CHAPTER 5

 

DISCUSSION

 

Evaluation of Hypotheses and Discussion of Results

      At the beginning of the methodology chapter it was explained that the hypotheses of this investigation will be stated in terms of research hypotheses instead of null hypotheses, as supported by Agresti and Finlay (1986). This convention will be followed in this discussion chapter, although it is understood that, technically speaking, the statistical tests used to analyze the data of this experiment can only provide evidence for rejecting or not rejecting the null hypotheses, and cannot directly evaluate the validity of the alternative, or research, hypotheses.

      Hypothesis 1: The maha mantra group will show significantly decreased stress from pretest to posttest compared with the alternate mantra group and with the control group at the .05 level of significance.

            This hypothesis was confirmed by the statistical analysis of the previous chapter. The maha mantra group decreased stress levels more than both of the other two groups, and effect sizes for group status on stress from pretest to posttest, 33% for partiallized values and 25% for observed values, are large according to Rosenthal (1997).

            According to Vedic theory, chanting the maha mantra causes a person to decrease stress and anxiety (Dasa, 1994). Thus, the results of this study, in regards to hypothesis 1, support Vedic guna theory.

            In behavioral terms, chanting the hare krsna maha mantra may have served as an effective positive replacement thought for stressful thoughts. According to Spiegler and Guevremont (1993), a common behavioral technique to reduce stress is to disrupt distressing thoughts and replace them with a competing, positive cognition. This stress-reduction technique has been successful in many environments, for diverse purposes and with several populations, including traumatology (Berk, 1998; Foa, 1997), rape victims (Foa, 1997), self-esteem development (Joiner & Sawyer, 1992), and post-traumatic stress disorder (Miller & Basoglu, 1991). According to the Vedas, the sound vibration of the hare krsna maha mantra can affect the mind to abandon stressful thoughts. Therefore, the results of this experiment with regard to hypothesis 1 are consistent with explanations based in behavioral and guna theory.

            Hypothesis 2: The maha mantra group will show significantly decreased depression from pretest to posttest compared with the alternate mantra group and with the control group at the .05 level of significance.

            This hypothesis is supported by the statistical analysis in the previous chapter. The maha mantra group decreased depression levels more than both of the other two groups, and effect sizes for group status on depression from pretest to posttest, 21% for partiallized values and 25% for observed values, are large according to Rosenthal (1997), and represent a strong association between the variables.

            Srimad-Bhagavatam (Prabhupada, 1976) hypothesizes that the hare krsna maha mantra affects the mind to abandon depressing thoughts. Results of this study, therefore, support guna theory in relation to Hypothesis 2.

            As described above in relation to stress, behavioral techniques such as thought stopping and thought replacement are also effectively used to treat depression (e.g., Albano and Morris, 1998; Freeston, Ladouceur, Provencher, and Blais, 1995). In such approaches, persons who habitually think depressing thoughts, such as “I’m worthless”, train themselves to replace these thoughts with thoughts that are rewarding and supportive of self-esteem. From a Vedic viewpoint, the sound vibration of the hare krsna maha mantra provides fulfillment for the mind that takes the place of depressing thoughts. Therefore, the results of this study in regards to depression are consistent with guna and behavioral theory.

            Hypothesis 3: The maha mantra group will show significantly increased sattva from pretest to posttest compared with the alternate mantra group and with the control group at the .05 level of significance.

            This hypothesis is supported by the results of this study. The maha mantra group increased its sattva level significantly more than the alternate mantra group and more than the control group. Effect sizes for group status on sattva from pretest to posttest were .23 for partiallized scores and .18 for observed scores. According to Rosenthal (1997) this indicates a medium to large effect size and a moderate to strong association between the variables. The results of this experiment in connection with hypothesis 3 can be understood in relation to the Vedic assertion that the maha mantra is predominantly in sattva guna. The alternate mantra, according to the Vedas, is not in sattva guna.

            Hypothesis 4: The maha mantra group will show significantly decreased rajas from pretest to posttest compared with the alternate mantra group and with the control group at the .05 level of significance.

            This hypothesis is not supported by the statistical analysis of the previous chapter. No group showed a statistically significant change in rajas scores from pretest to posttest compared with either of the other groups.

            From Vedic theory it can be understood that the formulation of hypothesis 4 may be faulty. As described above, the Vedas suggest that chanting the hare krsna maha mantra may transform rajasic and tamasic qualities into sattvic qualities. However, the Vedas also indicate that chanting the maha mantra will transform tamasic qualities into rajasic qualities. That is, Vedic teachings describe a progression from tamas to sattva, with rajas in the middle. Therefore, while some rajasic qualities are replaced by sattvic qualities, some tamasic qualities are transformed into rajasic attributes, and therefore the overall level of rajas is maintained. By continued chanting of the hare krsna maha mantra, according to the Srimad-Bhagavatam (Prabhupada, 1976), all rajas and tamas will be replaced by sattva. This study, however, entailed chanting the maha mantra for only one month, and thus a possible explanation could be that the full effects of the mantra were not allowed to take place, resulting in tamasic qualities transforming into rajasic qualities without sattva becoming prominent. For instance, the indolence of tamas may convert to the frenetic activity of rajas, before ultimately transforming into the calm determination of sattva.

