• Return to Medical Papers

    FULL PAPER (全文ー英語のみ)


    Yoshitomo Takahashi, M.D. (1), Douglas Berger, M.D. (1,2)

    PDF Format:

    (1) Tokyo Institute of Psychiatry, Tokyo Japan. (2) Department of Psychiatry, Albert Einstein College of Medicine, Bronx New York.

    Suicide and the Unconscious, Leenaars A. and Lester D. (Eds.), Northvale, Jason Aronson, pp. 248-258; 1996.

    An overview of all the intricacies of Japanese behavior and unconscious thought processes discussed in the literature would necessitate a book of its own. For the purposes of this chapter we would like to introduce a few common concepts related to the Japanese concept of the unconscious. First is that there is an overwhelming and intense desire for the establishment of an identity by belongingness to a group (Berger, 1985; Christopher, 1983; Doi, 1973; Sugiyama-Lebra, 1976; Takahashi, 1992). A sense of oneness "ittaikan" is generated unconsciously with the group and there is a social sensitivity to any possible disruptions in the harmony of this relationship. Ostracism from the group is to be avoided at all costs. Pressures to conform to a limited pattern of behavior and thought and an expectation for a total commitment to the group accompany this.

    A consequence of this is that both the pride and shame of an individual are shared by the group and vice-versa. Shame, conscious and unconscious, is thus quite powerful and is often an important factor in suicide in Japan. Guilt is involved in relationships of reciprocity where a favor "on" is accompanied by the burden of the duty of reciprocity "giri" which may be very intense and emotionally uncomfortable (this is in contrast to the western sense of guilt that stems from an internal sense that one has done something wrong). This sense of a never ending "owing" is quite pervasive in the unconscious of the Japanese especially in relation to those who have taken care of oneself.

    Suicidal Japanese have similar "risk factors" for suicide to those in other countries. These risk factors include psychiatric disorders, substance abuse, prior suicide attempts, lack of social support systems, male gender, older age, various kinds of loss, family history of suicide, accident proneness, etc. (Takahashi, 1992, 1993a, 1993b). Reports on suicide in Japan in English language often overemphasize the cultural difference.

    This paper will discuss some of the culturally-related dynamics and unconscious processes related suicide in Japan. After a review of the demographics of suicide in Japan, we will present two exemplary kinds of suicide peculiar to Japanese context, "shinju" and "inseki jisatsu."


    After World War II, there was a peak in the number and the rate of suicide in Japan in the 1950s, followed by a decline, then a second peak in mid-1980s (National Police Agency, 1993; Takahashi, 1993a). In 1986 there were 25,524 suicides in Japan, which was the highest number of suicides in any year since the war. However, since 1987 both the number and the rate of suicides gradually decreased despite various socio-economic problems such as changes in social values, family structure, increase in crime and substance abuse, severe competition in society, and a widening gap between the rich and the poor. All of these factors have been reported by suicidologists to be associated with an increasing suicide rate.

    In 1992 the declining tendency waned and there were 22,104 suicides, an increase of 4.8 percent. Economic recession might be an explanation for this increase, but interpretation should be postponed for several more years. In summary, there have been about 21,000 to 22,000 suicides in Japan annually in recent years, which represents a suicide rate of 17 to 18 per 100,000 per year. Although this number is about twice as many as that of those killed by traffic accidents, attention to suicide prevention in Japan has unfortunately not been adequate.

    As for the variation in suicide rates by age, in the 1950s and 1960s the suicide rate showed two peaks, one in young adults and one in the elderly. At that time, Japan had one of the highest suicide rates in the world. For the past three decades, the suicide rate in young people has been decreasing. The suicide rate in the elderly has also decreased, but this decrease has not been very large. Japanese still exhibit higher suicide rates with aging. In 1992, while those over 65 years old accounted for 13 percent of the population, suicides in this age group represented 27 percent of all suicides. While suicide is an especially serious problem for the elderly, it has been mostly ignored by Japanese society (Takahashi, Hirasawa, Koyama, Asakawa, Kido, Onose, Udagawa, Ishikawa, & Uno, in press).

