .This fact, on the one hand, refers to the importance of the problem in Russia; on the other hand, it reflects the presence of this problem in the communications on different levels.
The social representations are the form of commonsense knowledge worked out by people in everyday communications in order to give meaning to different objects, phenomena, events, etc. that are new, strange, unknown, threatening.As Moscovici underlines it, "…the purpose of all representations is to make something unfamiliar, or unfamiliarity itself, familiar.What I mean is that consensual universes are places where everybody wants to feel at home, secure from any risk…" (Moscovici, 2001, p. 37).The social representations transform the strangeness of such objects, phenomena, etc., by putting them into the existing frame.It is obvious that this transformation is defensive.Other functions of social representations are: the function of communication facilitation, the function of regulation of social behavior and practice, the function of social identity construction and support, and the function of justification of social relations (Abric, 1994;Breakwell, 2001;Doise, 1986;Moscovici, 1973).As Moscovici says, the social representations realize different types of functions: some of them realize cognitive ones, others -social in order to maintain the identity (Bonnec, 2002).
Thus, facing such a phenomenon as self-murder in everyday life, people try to explain its danger, to understand the reasons for committing it, and try to reveal the specificity of a self-murderer personality.
In the reported study, we followed the ideas of the structural approach developed by Abric (Abric, 2001).According to this approach, "the social representation consists of a body of information, beliefs, opinions, and attitudes about the given object.These elements are organized and structured so as to constitute a particular type of the social cognitive system" (Abric, 2001, p. 43).
The social representation has a particular structure; it consists of central and peripheral systems.The central system gives meaning to the whole representation, it maintains the stability of the representation, and it plays the main role in the organization of the representation; the peripheral system plays a crucial role in the protection of the central system stability (and, therefore, in the protection of the representation itself against any transformations); also, it plays an important role in the concretization and adaptation of the central system elements (Abric, 2001).The central system is "linked to collective memory and to the history of the group" (Abric, 1993, p. 76) that shares the representations, whereas, the peripheral system is linked to individual memory and experience.The central system is consensual and provides for the homogeneity of the group, whereas the peripheral system is not consensual, it provides for the heterogeneity of the group (Abric, 1993).As one can see, there is a hierarchy in the structure of the social representation: the central system is formed by the main elements and the peripheral systems are organized around the less important elements.
There are two main factors that influence the central system: social practice and social influence (Abric, 1994;Galand & Salès-Wuillemin, 2009;Moliner, 2001).The social practice determines the relation between persons and the object of the representations (Abric, 2001;Moliner, 2001).The social influence refers to the fact that social representations are constructed in the social interaction; the social influence functions through the communications concerning the object of social representation (Galand & Salès-Wuillemin, 2009).
There are several studies concerning the social representations of self-murder (selfmurderer has not yet become an object of studies in the social representations theory).Among the studied questions are the following3 : how do adolescents understand selfmurder (Cristea, Cicioc, & Alexoaie, 2008), how self-murder is represented by health workers (da Costa Araújo, Leal Vieira, & da Penha de Lima Coutinho, 2010; Matão, Miranda, Campo, de Sousa Borges, & Pereira, 2012), how does the cultural context affect the content of the social representations of self-murder (Arthi, 2008) etc.
Our objective in the present study was to analyze the specificity of social representations of self-murder and self-murderers in two groups of young people.The factor of social influence was implemented by the fact that one group of subjects was composed of friends and acquaintances of people with self-murder experience (committed selfmurder or attempted it).Another group was composed of people who had no friends or acquaintances with self-murder experience.