            Hypothesis 5: The maha mantra group will show significantly decreased tamas from pretest to posttest compared with the alternate mantra group and with the control group at the .05 level of significance.

            This hypothesis is partially supported by the results of this experiment. The maha mantra group decreased pretest-posttest tamas scores significantly more than the control group, though not significantly more than the alternate mantra group. Effect sizes for group status on pretest-posttest tamas scores were 21% for partiallized scores and 18% for observed scores. Effects of this magnitude are medium to large, indicating a moderate to strong association between the variables (Rosenthal, 1997). Differences between the alternate and control groups for pretest-posttest tamas values were not statistically significant.

            Tamas is characterized by a lack of determination, regulation and discipline. Persons predominated by tamas guna have difficulty fulfilling their responsibilities and gaining control over their life (Prabhupada, 1976). It may be conjectured that the regulated activity of chanting three japa rounds every day caused members of the alternate mantra group to reduce their level of tamas. That is, although the alternate mantra is not composed of sattvic vibrations, the disciplined act of chanting every day had the effect of increasing sattva sufficiently for the difference between the two chanting groups to not be statistically significant.

 

Secondary Hypotheses

 

            Hypothesis 6: The maha mantra group will show significantly decreased stress from pretest to followup compared with the alternate mantra group and with the control group at the .05 level of significance, though some diminishing of effects compared with pretest-posttest is expected.

            Hypothesis 7: The maha mantra group will show significantly decreased depression from pretest to followup compared with the alternate mantra group and with the control group at the .05 significance level, though some diminishing of effects compared with pretest-posttest is expected.

            Hypothesis 8: The maha mantra group will show significantly increased sattva from pretest to followup compared with the alternate mantra group and with the control group at the .05 significance level, though some diminishing of effects compared with pretest-posttest is expected.

            Hypothesis 9: The maha mantra group will show significantly decreased rajas from pretest to followup compared with the alternate mantra group and with the control group at the .05 level of significance, though some diminishing of effects compared with pretest-posttest is expected.

            Hypothesis 10: The maha mantra group will show significantly decreased tamas from pretest to followup compared with the alternate mantra group and with the control group at the .05 significance level, though some diminishing of effects compared with pretest-posttest is expected.

            None of these secondary hypotheses were fully confirmed by the results of this study, though hypothesis 7 and hypothesis 10 were partially confirmed. For pretest-followup depression and tamas scores the maha mantra group showed a statistically significant decrease in depression and tamas compared with the control group, though not compared with the alternate mantra group. In none of the pretest-followup comparisons did the alternate mantra group differ significantly from the control group. For tamas only the observed difference ANCOVA produced significant effects for group status, with group status accounting for 11% of the variance, which according to Rosenthal (1997) represents a moderate association between the variables and a medium effect size. For pretest-followup depression values effect sizes were 12% for partiallized scores and 20% for observed scores, representing a medium to large effect size (Rosenthal).

            Vedic theory suggests that when one withdraws from the association of the hare krsna maha mantra, or any sattvic influence, the effects of sattva will diminish, and effects of rajas and tamas, such as stress, lust, greed and depression, will reappear. Therefore the Vedas recommend that one remain constantly in sattvic association. Pretest-followup results of this experiment indicate that the effects of one month of chanting the maha mantra did diminish after a one month followup period in which no chanting was performed. However, as evidenced by depression and tamas scores, some effects did remain, though the effect sizes for these variables were less than the effects for the pretest-posttest period, as predicted by Vedic theory. It should be noted that the Vedas claim that depression is a symptom of tamas, and therefore the two pretest-followup variables that retained significant effects are theorized to be related. Providing evidence for this relationship is the significant correlation (r= .51; p= .000) between pretest-followup depression scores and pretest-followup tamas scores.

            From a behavioral perspective it can be understood that effects diminish when positive thoughts are discontinued. Thus, for example, it can be conjectured that stressful thoughts again became prominent for subjects when they ceased to chant the hare krsna maha mantra.

            Hypothesis 11: Gender will have no effect on any of the dependent variables at a .05 significance level.