    Although it has been suggested that attitudes toward suicide vary between cultures, there have been few comparative studies between Japan and other cultures. Domino and Takahashi (1991) studied medical students' attitudes toward suicide in Japan and the United States using Suicide Opinion Questionnaire and the findings indicated that the Japanese students were more likely to think that suicide is not always an abnormal behavior and it might be permissible in certain situations. In contrast, the American students were more likely to think that suicide is usually a result of some psychiatric problem with aggression expressed toward oneself, and that suicide should be prevented. Because the sample size was limited, in this study the findings should not be overgeneralized. The study is meaningful, however, because it utilized the same data base in both countries. The results suggest that, in a culture where suicide may be permissible, social factors may need to be taken into consideration. Attitudes toward suicide may differ in these cultures even if the suicide takes place while the person is in a normal mental state.


    The word shinju in Japanese, originally meant a mutual suicide agreement by lovers in order to prove the genuineness of their love to each other. According to the strict and original definition described by Ohara (1985), shinju is an act where more than two people commit suicide voluntarily at the same place, at the same time, and for the same purpose. The definition has become looser and now also includes murder-suicides where some of those involved are killed against their will (Fukushima, 1984). The definition of shinju now includes both a genuine suicide pact, extended suicide (assisted suicide followed by suicide), and murder-suicide in which the killer and the victim(s) have a strong emotional tie with each other before the act. The feeling of "oneness" of those involved in shinju is important.

    Shinju literally means "heart-inside" or "oneness of hearts" (Walsh, 1969) and this probably reflects a psychological joining of the participants. The participants may likely have a conscious awareness of wanting to join or unite with their partner in the afterlife, but unconsciously there may be a wish for infantile symbiosis and an intolerance of separation anxiety. The development of clear ego boundaries is probably impaired in these individuals though state-related regressions may also be at play in some individuals. Uniting with their suicide partner may be a reaction for failure to fit into society or a group.

    The Japanese language has diverse words for shinju. Shinju has been classified into two major categories, johshi (mutually consented lovers' suicide) and oyako-shinju (parent-child suicide), the latter of which is subclassified further such as boshi-shinju (mother-child suicide), fushi-shinju (father-child suicide), and ikka-shinju (family suicide). The number of johshi have been declining in the past three decades. Although the number of oyako-shinju has been declining since the 1950s as well, it is still a serious problem.

    Most cases of shinju are boshi-shinju in which the children, who are too young to decide on suicide themselves, are killed by their mothers. Ohara and Inamura have both pointed out that boshi-shinju and fushi-shinju have important differences (Ohara, 1963, 1965; Inamura, 1977, 1993). In boshi-shinju mothers in their 20s and 30s kill their children and then commit suicide. The children most often victimized in boshi-shinju are of preschool age. In fushi-shinju the fathers (who are usually older than the boshi-shinju mothers) kill their children (who are older than the victims of boshi-shinju), and then commit suicide. The most common reasons for boshi-shinju are psychiatric disorders and family conflicts, while those of fushi-shinju are financial problems and physical illness. Japanese often show considerable sympathy toward parents who are not able to find any other recourse but to commit suicide with her/his children.

    Japanese society as a whole fosters a mutual interdependency in the socialization process and this is quite a contrast to the emphasis placed in individuality put in the west (Christopher, 1983). In order to promote this mutual dependency there is probably some unconscious muting of the separation-individuation process on the part of caregivers as described by Margaret Mahler (1972). As a consequence, the boundaries, both conscious and unconscious, in ones nuclear family can be more blurred than in western society and this may have lead to the development of shinju as a Japanese cultural phenomena. Post partum depression may also be an important factor in boshi-shinju.

    Marzuk, Tardiff, and Hirsch (1992) have recently reviewed research on murder-suicide, but the term "murder-suicide" used by Marzuk et al. has some important differences from "shinju". Thus, it should be kept in mind that "murder-suicide" and "shinju" are not quite the same, an indication that suicide is often interpreted differently in different cultures and will be discussed below.

