From Self to Nonself: Death Anxiety as Mediator of the Relationship Between Renunciation of Desire and Mental Health


 In the present paper, we report two studies examining the relationships among renunciation of desires, death anxiety, and mental health. In the first study, we constructed the Desire Questionnaire (DQ), which measures success in renouncing certain desire. The DQ has satisfactory psychometric properties. In the second study, 507 adults from a Chinese society ranging in age from 17 to 84 years (M = 35.58, SD = 14.76) completed the DQ, the Death Anxiety Scale (DAS), and the Chinese Health Questionnaire (CHQ), which measures mental health and the presence of psychiatric symptoms. As predicted, DQ had significant negative correlations with both DAS (p < .05) and CHQ (p < .01). DAS had a significant positive correlation with CHQ (p < .01). In the linear mediation model, DAS was found to partially mediate the relationship between DQ and CHQ (\(\beta\) = .21, p < .001). These results suggest that the negative effect of poor control of desires on mental health can be partially accounted for by death anxiety. These findings imply that training in eliminating desires can be a successful strategy to improve mental health. These results also support the Nonself Theory as a theory of death anxiety and show its relevance to the relationship between nonself and mental health.

renounce desires by performing negative and positive duties 6 . This is why these precepts were chosen as subcategories for the development of Desire Questionnaire (DQ) in the present paper.
The Association Between Nonself, Desire, And Death Anxiety: From Self To Nonself There is no doubt that death is the paramount threat to the self or to one's identity. Our awareness that we will eventually die is a tacit recognition that life is fragile and that the self will someday disappear. The death anxiety causes our terror that affects everyone by pushing one to seek meaning in death [19][20][21] . The aim of this quest for meaning is to overcome death anxiety. It is important to understand that this meaning creation provides us with information about the presence of stable patterns and coherence of expectations and behaviors in the environment 22 , and it helps us cope with life's adversities [23][24][25] .
In our opinion the most well-reasoned theory attempting to explain what death means to people is the Terror Management Theory (TMT) 19,26,27 . The core postulate in the theory is that we draw upon our self-esteem, a feeling of self-importance or signi cance in the world, to overcome the fear of death. Selfesteem in TMT is a culturally grounded concept in the sense that we enhance it by convincing ourselves that our behavior meets the norms dictated by our culture 28 and that we are making important contributions to this culture, that the life we live is valuable and meaningful. TMT maintains that what drives human beings more than anything else is our need to conquer death anxiety. It is the role of selfesteem to activate our psychological defense mechanisms against death 29 .
For more than 2,500 years Buddhists have practiced a distinctive approach to coping with death anxiety, with the ultimate aim of overcoming and eliminating it 6,7,30 . It is of academic interest to explore the possibility that Buddhism provides an alternative perspective on the meaning of death and how to manage death anxiety. Thus, a major aim of this paper is to present a Buddhist perspective on the meaning of death. Buddhism shares with TMT the proposition that the awareness of death leads to anxiety. But Buddhism holds a different view than TMT on self-esteem and its role in dealing with death anxiety 30,31 . In contrast to self-esteem, nonself is a state in which a person has a sense of egolessness, which re ects awareness and realization of nonself 32,33 . Buddhists have long believed that nonself conquers death anxiety 31,34 .
The threat of death is always present and occurs in a variety of forms. If we succeed in maintaining or boosting self-esteem for the purpose of dealing with death anxiety, a new threat may supplant the previous one. We need countless action to boost and keep self so it can cope with death anxiety. TMT states that the prospect of death threatens the self-esteem. It is logic to infer if there is no self or identity, death can't threaten the nonself. Thus, there is no death anxiety. For this reason, based on the Nonself Theory 6 , the state of nonself helps one cope with death anxiety. Patients in hospice care who volunteered to practice Buddhist wisdom were found to have a decreased fear of death 35 . The same was found for a bone cancer patient 36 .
Accordingly, it is logical to assume that renunciation of desires approaching the nonself state is associated with a reduction in death anxiety. This was the second hypothesis tested in the present research.
The Role Of Death Anxiety Between Desire And Mental Health Death anxiety has been found to have a signi cant negative effect on mental health 37 . Death anxiety was also found to negatively in uence the assessment of one's mental health in a sample of deployed soldiers 38 and to be a strong predictor of mental disorders 39 . Death anxiety has been shown to have signi cant positive correlations with depression and anxiety symptoms 40 and post-traumatic stress disorder 41 . Taken together, these studies suggest that both desire and death anxiety are associated with mental health and that death anxiety can be overcome by minimizing self to attain the state of nonself.
Renunciation of personal desires, for example, is one of the best strategies to deal with inevitable death anxiety.
In Buddhism, self is considered to be the primary cause of death anxiety 7,30,31 . Recall that according to TMT, the death cannot threaten self-esteem if there is no self or identity, because there is no self-esteem that death can coerce. Based on the Nonself Theory 6 , there would be no unhappiness and no anxiety, but rather the greatest equanimity, contentment, happiness, and with no room to suffer the pain caused by death and life's adversities.
Accordingly, based on the assumption that death anxiety develops as the renunciation of desires is channelled by cultural, educational, and environmental in uences 6,7 , we predicted that renunciation of desire would account for the effects of death anxiety on mental health, but not vice versa. This study also predicted that individual differences in renunciation of desire would in uence the degree to which death anxiety affects mental health. In other words, we predicted that death anxiety would mediate some of the effects of renouncing desire on mental health. We tested these predictions using a mediation model (see Figure 1.), for which it was assumed that the effect of the independent variable (renunciation of desires) on the dependent variable (mental health) is mediated by the action of the mediating variable (death anxiety). In path terms, renunciation of desires reduces death anxiety which in turn improves mental health. Con rmation of the model would support this hypothesis. Finally, we predicted that the demographic characteristics age, gender, and education would be associated with measures of death anxiety 26 and mental health 42 , and renunciation of desires. Thus, we included age, gender and education as covariates in our analyses.