            When ANCOVAs resulted in a significant group status effect, gender did not have a significant effect on the dependent variable. Therefore hypothesis 11 was supported. When group status had a significant effect on the dependent variable, the non-significant effect sizes associated with gender never had an effect size greater than 1%. Therefore it can be concluded that gender did not substantially effect the dependent variables, though including gender as a controlled variable in this study successfully removed the effects of gender as an unknown and uncontrolled source of variation, thus adding to the explanatory strength of the group status results. The correlations of covariates section of the results and data analysis chapter shows that for the control group gender did have 5 significant correlations, indicating that gender and the gunas may not be completely unrelated. In psychometric research on the VPI (Wolf, 1998) it was also found that gender was not completely neutral in relation to the gunas, with higher sattva scores for females and higher tamas scores for males.

            Hypothesis 12: Age will have no effect on any of the dependent variables at a .05 significance level.

            Age had a significant p value for pretest-posttest stress and depression scores, though the largest effect size for age among all the ANCOVA analyses was 6%, which was for pretest-posttest partiallized difference scores for stress. Pearson r for age correlated with pretest-posttest stress difference scores was  -.26 (r2= .07), and r for age correlated with pretest-posttest depression difference scores was  -.25 (r2= .06). These correlations, along with the significant p values for the F tests mentioned above, demonstrate that older participants decreased their stress and depression from pretest to posttest more than younger participants. As described in the correlations of covariates section of the previous chapter, age did not significantly correlate with any of the pretest, posttest or followup scores for any of the three groups for any of the dependent variables. Overall, ANCOVAs indicate that age did have some effect on some of the variables, contrary to hypothesis 12, and controlling for age in the statistical analysis increases the explanatory strength of the group status analyses.

            Hypothesis 13: Chanting frequency for the maha mantra group will be positively correlated, at the .05 significance level, with dependent variables in the same direction that the maha mantra group is related with the dependent variables. That is, increased chanting frequency for the maha mantra group will correlate with increased sattva and decreased stress, depression, rajas and tamas. Chanting frequency for the alternate mantra group will have no correlation at the .05 significance level with any of the dependent variables.

            Out of the six analyses resulting in significant p values for group status effects, chanting frequency had significant p values for three of the analyses. These three were pretest-posttest depression, pretest-posttest tamas, and pretest-followup depression (for partiallized scores only). For the observed difference pretest-posttest depression ANCOVA, chanting frequency had an effect size of 17%, which was the highest of the effect sizes for chanting frequency. For some analyses chanting frequency had a medium or even large effect size, but the p value for the F test was not significant. For instance, for partiallized pretest-posttest stress scores chanting frequency had an effect size of 8%, and the p value was .581, for observed differences pretest-followup depression scores chanting frequency had an effect size of 16%, and the p value was .106, for observed differences pretest-posttest sattva difference scores chanting frequency had an effect size of 7%, with a p value of .061, for partiallized pretest-followup differences for sattva chanting frequency had an effect size of 11% and a p value of .380, and for observed differences pretest-followup tamas scores chanting frequency had an effect size of 12% and a p value of .124. These noteworthy effects with non-significant p values suggest that a study with a larger sample size, and thus greater power, might detect more significant differences with chanting frequency in relation to the dependent variables.

            Also, in some cases chanting frequency had a significant p value, though the effects of group status were not significant. For example, the observed differences pretest-followup stress analysis resulted in an effect size of 16% for chanting frequency, with a p value of .004, and the p value for chanting frequency in the pretest-followup partiallized differences rajas scores was .048. When group status did not have a significant effect on the dependent variable, thorough analysis of effects of chanting frequency were not particularly relevant, even if chanting frequency had a significant p value, because it is not especially consequential to discuss effects of chanting frequency if the overall effects of chanting were found to be nonsignificant.

            Table 11 shows that chanting frequency significantly correlated with several of the alternate group dependent variable scores, and with none of the maha mantra group dependent variables. This means that in the ANCOVAs with a significant chanting frequency p value, the effects were more prominent for the alternate group than for the maha mantra group, contrary to hypothesis 13. Apparently there was some placebo effect in this study, as dependent variables were effected by changes in chanting frequency of the alternate mantra group, although, as described in the section in this chapter entitled major hypotheses, in no case did the alternate and control groups significantly differ in pretest-posttest or pretest-followup differences in dependent variables.

            According to Vedic theory, increased chanting of the hare krsna maha mantra will correspond to increased effects of the mantra (Prabhupada, 1976). In this experiment all subjects were instructed to chant three rounds of japa every day, and the actual range of chanting frequency for the maha mantra group was 2.46 to 3.00. It is suggested that this range is not adequate to obtain a true picture of the effects of chanting frequency. A range of 1 round per day to 30 rounds per day, with groups instructed to chant different numbers of rounds per day than other groups,  would be more appropriate for a study of chanting frequency effects.