    Marzuk et al. reviewed the studies on suicide published in the United States over the past 30 years and found that the incidence of murder-suicide was 0.2 to 0.3 per 100,000 per 100,000. The overall suicide rate, which has been fairly stable over this period in the United States, was about 12. They estimated, therefore, that the portion of murder-suicide among all suicide was about 1.7 to 2.5 percent. They also reviewed 17 papers published from 1900 to 1979 in various countries in North America and Europe and found that the rates of murder-suicide in those countries were similar, 0.2 to 0.3. While the reason for this similarity is not clear, Marzuk et al. pointed out that, although the incidence of murder-suicide is similar, the type of murder-suicide may differ from country to country. For example, in the United States one-half to three-quarters of murder-suicides are those in which a husband murders his wife and then kills himself. In Britain and Japan, most murder-suicides are those in which a mother kills her children and then commits suicide.

    Inamura (1993) reported that shinju has accounted for 1.6 percent of all recent suicides. Based on this figure, the rate of shinju among all suicides is estimated to be 0.29, which is similar to that reported by Marzuk et al. In Japan, the type of shinju has been changing; in 1950s johshi was the most prevalent, but since the 1960s the number of boshi-shinju has become the most common.

    While the incidence of murder-suicide is not all that different in different cultures, each society deals with it differently, as Marzuk et al. have noted. Aizawa (1975) also pointed out in his paper on shinju that "The important thing in studying the relationship between shinju and Japanese culture is not to focus only on statistical analysis and literature on shinju in Japan and other cultures, but to concentrate on the sympathy Japanese have toward shinju." (p. 143) Similar cases may be interpreted differently in different cultures. Most Western cultures would consider this to be a murder-suicide, and not a parent-child suicide, because the children usually do not commit suicide on their own free will but are killed by their parents. The Santa Monica (California) case was a typical example of this kind of suicide (Japan Times, 1985). On January 29, 1985, a 32-year-old Japanese immigrant woman tried to drown herself, her infant daughter and her four-year-old son by entering the ocean on Santa Monica Beach. Although they were quickly pulled out from the water, only the mother survived. She was tried in a California court for child abuse and first-degree murder. When this was reported in Japan, it shocked Japanese society. There was also the added feature for the Japanese of being subjected to foreign censure in this case, something not easily accepted.

    This mother had attempted oyako-shinju about a week after discovering that her husband had been having a secret extramarital affair for years, leaving her depressed and ruminating about suicide. The reasons for her despair were personal, and although maladaptive, the method she chose to resolve it was cultural, and very Japanese. Although she had lived in the United States for 14 years, she remained Japanese in her thinking and life style, isolated from American culture. She did not drive, spoke little English, knew nothing of her husband's business, and had no hobbies or close friends outside the family. In other words, she was virtually without any kind of support system which might have sustained her in time of emotional distress. Social supports have also been found to be important for preventing suicide in Western society (Berger, 1993).

    In Japan, the mother-child bond and the mother's dedication to the child are very important. Why then, is infanticide committed by the mother relatively common in Japan? Paradoxically, it is this very bond between mother and child that causes oyako-shinju. According to Japanese logic, the suicidal mother cannot bear to leave the child to survive alone; she would rather kill the child because she believes that nobody else in the world would take care of the child better than she, and that the child would be better off dying with her.

    At this point we should introduce the concept of amae which was popularized by Takeo Doi a well-known Japanese psychiatrist (Berger, et. al., 1994; Doi, 1973). Amae may be described as a mutual dependency where the assurance of another's good will permits a certain degree of self-indulgence. Much of this process is unconscious and partly explains the insularity of Japanese society because foreigners cannot readily fit into this mutual good will interaction.

    A western concept that Doi felt was equivalent to amae is that of "passive object love" described by Michael Balint (1965). The importance of amae is that it is a harmonious state of affairs where one can be relieved of the burden of "on" and "giri". When the unconscious expectations of amae fall apart, or if for example the mother feels she cannot provide for an amae relationship for her children in the face of severe family conflict, this may be a risk factor for shinju. Shame and severance of ones connection to the group (family) are important factors as discussed above.