Overview Of The Studies
In the present paper we report two studies. The aim of the rst study was to develop the Desire Questionnaire (DQ), a measure of the renunciation of desires with ve subscales: No Killing, No Harmful Speech, No Sexual Misconduct, No Intoxicants, and No Stealing. Next, we assessed the reliability and construct validity of the DQ. The aims of the second study were to provide criterion validation for the DQ by demonstrating signi cant relationships among renunciation of desires, death anxiety, and mental health, and to test the primary hypothesis that a signi cant positive effect of renunciation of desires on mental health is mediated by reduced death anxiety.
Both studies were conducted in single sessions. Two independent assistants checked data records against responses on the questionnaires to eradicate possible recording errors. Both studies were approved by the National Cheng Kung University Human Research Ethics Committee. All participants gave informed consent. Prior to identi cation of the subscales, 243 adults lled out the 29-item scale ( Table 1). The responses were subjected to exploratory principal components factor analysis with an oblique rotation for the purpose of determining the most appropriate solution, and thus the number of factors, and then which items should be retained with this solution. An oblique rotation was used because the factors were hypothesized to be correlated. The following criteria were used to select the best model: (a) minimum eigenvalue of 1, (b) minimum factor loading of .30 and (c) good factor interpretability.
The best model was found to be the ve-factor model (see Table 2). For this model, Bartlett's test of sphericity indicated strong correlations between the items in each subscale, = 3491.10, p < .001. The Kaiser-Meyer-Olkin value was 0.79, which is greater than 0.60, indicating that these factors are distinct and reliable 46 . Cronbach's was .84 for the total scale. These results justi ed the conduct of a con rmatory factor analysis.  With a cut off of .7, all ve items loaded highly (> 0.72) on the rst factor, four of the ve items loaded highly (> 0.75) on the second factor, two of the three items loaded highly (> 0.86) on the third factor, two of the three items loaded highly (>.95) on the fourth factor, and one of the four items loaded highly (= .98) on the fth factor. Con rmatory factor analysis A con rmatory factor analysis (CFA) was performed on the ve-factor model using data from a new sample of 248 adults using AMOS 22.0 software (Table 3). Maximum likelihood was used to estimate the parameters of the model. The null hypothesis that the data would t the model perfectly was rejected, 160 = 385.25, p < .001. We used several goodness-of-t indices to examine how accurately the present data t the ve-factor model ( Table 4): The /df ratio was less than 3, which is considered moderate (see Wheaton, 1987); the parsimonious normed t index (PNFI) of 0.73 also was moderate ( > 0.5; see James et al., 1982); the comparative t index (CFI) was high (0.92); the root mean square error of approximation (RMSEA) was less than 0.08, indicating a good t 47 ; and the value for the Akaike information criterion (AIC) was substantially smaller for the independence model than for the proposed model. The t with the ve-factor model was satisfactory. Internal consistency (Cronbach's ) was .83 for the total scale.  Note: AIC = Akaike information criterion; CFI = comparative t index; PNFI = parsimonious normed t index; RMSEA = root mean square error of approximation. ***p < .001.
The DQ was validated by both the exploratory factor analysis and the con rmatory factor analysis. The items were the same for each analysis, but the samples were different. As the analyses provided evidence for adequate psychometric validity of the DQ, we used it in Study 2.

Study 2
The aim of the second study was to test for possible mediation by death anxiety of a possible relationship between renunciation of desires and mental health. There were three speci c hypotheses: Hypothesis 1: Renunciation of desires is negatively associated with death anxiety and positively associated with mental health.
Hypothesis 2: Death anxiety is negatively associated with mental health.
Hypothesis 3: (Low) death anxiety partially mediates a positive effect of renouncing desires on mental health.