            Also, it should be noted that although all of the groups began with equal numbers of subjects, the maha mantra group retained 24 members, while the other two groups each retained 19 members. Cook and Campbell (1979) note that when the treatment group keeps more members than other groups, it may be regarded as evidence of the hypothesis that the treatment is effective. That is, the comparative group dropout results of this study suggest that participants derived more value from the maha mantra, and therefore they were more likely to remain in the maha mantra group compared with the other two groups.

 

Limitations of the Study

 

            Rosenthal (1997) explains that generalizability of a study is the degree to which similar findings will be observed in a different context or setting. As this study did not employ random sampling, the sampling method of the experiment does not allow statistical generalization to any population other than the 62 subjects who participated in the study. However, generalizability of a study also contains a qualitative component, requiring the researcher or practitioner to apply his/her expertise in an area to determine the degree of similarity of contexts. Therefore, a practitioner or researcher desiring to assess this study for applicability can evaluate the similarity of the participants and setting of this study to the context in which s/he wishes to apply the techniques and findings. Generalizability of course is enhanced or reduced by replication studies, none of which have yet been performed for the hare krsna maha mantra.

            Another deficiency of this study in regards to generalizability is reactive effects of the pretest, which was not resolved with the experimental design utilized in this investigation. The effects of the pretest may have affected subjects in such a way that they were no longer representative of any larger population, and thus the effects of the interventions may only apply to those who completed such a pretest, and not to others (Royse, 1995).

            Bracht and Glass (1968) explain that external validity, or generalizability, of a design can be divided into two categories- population validity and ecological validity. Population validity refers to the ability to make inferential leaps to a larger population, and the major threat in this experiment to population validity is lack of random sampling, and another threat is reactivity to pretests. Bracht and Glass describe novelty effects as an additional threat to population validity, and this may be a defect in the representativeness of this experiment. Specifically, the effects of chanting may have to some degree been caused by the novelty of the experience. If this is the case, then after the novelty effect wears off, the effects of the chanting process can be expected to diminish. Thus, the results may not apply to a general population, but only to the studied population in which the novelty effects were active.

            Ecological validity refers to the degree to which the experimental effect is independent of the experimental environment. In this study, the Hawthorne effect, referring to the subjects’ knowledge that they were participating in an experiment, may have been active, thus decreasing the generalizability of the findings (Bracht and Glass, 1968).

            Bracht and Glass (1968) also describe experimenter effects as a threat to ecological validity. In this study the researchers were aware of the group status of each participant, and therefore experimenter effects must be considered. That is, the researchers were part of the setting of the experiment, and it is possible that different results would be achieved with different researchers, and this was a shortcoming in the ecological validity of the study. Specifically, the researchers may have behaved one way towards members of one group, and differently towards members of another group, thereby causing effects that are not directly attributable to any of the independent variables in the design. The design of this study attempted to control for experimenter effects by standardizing the training of each researcher to ensure consistent presentation to all members of each group of subjects.

            While acknowledging the threat of experimenter effects to ecological validity, overemphasis on the effectiveness of double-blind studies must also be avoided. In the field of medical research, for instance, there is evidence that a physician’s beliefs can alter the results of double-blind studies. For example, Dossey (1993) describes double-blind studies of the use of vitamin E in treating angina pectoris. In a double-blind study involving an enthusiastic doctor who believed in vitamin E the treatment was found to be more effective than a placebo, while two double-blind studies conducted by skeptics showed no effect. Many analyses of drug research development show similar results, with enthusiasts arriving at opposite conclusions to skeptics, both using double-blind studies. Regarding the drug meprobamate, Dossey writes “Overall, therefore, three of the four meprobamate studies suggested strongly that the effectiveness of the drug over the placebo was correlated with the physician’s attitudes and beliefs toward it, and that the beliefs of the prescribing physician can somehow penetrate the double-blind conditions of the experiment and shape the action of the drug” (p. 136). After reviewing many double-blind studies in the fields of medicine, neuropsychopharmacology, and psychiatry, Solfvin (1984) concluded “Studies with a wide variety of treatments have conclusively affirmed that the administering physician or researcher is not independent of the results in double-blind treatment effectiveness studies...As a general rule, the double-blind cannot any longer be assumed to guarantee the exclusion of the nonspecific effects of the treatment, especially when the actual treatment has a weak or variable effect.” (p. 56). Still, double-blind replication studies conducted by researchers with a different world view than this author will surely add valuable information to the body of literature on the hare krsna maha mantra.

            Another drawback to this investigation is that it did not involve a clinical sample. Therefore, practical significance of the statistically significant findings is difficult to determine. For instance, if the hare krsna maha mantra was tested with persons diagnosed with anxiety disorders or depressive disorders, then the results would indicate whether subjects improved sufficiently to tangibly enhance daily functioning, and whether the intervention effectively impacted the disorder. Lacking a clinical sample, interpretation of clinical or practical significance is limited to analysis of the reported effect sizes, and cannot be directly generalized to clinical settings.