    This feeling of oneness (which may be delusional in quality) and symbiosis between a mother and her children has intensified as a result of the breakdown of the traditional community in which children belonged to a wider circle where they had fictive as well as substitute parents along with their real parents. Today, because of the development of the nuclear family, children, particularly those of preschool age, belong to their parents alone. The mother-child bond may also be reinforced in Japanese society because females do not usually have the emotional and social supports afforded by a career owing to the male-focused orientation in the work place, lack of an outside social support system and a 24-hour focus on the family. When boshi-shinju takes place, sympathy is usually given to the mother who was not able to think about ways other than shinju to solve her problems. Japanese society in general rarely accuses the mother of infanticide.

    In Japan, or in other Asian countries where Confucianism is prevalent, conscious and unconscious, symbiotic ties between oneself and significant others are very common and these kinds of relationships may be valued highly (Takahashi, 1989). In this situation, murdering one's children can be unconsciously regarded as murdering a part of oneself. The mother may not necessarily realize that she is killing another human being separate from herself, but unconsciously rather feels as though she is killing a part of herself. In the psychology of the mother who commits boshi-shinju, killing her own children is equivalent to actually killing herself, and this is interpreted as a kind of "extended suicide". It is impossible for her to imagine a world for her children after her death. Japanese society is often sympathetic to this dynamic. In addition, Japanese have a general reluctance to criticize each other which may be related to a filial piety to the group. However, we should also note the proverb "Shinu kiga areba nandemo dekiru" which means "If one has a will to die, one could do anything." The existence of this proverb probably reflects Japanese society's ambivalence to shinju, both praising and demanding suicide in the same proverb.

    Some boshi-shinju may be a way for a wife to get revenge on her husband. She may react to the discover of her husband having an extramarital affair or demand for divorce by killing her children in order to punish her husband after she commits suicide. In most of cases of boshi-shinju, however, there is a strong tendency for the mother to consider the children as an essential part of herself. In Japan, this tendency is observed even among parents with much older children. It is believed that children cannot or should not be left alone in the world where parent(s) have killed themselves. The children are killed before the parent commits suicide, because they are loved deeply. It may also be that because they are considered a part of the parent, she/he should take care of them by themselves. An alternative interpretation could be that pathological separation-individuation (in the Western sense), or an overly symbiotic oneness (in the Eastern sense) could have caused blurring of unconscious ego boundaries.

    Western society, in contrast, would usually consider a child, even an infant, as having a separate existence from the parent. The concept of oyako-shinju would not easily take root in this cultural context. In Western society it would be felt that the suicidal mother chooses to kill her children for selfish motives. The converse is true in Japan where the concept of children separate from their parents has not taken root. (Though there is intervention by social agencies, children are rarely taken from their parents in abusive situations). In the Western cultural context the mother is usually held responsible, and may be tried for murder. In Japan there would also be a court case, though the sentencing would likely be lighter and more sympathy given to the mother.

    Though there are different unconscious dynamics involved here from those of Western psychology it should be emphasized that the adults involved in shinju do not represent the norm of Japanese society. These are usually individuals who become desperate due to a combination of a life stressor, concomitant psychiatric illness such as depression or psychosis, and premorbid personality vulnerabilities (possibly personality disorder) which then interact with certain cultural tendencies. The average psychologically healthy Japanese would not consider shinju as a solution to their problems (The vast majority of Japanese who experience the family stressors thought to be associated with shinju do not engage in this behavior).

    As clinicians practicing in Japan, we have to weigh the risk of shinju carefully when we evaluate patients who may be at risk. If the patient has a poor social support system, therapists have to mobilize available resources and solicit as much support from relatives as possible in order to prevent shinju.


    Throughout the history of Japan, there has been almost no period when suicide was prohibited by law. The exception was in the early 18th century when a cluster of joshi suicides (lovers' suicide pact) was triggered by melodramas written by Chikamatsu Monzaemon. The government prohibited johshi in order to prevent these suicides. If a joshi occurred, a funeral was prohibited, and the bodies were left in the public view at Nihonbashi Bridge for three days. If one died and the other survived, the survivor was convicted of murder.

    The most widely known form of suicide outside of Japan is harakiri, or seppuku, which means suicide by stabbing one's abdomen. Harakiri was the ritual form of suicide practiced by feudal warriors to show that they accepted responsibility for their actions (Fus・ 1985), and had its beginnings about 1,000 years ago. Japanese warriors used to respect the abdomen because it was considered to be the seat of the soul, so when they assumed responsibility for some serious action or course of conduct, they would cut open their abdomen to demonstrate their innocence and bravery. Cutting one's abdomen itself was not a very effective way to kill oneself so that in the ritual another person severed their neck.