Applying the ideas of the social representations theory (Moscovici, 1961(Moscovici, , 2001) ) and taking into consideration the results ob-tained in a study on social representations of AIDS and of People living with HIV/ AIDS (PWA) where it was shown that the social representations of PWA were more simple in case of students (the biggest distance from the AIDS problems) and more complex in groups of homosexuals and drug addicts (the least distance towards the AIDS problems), on the one hand (Bovina, 2007), and following the methodology of the structural approach (Abric, 2001), on the other hand, it was supposed that the social representation of self-murder and of self-murderers would be less shared in the group of subjects who had friends or acquaintances among people with selfmurder experience (committed self-murder or attempted it); it means that the structure of social representations would be more complex (the central system would be composed of more elements that correspond to different themes of self-murder), whereas, the social representation of self-murder and of self-murderers in the group of subjects who had no friends or acquaintances with self-murder experience (committed selfmurder or attempted it) would be more shared (the central system would consist of less themes) and less complex.Our study was descriptive and exploratory in its nature, that's why the only general supposition was formulated.

Sample
A total of 106 4 young Russians aged 18-35 years (М = 23.48,SD = 4.36) participated in the current study (67 females and 39 males).Group 1 comprised friends and acquaintances of people who committed or attempted to commit self-murder (44 subjects); group 2 was a control group without friends and acquaintances of persons who committed or attempted to commit self-murder (62 subjects).The study was totally anonymous, subjects participated voluntarily, they were recruited at several universities of Moscow.The sample was composed of students from the second to fifth years of studies in social sciences.

Method
Following the ideas of the structural approach of social representations theory (Abric, 2001), we used the free association technique.The free association technique is usually used in the analysis of social representations, because it reveals the latent dimensions of the semantic universe of the studied object (Abric, 2003).In the numerous studies where the free association technique is employed, the subjects are presented with a key word (or key words) and are asked to give three to five words that come to their mind immediately (for example, Bonnec, 2002;Dany, Urdapilleta, & Lo Monaco, 2014;Flament & Rouquette, 2003).From these data, the structure of social representations was revealed.

Procedure
The paper-and-pencil questionnaire was administered in the study.The study was presented as a survey on youth health problems.The questionnaire consisted of three parts: in the first part, the free association technique was used with two key words -"self-murder" and "self-murderers", and some open-ended questions about self-murder and self-murderers were asked; in the second part, the free association technique was also used with two key words "drug addiction" and "drug addicts", and some openended questions about drug addiction and drug addicts were asked; finally, in the third part, there were some open-ended questions about health and health promotion.
In this paper, we present the results obtained only in the first part.
The questionnaire was administered to small groups from 2 to 16 persons; each person completed an individual copy of the questionnaire.In accordance with the free association technique, the subjects were asked to write down no less than five words that first came to their mind when the key word self-murder was presented to them.The same technique was used for the second object of social representationsself-murderers.The participants were asked about their knowledge of persons who committed or attempted to commit self-murder.Some socio-demographic questions were also asked.The participation in the study took from 15 to 20 minutes in general.
The key word "self-murder" was used because of the frequency of its usage (two times more frequent than the word "suicide") in public discourse, according to the Engine systems Yandex.ru,Mail.ru,Rambler.ru,Google.ru.
The prototypical analysis proposed by Vergès (Vergès, 1992) was applied to the collected data.The so-called rank-frequency method (Dany et al., 2014) was used to reveal the structure of the social representation of self-murder and the structure of the social representation of self-murderers.The rank-frequency method takes into account two parameters: 1) associations' frequencies (quantitative parameter5 ); 2) rank of their appearance (qualitative parameter6 ).The usage of these criteria allows us to reveal the most salient words.
The combination of these criteria gives us a four-cell table that corresponds to different zones of social representation.As it was said before, the structure of the representation included the central and the peripheral systems.In the peripheral system, two parts could be distinguished: the potential change zone and the peripheral system.The potential change zone is part of the peripheral system that could bring changes to the central system and change the representation as a result of it.The further theoretical analysis is needed in order to understand the inner structure of the peripheral system (Sa & Oliveira, 2002).