            When considering these shortcomings in the generalizability of the study, there should also be consideration of Mook’s (1983) contention that generalizability is often overemphasized, and that it is important to realize that generalizability is sometimes not the prime intent of the experiment. Mook points out that important phenomena such as biofeedback could never have been discovered by sampling or mimicking natural settings, but only through laboratory experiments with a very controlled setting. Applying this idea to the current investigation, the fact that the study indicates that chanting the hare krsna maha mantra worked to some degree with the tested population has significance in itself, despite the lack of formal generalizability inherent in the design.

            Royse (1995) explains that internal validity of a design assesses whether an intervention was truly responsible for the observed differences in the experimental group. Experimenter effects, described above in regards to ecological validity, were also an internal validity shortcoming of this study. That is, it is unclear to what extent the design detected the effects of the maha mantra on the dependent variables, or the effects of the researchers’ influence on the subjects.

            Though the random assignment of this study addressed some threats to internal validity, such as history and maturation, other threats, such as diffusion and resentful demoralization of respondents receiving less desirable treatments, were not resolved by random assignment. Though the study was designed so that researchers met individually with subjects, rather than in a large group, there were still a few roommate pairs, and they may have communicated with each other about effects they were experiencing, thus creating the shortcoming of diffusion. Also, it is possible that some subjects felt disappointed that they were not in a particular group, and the dependent measures may have assessed this disappointment, which compromises the validity of the study. Additionally, though the selection threat to internal validity is usually not a factor in studies utilizing random assignment, analysis of the pretest data of this study indicates that randomization may have been ineffective in creating group equality. Therefore it is possible that there were different kinds of people in the groups, which makes the design more like a quasi-experimental design rather than an experimental design, though the statistical procedures applied in the previous chapter are applicable to quasi-experimental designs. Related to selection, there may be other threats to the internal validity of this study, such as interactive effects of selection and history, and selection and maturation (Cook & Campbell, 1979).

            With regards to procedures for random assignment, it should be noted that randomization was performed before pretest. Twelve subjects left the study after random assignment and before the pretest was administered. These 12 were not equally divided between the groups. Alternatively, randomization could have been performed after pretest, to ensure equal numbers of subjects in each group at the beginning of intervention. Logistical considerations did not allow randomization after pretest in this study. Specifically, time constraints related to schedules of the research team and of many of the subjects who were university students, prevented the scheduling of an additional meeting after pretest specifically for introduction of the intervention. Therefore, randomization had to be performed before pretest, and thus the pretest meeting also served as the first day of the intervention. This ordering of random assignment before pretest may be regarded as a procedural shortcoming and could be related to pretest group inequalities.

            Though setting factors generally relate to external validity, there is a setting consideration of this study that is connected with internal validity. Meetings between researchers and subjects were sometimes held in a public place, such as the main plaza on the university campus, and sometimes in the home of the subject or the office of the researcher, depending on practicality and convenience of the research team member and subject. The salient point is that meeting place was not controlled in this study, and all three research team members noted that subjects seemed to respond best when the meeting was held in their home, and subjects seemed least comfortable when meetings were held in public. Thus, effects of meeting place may not be equally distributed across groups.

            The factors of the timing of random assignment and uncontrolled settings may be related to pretest inequalities that were found in four of eight variables, including four of five of the dependent variables. Also, these pretest inequalities may indicate that unknown, uncontrolled variables affected pretest scores. This is a shortcoming of the study, though the ANCOVA statistical procedure, especially when pretest scores are partiallized, tends to neutralize the effects of pretest inequalities.

            Concerning internal validity, it should be noted that the alternate mantra had the same syllabic pattern as the maha mantra. This controlled for the effects of syllabic pattern, and helped to isolate the effects of the mantras themselves.

            Power of the statistical tests used in this study was sufficient to detect differences that supported four out of five of the major hypotheses, as well as some of the secondary hypotheses. However, analyses in this study that were found nonsignificant at the .05 level include p values such as .096, .061, .058, .056, .054, .053, and even .051. Therefore, it is reasonable to conclude that greater power, achieved by a larger sample size (Orme and Combs-Orme, 1986), may have detected a greater number of significant differences than were found in this experiment. For instance, the maha mantra group decreased pretest-posttest tamas scores significantly more than the control group, though not significantly more than the alternate mantra group, though the decrease in tamas scores for the maha mantra group was greater than the decrease for the alternate mantra group (see Table 25). This lack of a significant finding may be due to an inadequate sample size and consequent deficiency in statistical power. With regards to chanting frequency, the discussion of hypothesis 13 earlier in this chapter reports that for some analyses chanting frequency had a medium or even large effect size (Rosenthal, 1997), although the p value was not significant. This suggests that with a larger sample size there may have been a greater number of significant relationships discovered between chanting frequency and dependent variables.