    Since the late 19th century when the Edo era ended, harakiri has rarely been practiced either as a form of suicide or as a punishment. The harakiri suicide of the famous Japanese novelist Yukio Mishima in 1970 was an extremely exceptional case, and astonished even the Japanese.

    While contemporary Japanese do not commit suicide by harakiri, inseki-jisatsu is a form of suicide sometimes regarded as a way of taking responsibility. Inseki means taking responsibility and jisatsu means suicide in Japanese. Toyomasa Fuso, a sociologist and suicidologist, has been conducting suicide research from a cross-cultural perspective and suggested that suicide often takes place when political or social scandals occur in Japan.

    Fuso (1985) noted an interesting difference here between Japan and the United States. Some of the officials who were found guilty in the Watergate Scandal, when on bail or after release from prison, wrote memoirs, and gave lectures.....Almost no one killed themselves suffering from the crime they committed. When scandals occur in Japan, persons who hold important information and feel loyal to the key figures who actually control from behind the scenes often commit suicide. It is rare for the key figures themselves to commit suicide. (p.208) The key differences here may relate to the strong bond people have to members of the same group in Japan, reminiscent of the bond to the mother (the other side of the coin is the strong exclusion of those not in the group).

    The Japanese individual will feel an intense indebtedness to their group, and as noted above their unconscious need to relieve this burden can result in their taking responsibility as a way to prevent ostracism from the group. This can be seen in everyday life where workers may work late into the night for fear that their coworkers would resent them if they left earlier than the group as a whole. While superficially it seems that the individual is sacrificing for the group it is actually the unconscious and preconscious needs of the individual to maintain group acceptance that is expressed in manifest self-sacrifice.

    Ono has stressed how Japanese evaluate themselves in his discussion of the difference between the western concept of social phobia and the Japanese concept of social phobia or "taijin kyoufu" (Ono, 1991). The western social phobics are concerned about their behavior as assessed by their own standards while the "taijin kyoufu" patients assess this by the standards of others. This is related to two kinds of shame. In western shame the feeling comes from the discrepancy between the person's own ego ideal and his behavior, resulting in anxiety that this discrepancy will be noticed by others. The Japanese type of shame on the other hand is characterized by the feeling that comes from the concern about how others view their behavior and the fear that as others become aware of their shortcomings they will be excluded from a group of significant others.

    The mass media in Japan has not recently reported suicide cases in an exaggerated fashion. This does not hold, however, for cluster suicides in young people or for suicides of those involved in political scandals. The mass media report such cases repeatedly and sensationally. While only the facts are briefly reported for ordinary suicides, great details of the behavior before the suicide and the methods of the suicide are often reported for inseki-jisatsu. They usually do not touch upon the psychiatric problems that might have existed in those killing themselves. There is also a trend for the suicide to be reported as a way of taking responsibility for some wrong doing.

    Often, someone who has important information about the facts of the situation commits suicide and leaves a note saying something like "I did nothing wrong. However, I have caused a great deal of trouble to my organization. Therefore, for this I will take responsibility by committing suicide." De Vos (1968) called exaggerated self-identification to a role given by others or an organization to which an individual belongs, "role narcissism." De Vos pointed out that suicide may take place because unconscious identification to a group (or its leader) is so strong that it becomes almost impossible to imagine dissolution of the group.

    In Western culture, the high value placed on individuality does not lend itself to the type of suicide resulting from over-identification to a group (such as inseki-jisatsu), consequently, statistics are not available to compare with inseki-jisatsu in Japan.


    Suicide is a complex human behavior that includes multiple unconscious processes and needs to be interpreted multidimentionally from a biopsychosocial perspective. Suicide should not be interpreted on a psychiatric or a cultural perspective alone but by an integrated view of these variables. We have presented a general overview of those unconscious and cultural factors involved in suicide in contemporary Japan, though by no means the only important aspect involved in any individual suicide.