1) the central system zone (the cell contains associations with the high frequency and low rank of appearance; in other words, these associations come to mind of many people immediately when they think about the object of the representation, the consensual elements are situated in this cell) (Table 1); 2) the potential change zone (this zone unites two cells; the first cell contains associations with the low frequency and low rank of appearance; it means that these words are shared by the minority of people, these associations come to subjects' mind immediately when they think about the object of the representation; the second cell contains associations with the high frequency and high rank of appearance; in other words, these associations are shared by many people but do not come to the subjects' mind immediately when they think about the object of the representation) (Table 1); 3) the peripheral system (the cell contains associations with the low frequency and high rank of appearance (Table 1); in other words, these associations are shared by few people and these words do not come to the subjects' mind immediately when they think about the object of the representation.
Once again we will remind here that the central system of representation contains the elements that give the meaning to the whole representation and organize it; also, they determine the stability of the representation (Abric, 1993;Flament & Rouquette, 2003).While the peripheral system unites elements that function as an interface between the central system and the reality, the peripheral system unites the elements that are linked to the subjects' individual experience and individual memory.

Results
Social representations of self-murder in two groups.A total of 213 associations were revealed using the free associations technique with the object "self-murder" in group 1, and 294 associations in group 2. The prototypical analysis revealed the structure of the representations, the answers that were done by at least 5% of subjects were analyzed here7 ; the volumes of the central and the peripheral systems are 51.64% and 47. 96% correspondingly in group 1 and group 2.
The peripheral system was not revealed.
In group 2, the central system of the social representation of self-murder consisted of four elements (see Table 3): "death", "hopelessness", "desperation", and "morally weak person".The potential change zone united the following elements: on 8 This difference in the potential change zone is made because these elements occupy different places in the structure of the social representations.The associations with the low frequency and low rank of appearance can be seen as important elements (according to their rank of appearance) shared by the minority, while the associations with the high frequency and high rank of appearance can be seen as less important elements shared by the majority.Note: For each association, the frequency and the rank of its appearance are indicated in parentheses.Note: For each association, the frequency and the rank of its appearance are indicated in parentheses.Note: For each association the frequency and the rank of its appearance are indicated in parentheses.Note: For each association, the frequency and the rank of its appearance are indicated in parentheses.
the one hand "mentally ill person", "horror", and on the other hand "foolishness", "why?", "fear".The peripheral system included elements: "pain", "jump from a height", "unanswered love", "loss of the meaning of life", "lack of understanding".Social representations of self-murderers in two groups.A total of 204 associations were revealed as a result of the free associations technique with the object "selfmurderers" in group 1, and 271 associations were revealed in group 2. The prototypical analysis revealed the structure of the representations; the volumes of the central and the peripheral systems are 31.64%and 30.99%, respectively, in group 1 and group 2.
In group 1, the central system of the social representation of self-murderers consisted of the following five elements (see Table 4): "madmen", "morally weak persons", "desperation", "unhappy", and "adolescents".
The potential change zone united the following elements: on the one hand, "foolishness", "lost", "pity" and on the other hand, "loneliness".The peripheral system included the following elements: "fear", "grief", and "sorrow".
In group 2, the central system of the social representation of self-murderers was composed of the three following elements (see Table 5): "morally weak persons", "stupid", and "fool".
The potential change zone united, on the one hand, "pity", "adolescent", and "death", and on the other hand -"mental disorder".The peripheral system included only one element: "lack of understanding".

Analysis
Social representations of self-murder in two groups.In the central system of the representations of self-murder in group 1, there is the element "desperation".It indicates the state of a person who is going to murder himself (according to the analysis of the studies on self-murder, for example, Heden, 2008;Fowler, 2012, it is one of the factors that predict the self-murder attempt).The common sense also associates this state with self-murder.In the peripheral system, there are other references to the state of the person: "helplessness", as well as some more predictors of self-murder: "loneliness", "problems", "mental disorder".The other element of the central system -"foolishness" -refers to the evaluation of the act of self-murder; in the peripheral system, there are other elements that evaluate the act of self-murder.They are "escape from the problems" and "thoughtless action".The self-murderer is evaluated as a "morally weak person" (this element is situated in the peripheral system).