 

Clinical Significance and Application of Techniques

 

            Since this experiment did not involve a clinical sample, a main call to action of the findings is to research the hare krsna mantra in a clinical setting, such as with a sample of clients experiencing depressive, stress-related or anxiety-based disorders. Clinical studies have not yet been conducted with the maha mantra, though there are a few practitioners, such as the author, with some experience in using the maha mantra with clients.

            If a practitioner wishes to try the technique of chanting the hare krsna maha mantra with a client, it is important that the client feels comfortable with the method and voluntarily agrees to practice the mantra. The practitioner can demonstrate how to chant the maha mantra according to the description provided in the procedures section of the methodology chapter. As many other interventions of Eastern origin have been incorporated into the mental health professions, the hare krsna maha mantra can also be presented as potentially effective therapy, without reference to sectarian conceptions.

            Applications of the hare krsna maha mantra can include the same target problems that many other yoga, mantra and meditational techniques have been effective in treating. This japa study specifically dealt with stress and depression, both of which have been treated with various Eastern-style techniques. The spiritual intervention articles section of chapter two describes several studies reporting effective use of yoga, mantras and meditation in the treatment of stress and depression. Janowiak and Hackman (1994) found that meditation is effective in relieving stress, and Kaye (1985) successfully used mantra therapy to relieve depression and anxiety in an elderly population. Pearl and Carlozzi (1994) and Miller, Fletcher, and Kabat-Zinn (1995) found that Eastern-style interventions are effective in reducing stress, and Kutz et al. (1985) discovered that meditational techniques are effective as an adjunct to psychotherapy in relieving depression and anxiety. The literature reviews of Delmonte (1983) and Delmonte and Kenny (1985) concluded that mantras and meditation are effective in reducing anxiety and drug abuse, and also in improving physiological indicators. Though Delmonte found that a mock mantra is as effective as any other mantra, the results of this japa study indicate that the maha mantra produces desirable effects that an alternate, or mock mantra does not generate.

            Many other studies also support the use of Eastern-style interventions in treating stress and depression, and, combined with the findings of this study, they support the experimental use of the hare krsna maha mantra for alleviating stress and depression. Wardlaw (1994) effectively used yoga techniques to treat stress-related symptoms such as bruxism. Khumar, Kaur, and Kaur (1993) successfully treated clinically depressed college students with yoga methods. Joseph (1998) asserts that throughout the mental health fields there is growing interest in yoga techniques as a stress reliever. The results of this study indicate that the maha mantra is a yogic method that should be tried with problems related to stress and depression.

            Stress is often regarded as the underlying factor in substance abuse and addictions (Ellis and Corum, 1994), which points to other potential uses for the maha mantra. Many researchers and practitioners, such as Kremer, Malkin, and Benshoff (1995), Alexander, Robinson, and Rainforth (1995), Karel (1993, and Sands (1994), have been successful in using mantras and other yoga techniques in the prevention and treatment of substance abuse. According to Vedic theory, substance abuse is indicative of tamas guna, and this study indicates that the maha mantra decreased the level of tamas guna. A. C. Bhaktivedanta Swami Prabhupada (1976) suggests in the Srimad-Bhagavatam that chanting the hare krsna maha mantra is specifically effective for prevention and treatment of drug abuse. Eastern-style interventions have been efficacious in many other ways also, including development of empathy and prevention of stress and burnout in mental health professionals (e.g., Karel, 1993; Shapiro, Schwartz, and Bonner, 1998; Keefe, 1996), treatment of anxiety disorders with a clinical population diagnosed with panic disorder and generalized anxiety (Kabat-Zinn, Massion, Kristeller, & Peterson, 1992), treatment of phobias (Snaith, Owens, & Kennedy, 1992), trauma intervention (Urbanowski & Miller, 1996), treatment of asthma (Lane, 1994), and treatment of posttraumatic stress disorder (Canda & Phaobtong, 1992).

            Vaisnavas, adherents of Vedic culture and philosophy, have been chanting the hare krsna maha mantra and utilizing other techniques described in the Vedas for many of the above purposes for centuries. Rangaswami (1996) explains that meditation, mantras and yoga are an integral part of the Indian system of psychotherapy. Canda and Phaobtong (1992) describe meditational techniques specifically implemented for treatment of posttraumatic stress disorder among Buddhist refugees. Similarly, a person experiencing a crisis or severe anxiety from any source can be guided to chant the hare krsna maha mantra for alleviation of stress and development of sattvic symptoms, such as serenity and contentment. Of course, as described above, the counselor must be confident that this technique is appropriate for the particular client, and that the client consents to practice the method.