Method
Participants A convenience sample of 507 adults (166 males and 335 females) ranging in age from 18 to 85 years was recruited. Six were removed from the sample for missing data, leaving a nal sample of 501.

Measures
The demographic variables were age, gender, and education, and religion. The education levels were elementary school, junior high school, high school, college, graduate school, and postgraduate. The religions were Christianity, Buddhism, traditional Chinese religions, and atheism.

Desire Questionnaire
We used the Desire Questionnaire developed in Study 1.

Death Anxiety Scale
We used the Death Anxiety Scale (DAS), developed by Templer 48 and widely employed to measure death anxiety 49 . We used the Chinese translation of the DAS, which has good reliability and validity 50 . The DAS has 15 items rated as "true" (1) or "false" (0), with a higher score indicating greater death anxiety.
However, some researchers suggested to use a ve-point Likert scale with response alternatives of least anxiety (1) and most (5) to re ect real situation 51 . For this reason, we adopted the ve-point Likert scale. We performed an extra analysis to demonstrate a satisfactory internal consistency (Cronbach's α) of .76 for the DAS.
Chinese Health Questionnaire (CHQ) As our measure of mental health, we used the 12-item Chinese Health Questionnaire (CHQ), which was developed as a screening measure for minor psychiatric illness 52 . The CHQ is a modi cation based on the concepts and structure of the General Health Questionnaire (GHQ) 53 and re ecting cultural differences between Chinese and Western culture 54 . Participants respond to the 12 items on a 4-point Likert scale with response alternatives of "not at all" and "same as usual" (both scored as 0), and "more than usual" and "a lot more than usual" (both scored as 1), with a higher score indicating more severe psychiatric symptoms. The CHQ demonstrated an internal consistency (Cronbach's α) of .79 52 .

Procedure
A research assistant informed participants of the nature of the research. Participants received a booklet including the demographic items, followed (in order) by the DQ, CHQ, and DAS.

Data analysis
Means, standard deviations, and percentages were calculated for the demographic variables. Pearson correlation coe cients were calculated to measure the pairwise linear relationships between scores on the DS, DAS and CHQ. To test the mediation, we used stepwise multiple regression and Sobel test.

Descriptive data
Descriptive data for Study 2 are presented in Table 5. Note. For continuous variables, the mean is followed by the standard deviation in parentheses.

Correlational analyses
The partial correlations between the Desire Questionnaire, The Death Anxiety Scale, and the Chines Health Questionnaire, controlling for age, gender, and education, are shown in Table 6. As expected, the DQ had a signi cant negative correlation with the DAS and a signi cant positive correlation with the CHQ. The DAS had a signi cant negative correlation with the CHQ. These results support Hypotheses 1 and 2. Table 6. Partial correlations between Desire Questionnaire, Death Anxiety Scale, and Chinese Health Questionnaire scores, controlling for age, gender, and education.

Mediation Analyses
We conducted a linear mediation analysis to determine whether death anxiety acted as a partial mediator of the effect of renouncing desires on mental health. Figure 1 presents a schematic of the model and Table 7 summarizes the results of each step.
Step 1 estimated the "c" path by regressing CHQ scores on DQ scores.
Step 2 estimated the "a" path by regressing DAS scores on DQ scores. Step 3 estimated the "b" path by regressing CHQ scores on DAS scores controlling for DQ scores.
Step 4 estimated the "c'" path, which distinguishes between partial and complete mediation based on whether the demonstrated effects of renunciation of desire on mental health are reduced to zero when DAS scores are included as a predictor in the equation.
As seen in Table 7, the relationship between DQ and CHQ scores is partially mediated by DAS. These results support our Hypothesis 3. Table 7. Summary of analysis of the effect of renunciation of desire (DQ scores) on mental health (CHQ scores) mediated by death anxiety (DAS scores).