The element of the central system "jump from a height" is related to the way of selfmurder.Other element of the central system, "grief", is associated with the reaction of the inner circle of the self-murderer to his action.
Finally, the element "death" refers to the result of self-murder.
Looking differently at the elements of the central system, it is possible to notice that there is a script of self-murder in them: desperation -jump from a height -deathgrief.
In the central system of the social representation in group 2 there are two references to the state of a person who is going to commit self-murder -"hopelessness" and "desperation".Once again, as it was found in the literature (for example, Henden, 2008;Fowler, 2012), certain characteristics of the state of a person were seen as predictors of self-murder risk.Of course, social representations are not a simple replication of scientific ideas; they are a complex mixture of different knowledge systems.
The subjects in this group associated self-murder with other reasons to die: "ill person", "unanswered love", "loss of meaning of life", "lack of understanding" (these elements are situated in the peripheral system).Another element of the central system of the representation is an evaluation of the actor of self-murder -he is seen as a "morally weak person", the evaluation of selfmurder itself is situated in the peripheral system: "foolishness".
The element "death" refers to the result of self-murder.
At the first glance, there is a certain similarity between the elements in case of social representations of self-murder in two groups, namely: "desperation", "death", "jump from a height"9 , "hopelessness", "foolishness", "morally weak person", "pain", "mental disorder" (or "mentally ill person").However, as explained before, the structure of the social representation has a certain hierarchy, so a comparison of representations should take into consideration the positions of the elements in the structure of representation.The central system of social representations of self-murder in group 1 is different from the one in group 2. As has been said before, the central system is consensual, and it provides for the homogeneity of the group; whereas, the peripheral system is not consensual, it provides for the heterogeneity of the group (Abric, 1993).As a result, the comparison of the peripheral system is not important in order to compare the representations.
The key themes of the representation of self-murder in group 1 were: 1) state of a person who is going to murder himself, 2) way to commit self-murder, 3) evaluation of the act of self-murder, 4) reaction of the inner circle of the self-murderer to his/her action, 5) the result of self-murder.The main themes of the representation of self-murder in group 2 were: 1) the state of a person who is going to murder himself, 2) evaluation of self-murderer, 3) the result of self-murder.As a result of comparison of these themes, the first conclusion can be formulated: the social representation of self-murder in group 1 is less shared in comparison with the social representation of self-murder in group 2. Subjects in group 1 revealed more factors that lead to self-murder than the subjects in group 2 did.Another observation could be mentioned here that the central system of the representation in group 1 contains the element "grief" which indicates the relations towards self-murderers; this element does not exist in the representation in group 2.
Finally, one interesting observation can be mentioned here.In both groups of subjects there were no references to the macro-social or economic aspects of the situation that could be the reasons for selfmurder.The self-murder is seen almost as an individual act with references mostly to the individual construct; the mass suicide was not mentioned by the subjects.
Social representations of self-murderers in two groups.The element "madmen" of the central system of the social representation in group 1 refers to the reason to commit a self-murder.Two other elements of the central system -"desperation" and "unhappy" -reveal the state of a person that leads to self-murder.A self-murderer is characterized as a "morally weak person"the element of the central system; the other characteristics are situated in the different zones of the peripheral system -"foolishness", "lost", "loneliness".The last theme, which is introduced in the central system by the element "adolescents", relates to the group of risk.The presence of this element in the central system could be interpreted in the following manner: on the one hand, it may be a mere fact that people who tried or commit self-murder are more seen as people from this age group; on the other hand, the references towards a certain group may have a defensive function, protecting the subjects from danger of self-murder because it is attributed to the other group.
The theme of the affective reactions towards self-murderers is presented by the elements "pity"10 , "fear", "grief", "sorrow", situated in the different zones of the peripheral system; the positions of these elements in the structure of the representation indicate the secondary role of this theme in the representation.