    Aizawa, S. (1975). Cultural background of shinju. In Suicidology: Suicide and Culture, ed K. Ohara, pp. 142-182. Tokyo: Shibundo. (in Japanese)

    Balint, M. (1965). Primary Love and Psychoanalytic Technique. New York: Liverwright.

    Berger, D. (1985). On the practice of medicine and on the culture and customs in Japan. Tokai Journal of Experimenal and Clinical Medicine, 10: 637-645.

    Berger, D. (1993). Suicide evaluation in medical patients: a pilot study. General Hospital Psychiatry, 15: 75-81.

    Berger. D., Ono, Y., Kumano, H., & Suematsu, H. (1994). The Japanese concept of interdependency (letter). American Journal of Psychiatry, 51: 628-629.

    Christopher, R. C. (1983). The Japanese Mind. Tokyo: Charles E. Tuttle.

    De Vos, G. A. (1968). Suicide in cross-cultural perspective. In Suicidal Behaviors: Diagnosis and Management, ed. H. L. P. Resnik, pp. 125-146. Boston: Little Brown.

    Doi, T. (1973). The Anatomy of Dependence. Tokyo: Kodansha International. Domino, J., & Takahashi, Y. (1991). Attitudes towards suicide in Japanese and American medical students. Suicide and Life-Threatening Behavior, 21: 345-359.

    Fukushima, A. (1984). Shinju. In Kodansha's Comprehensive Dictionary of Psychiatry, ed. N. Shinfuku, pp. 430-431. Tokyo: Kodansha (in Japanese)

    Fuso T. (1985). Suicide and Culture. Tokyo: Shincho-sha. (in Japanese)

    Inamura, H. (1977). Suicidology: Treatment and Prevention. Tokyo: Tokyo University Press. (in Japanese)

    Inamura, H. (1993). Recent characteristics, risk estimations and preventive methods of shinju (double suicide) in Japan. Archives of Psychiatric Diagnostics and Clinical Evaluation, 4: 173-184 (in Japanese)

    Japan Times (1985) January 31.

    Mahler, M. (1972). Rapprochment subphase of the separation individuation process. Psychoanalytic Quarterly, 41: 487-506.

    Marzuk, P. M., Tardiff, K., & Hirsch, C. S. (1992). The epidemiology of murder-suicide. Journal of the American Medical Association, 267: 3179-3187.

    National Police Agency (1993). National Police Agency's 1990 Annual Report, Tokyo: Printing Section of Ministry of Finance. (in Japanese)

    Ohara, K. (1963). Characteristics of suicide in Japan: Especially on parent-child double suicide. American Journal of Psychiatry, 120: 382-385.

    Ohara, K. (1965). Suicide in Japan: Understanding Solitude and Anxiety. Tokyo: Seishin-shobo. (in Japanese)

    Ohara, K. (1985). Shinju. In Encyclopedia of Psychiatry (Revised Edition), ed. M. Kato, H. Hozaki, Y. Kasahara, T. Miyamoto, & K. Okonogi, pp.330-331. Tokyo: Kohbundo. (in Japanese)

    Ono, Y. (1991). Anxiety and avoidant personality disorder. presented at the 2nd International Congress on The Disorders Of the Personality, Oslo, 1991.

    Sugiyama-Lebra, T. (1976). Japanese Patterns of Behavior. Honolulu: University of Hawaii Press.

    Takahashi, Y. (1989). Suicidal Asian patients: Recommendations for treatment. Suicide and Life-Threatening Behavior. 19: 305-313.

    Takahashi, Y. (1992). Clinical Evaluation of Suicide Risk and Crisis Intervention. Tokyo: Kongo-shuppan. (in Japanese)

    Takahashi, Y. (1993a). Suicide Prevention in Japan. In Suicidology: Essays in Honor of Edwin S. Shneidman, ed A. A. Leenaars, pp. 324-334. Northvale: Jason Aronson.

    Takahashi, Y. (1993b). Depression and suicide. In Affective Disorders: Perspectives on Basic Research and Clinical Practice, ed. T. Kariya, & M. Nakagawara, pp.85-98. New York: Brunner/Mazel.

  • Return to Medical Papers