Relevance to Social Work

 

            Keefe (1996) describes social work applications of Eastern-style interventions in treating depression, substance abuse, and excessive anxiety. Further, he asserts that these techniques are potentially important in development of social work skills in professional training. Canda (1988) claims that there are many meditative practices that have not yet been applied in social work. Meditation on the hare krsna maha mantra is one of these techniques, and social workers with an inclination towards innovation and who feel comfortable with Eastern-style mantra meditation can apply this technique to help clients with stress, depression, and related problems.

            Social workers are active in many fields, and yoga techniques are becoming increasingly popular in these fields. For example, medical social workers can note that Derr, Shaikh, Rosen, and Guadagnimo (1998) surveyed medical students and found that a majority of the students identified yoga as a beneficial complementary therapeutic technique. Cormier and Cormier (1997) describe yoga as an important and effective skill for mental health professionals utilizing behavioral interventions. Many hospice workers are trained and educated as social workers, and spirituality has been found to be an important component in hospice work (Millison, 1995). According to the Vedas, chanting the maha mantra is a spiritual activity, based in sattva guna, or suddha sattva. For centuries Vaisnavas have chanted the hare krsna maha mantra to deal with issues of death and dying. By becoming conversant in techniques such as chanting the hare krsna maha mantra, social workers can add a useful tool to their repertoire of skills that can be implemented in diverse settings with various populations.

            Cultural diversity is a key element of the social work profession (National Association of Social Workers, 1990), and this entails respecting the world view and practices of cultures different from that of the social worker. In dealing with psychological issues, social workers should understand that Western paradigms may not be applicable for explicating and comprehending the psychology of indigenous peoples (Gergen, Gulerce, Lock, and Misra, 1996). Laungani (1993), in his examination of differences in cultural perspectives on stress, emphasized that in India people have been relying on yoga and mantra techniques to relieve stress long before the development of Western psychology. Familiarity with these techniques, such as chanting the hare krsna maha mantra, will help social workers to appreciate and effectively work with persons from different backgrounds. Also, by incorporating methods from a different tradition, adherents of that tradition in the West will feel supported and validated, as individuals and as an ethnic group. Such multicultural upliftment is an important aspect of the mission of social work (National Association of Social Workers, 1990).

            Though the maha mantra can be applied in a behavioral context, it is, according to the Vedas, a spiritual approach to psychological and social health. As described from the beginning of this dissertation, there is a need for social workers to explore spirituality and spiritually-based interventions more than they do at present (Canda, 1988). Lloyd (1997) urges social workers to broaden their perspective to incorporate the spiritual dimension of life, and he emphasizes that a spiritual perspective is especially important in social work surrounding issues of death and dying. Diagnostic and Statistical Manual of Mental Disorders-IV (American Psychiatric Association, 1994) includes spiritual problems as a focus of clinical attention, and this has raised the awareness of mental health professionals regarding the importance of considering a person’s spiritual values in psychosocial development. Among helping professionals social workers are especially known for considering all dimensions of persons and problems. Jacobs (1997) concluded that attending to the spirituality of helping situations invites social workers to expand their personal and professional boundaries. Considering this, there is clearly benefit for social workers to learn and teach spiritual techniques such as chanting the hare krsna maha mantra, when it is compatible with the world view of the client.

            Bullis (1996) asserts that, philosophically, both social workers and spirituality promote common interests and self-respect. Social work and spiritual professionals have similar goals, such as personal healing and alleviation of community strife, violence and ignorance. Bullis claims that spirituality offers social workers a way of transformation for their clients, and he offers several suggestions for spiritually-based assessment and interventions in social work practice. He recommends that meditation and prayer be incorporated in social work interventions, and the hare krsna maha mantra is a type of meditation and prayer (Goswami, 1977). Bullis writes “...meditation or deep prayer is a safe, fun, and effective way to access the spiritual dimensions” (p. 62). Further, he states, “...social work should abandon the strictly physical cosmology in favor of a post-positive, spiritual cosmology...the great divorce between social work and spirituality is inauthentic and unsupportable. Clients who have a desire, articulated or not, to have their spiritual concerns addressed, deserve to be heard by a competent and thorough professional” (p. 145, 163).

Suggestions for Further Research

 

            Bullis (1996) contends that the spiritual dimension is not well understood in the social work profession, and that much research needs to be conducted on spiritual assessment and intervention. Considering this, further research can be conducted on the VPI (Wolf, 1998) as a tool for assessment and evaluation in social work, and on the maha mantra as an intervention.