Predictor
Mediator Step 1 Path c Step 2 Path a Step 3 Path b Step

General Discussion
The two studies presented in this article were designed to develop the DQ as a measure of renunciation of desire and use it to test the relationships of renunciation of desire with death anxiety (measured by DAS) and mental health (measured by CHQ). Study 1 provided evidence for the psychometric adequacy and validity of the DQ by exploratory factor analysis and con rmatory factor analysis. The DQ was shown to be a reliable and valid instrument for use with the general adult population in Taiwan. In Study 2, the signi cant negative correlation of DQ with DAS and the signi cant positive correlation of DQ with CHQ demonstrated the criterion validity of DQ as measuring a construct (renunciation of desire) that is relevant to mental health. The results of these analyses demonstrate satisfactory construct and criterion validity for the DQ.
The second study was the rst to investigate the relationships among renunciation of desires, death anxiety, and mental health. As expected, renunciation of desires had a positive association with mental health and a negative association with death anxiety. Our nding of a signi cant negative correlation of death anxiety with mental health is consistent with previous studies [37][38][39] .
Death anxiety was found to partially mediate the relation between renunciation of desires and mental health. Consistent with our prediction, these ndings suggest that the effect of renouncing desires on mental health can be accounted for at least partially by the effects of reduced death anxiety caused by renouncing desires. Though the role of death anxiety in mediating the relationship between renunciation of desires and mental health was rst discovered in the present study, this nding needs to be replicated before de nitive conclusions can be drawn.
The present study supports the Nonself Theory 6 as a theory of death anxiety and shows its relevance to nonself and mental health that Buddhism provides an alternative perspective on the meaning of death and the Buddhist tenet that self is the main cause of death anxiety for more than 2,500 years 7,31 . Undoubtedly, everyone experiences death anxiety at some time in life, but it can be reduced or overcome by moving self towards the nonself state by severing all ties to the world, renouncing unnecessary personal desires 30 , and practicing the Buddhist wisdom. According to Buddhism, death serves as both a continuous reminder that life is nite; recognition of this directly leads to the search for the meaning of life, conducted mainly by cultivating self and moving it towards the nonself state as a way to overcome the death anxiety. Our results demonstrate that although people indeed perceive death as a threat, it is not necessary as the time of death approaches to invoke a defensive response by moving mortality away from our focal awareness or appealing to our self-esteem, as described by TMT 26-28 . But avoiding this requires that one adopt a different explanation for the meaning of death. Our research can help in this regard by providing evidence for an alternative framework, based on Buddhism, for understanding at least some of the effects of the salience of mortality.

Limitations and suggestions for future research
The DQ is still in its infancy and our results suggest a number of opportunities for further investigation. The subjects for its validation were recruited from Chinese societies, and we suggest that further research on the DQ be conducted with more diverse samples so as to generalize the ndings cross-culturally. cultures. Based on the Nonself Theory 6 , the wisdom of Buddhism provides a sophisticated framework for explaining a possible mechanism for this phenomenon of renunciation of desires, as well as its bene ts for mental health, which is an important topic for further empirical investigation. Even more so, investigation of how development of the avoidance of desire-driven pleasures promotes the nonself should be a next step in this line of research. Future research could also pro tably pay more attention to clinical application of this avoidance, for example, on whether the avoidance of desire-driven pleasure is related to the successful application of strategies for coping with events that lead to an increase in death anxiety. Some might think our results support TMT. For example, based on the TMT, by following the ve precepts, one might feel that he/she is living up to the standards of value inherent in his/her cultural worldview and therefore resulted in higher self-esteem. However, based on the Buddhism, the purpose of obeying the ve precepts is not to boost self-esteem. More studies are needed to approve or disapprove this issue. Furthermore, future research should consider the effect of social desirability response. For example, one might increase their prosocial response while lling out the questionnaire. Finally, precisely how reduction of desires promotes good mental health is still not clear, making this another important topic for future studies.

Page 14/18
The purpose of the two studies reported in this paper was to demonstrate the role of renunciation of desire in the reduction of death anxiety and promotion of mental health. As mentioned earlier, based on Nonself Theory 6 , a reduction in the level of desire linked to the nonself state leads to reduced death anxiety and improved mental health. These results also support the Nonself Theory as a theory of death anxiety and show its relevance to the relationship between nonself and mental health. Although there have been many studies of Buddhism, which has been widely practiced for more than 2500 years, the majority of them have been focused on meditation and its effects, such as increased emotional stability, heightened positive emotion, and improved attention. There have been few empirical studies or theories directly relating Buddhist teachings to lifestyle issues. The present research is among the rst attempts to study Buddhist teachings in relation to death anxiety in this di cult time of pandemic. Our hope is that this research has helped to ll these gaps, because it suggests that Buddhism provides a reliable and useful way to cope with life's adversities, which contribute to a variety of mental health problems. Further research on these attributes could open signi cant new avenues for mental health research and unravel the secrets of why Buddhism has lasted for thousands of years.

Declarations
Ethical Approval The two studies and their relevant details were approved by the Human Research Ethics Committee of National Cheng Kung University in Taiwan. All studies were performed in accordance with relevant guidelines and regulations. The volunteer participants were informed of the nature of the research and given assurances of con dentiality; the researchers then obtained their informed consent.
Informed Consent Informed consent was obtained from all individual participants included in the study.
Con ict of Interest The authors have no con icts of interest to declare that are relevant to the content of this article.
Data availability The datasets analyzed during the current study are available from the corresponding author on reasonable request.