In case of the social representation in group 2, it is easy to reveal the only theme that organizes this central system -characteristics of self-murderer: "morally weak persons", "stupid", and "fool".The other themes play a less important role in the representation of self-murderers because of their position in the structure of the representation.They are: 1) affective reactions towards self-murderer ("pity"); 2) the reasons to commit self-murder ("mental disorder", "lack of understanding"); 3) group of risk ("adolescent").
In a similar way as in the case of social representations of self-murder, it is easy to notice that there are some similar elements in representations of self-murderers in both groups, namely: "morally weak persons", "fool" ("foolishness"), "adolescent" ("adolescents"), "pity", "madmen" ("mental disorders").However, these elements are situated in different systems of social representations, and the difference of the central systems of representations explains the difference of representations themselves in the two groups.
The key themes that form the representation of self-murderers in group 1 were: 1) reason to commit a self-murder; 2) the state of a person who is going to murder himself; 3) characteristics of a self-murderer; 4) the group of risk.The main theme of the representation of self-murder in group 2 was characteristics of a self-murderer.
A comparison of themes situated in the central systems of two representations shows that there are four themes in the central system of social representation of self-murderers in group 1, and there is only one in case of social representation in group 2; as a result, the representation of self-murderers in group 1 is less shared than in group 2.
One particular observation could be discussed here.In both groups, the selfmurderers feel, they suffer, but they "do not think", references towards the cognitive sphere were not revealed.
Putting together the results of comparisons of social representations of self-murder and self-murderers in two groups, we can conclude that our general supposition (that the social representation of self-murder and of self-murderers would be less shared in the group of subjects who have friends or acquaintances among people with selfmurder experience (committed self-murder or attempted it); in other words, the structure of social representations would be more complex (the central system would be composed of more elements that correspond to different themes of self-murder and of selfmurderers), whereas the social representation of self-murder and of self-murderers in the group of subjects who have no friends or acquaintances with self-murder experience (committed self-murder or attempted it) would be more shared (the central system would consist of less themes and be less complex) got empirical support.

Conclusion
We conducted an exploratory study following the ideas of the social representations theory and using the methodological strategy of the structural approach of this theory in order to reveal the specificity of representations of self-murder and of selfmurderers kept by two groups of young Russians.The lack of research concerning the social representations of self-murder and of self-murderers did not allow us to put forward more precise suppositions.In our study, it was revealed that the central system of the social representation of selfmurder in group 1 the was composed by the following themes: 1) state of a person who is going to murder himself, 2) the way of self-murder, 3) evaluation of the act of self-murder, 4) reaction of the inner circle of the self-murderer to his action, 5) result of self-murder.The main themes of the representation of self-murder in group 2 were: 1) the state of a person who is going to murder himself, 2) evaluation of selfmurderer, 3) the result of self-murder.The structure and its complexity are different in two groups of subjects.So, in each group the representation is crystalized around these main themes.
In case of the social representation of self-murderers, the main themes that are situated in the central system of the representation group 1 were: 1) the reason for committing a self-murder; 2) the state of a person who is going to murder himself; 3) characteristics of self-murderer; 4) group of risk.In case of group 2, the only theme was situated in the central system (characteristics of self-murderer).
Our study has some limitations.First of all, in this exploratory study the sample was composed only of young people, mostly students of social sciences; the future study should comprise a bigger sample.
The second limitation is related to the lack of data obtained from people who attempted to commit self-murder.The future studies that will include this group of subjects will highlight the transformation of the social representations of self-murder and of self-murderers as a result of transformation of positions from the "observer" to the "actor".
The third limitation is related to the usage of the "rank-frequency method".More precise attention should be paid to the methodological aspects of the "importancefrequency method" (Dany et al., 2014).
Finally, the obtained results do not allow us to formulate some concrete preventive recommendations concerning the selfmurder; however, the social representations theory has an appropriate potential for the recommendations of preventive measures (for example, Dany & Apostolidis, 2002;Paicheler, 1993), and the future studies will provide us with the sufficient data for it.
Nevertheless, the objective of our exploratory study was reached; the first steps towards the comprehension what representa-tions about self-murder and self-murderers are used in everyday life have been made.