            Path analysis is a statistical method for determining causal relationships (Agresti & Finlay, 1986), and this technique could be applied to research on the VPI (Wolf, 1998) and the maha mantra. Whereas the current study indicates several relationships of association between variables, path analysis could examine more deeply whether these relationships are causal. Even with the data of this maha mantra study path analysis could be applied in order to glean further information. For instance, each guna subscale of the VPI is composed of several attributes, with each attribute derived from the Bhagavad-gita (Prabhupada, 1972) and presented as an item or items on the subscale. The sattva subscale, for example, contains items on satisfaction, preferred foods, view towards violence, cleanliness, spirituality, intelligence, willpower, verbal gentleness, self-control, and dutifulness. Responses on each item could be averaged and causal relationships with dependent variables could be assessed through path analysis. This could potentially reveal much useful information about the detailed effects of the maha mantra, and also about the components of the VPI subscales. Path analysis involves explicitly defining the presumed causal relationships, and this will necessitate the researcher delving further into Vedic theory to ascertain hypothetical relationships. As an example, a fairly intricate theoretical relationship between the three gunas is described above in the discussion on hypothesis 4. This relationship can be tested with the VPI (Wolf) and path analysis.

            Nugent (1996) describes the integration of single-system design and group-comparison methods using the Hierarchical Linear Models (HLM) analytical method. HLM allows the researcher to aggregate data from single-case and group designs, thereby minimizing, to some extent, the drawbacks of both approaches. With single-case experiments, for instance, it is generally difficult to analyze data for purposes of generalizability, due to lack of effective statistical methods that can be used with a small number of subjects. With group designs there are usually a very small number of data points for each subject, and therefore much information about each individual subject is lost. Thus, the dynamic nature of subject responses often goes unnoticed with group designs. In this dissertation both a single-system design and a group design examining the maha mantra are described. Consequently, the data from these experiments may be ideally suited for HLM.

            Future hare krsna maha mantra japa studies can contain increased rigor, including elements such as a double-blind design, better control of location for meetings between researchers and subjects, a greater range for chanting frequency built into the design of the experiment, a clinical sample, and a larger sample size. Also, the alternate group can receive a treatment that has been proven effective, and then the researcher could assess whether the maha mantra is more effective in treating the target dependent variable than the other intervention. Japa studies can be performed in various settings, with various populations, and tested with different target variables. For instance, from guna theory many dependent variables, such as anger control and relationship satisfaction, can be incorporated into a maha mantra study. Future studies can also experiment with different time periods for the intervention, such as one week, eight weeks, or twelve weeks. In this way, the applicability and effectiveness of this technique can be evaluated across several dimensions.

            Additionally, future studies can incorporate greater control for the factor of past experience of subjects with chanting and related techniques. As described in the sampling section of the methodology chapter, the initial phone conversation with each subject included a question concerning whether they had prior experience with chanting, biofeedback, meditation, or other yoga techniques. Answers to this question were used to create two blocks, “past experience” and “no past experience”, for the process of random assignment. These blocks, however, controlled for the variable of past experienced based on a yes or no question, without differentiating between length or intensity of experience. Future studies on the maha mantra can include, for instance, a group with experience chanting the maha mantra, a group with experience chanting a different mantra, a group with silent meditation experience, and a group with biofeedback experience. In this way, differential effects of the maha mantra can be more precisely analyzed.

            In the Vedas, chanting the hare krsna maha mantra is described more as a component of a spiritual lifestyle than as a therapeutic intervention. All aspects of this lifestyle are based on guna theory, and thus many aspects of the Vedic lifestyle can be assessed by the experimental method. With regards to further examination of the maha mantra, future studies can incorporate a group that continues to chant japa, without discontinuation at the end of an intervention period. As described earlier in this chapter with regards to hypotheses 6-10, effects of the maha mantra diminished during the followup period. In Vedic culture, many people daily chant a fixed number of japa rounds of the hare krsna maha mantra. Persons who have been regularly chanting in this way for years, as well as persons who have recently begun regular chanting, can be compared with chanters who cease chanting after a pre-determined time period, and with a control group, an alternate mantra group, as well as with alternate therapy groups. Alternate therapy groups would involve an intervention that has already been shown to be effective in addressing the dependent variable. For instance, the maha mantra could be compared with an established cognitive-behavioral therapy for relieving stress.

            Many other aspects of a Vedic lifestyle could also be researched. For example, the Vedas explain that each time period of the day has predominant modes. The hour before sunrise, for instance, is called the brahma-muhurta, and is predominated by sattva guna, according to Vedic descriptions (Prabhupada, 1976). Therefore, one who chants the maha mantra at this time is predicted to achieve greater benefit, in terms of any of the dependent variables connected with increased sattva, such as reduction of stress and depression, compared with one who chants at a different time of the day. A study could be designed incorporating this aspect of Vedic guna theory.

            As a final example, the Vedas assert that there is great benefit in eating prasadam, which is food that is prepared according to certain principles and is described in the Vedas to be completely in the mode of goodness, or sattva. Krsna indicates in the Bhagavad-gita (Prabhupada, 1972) that all foods are situated in the modes of nature, and to develop sattvic qualities one should eat sattvic foods, such as prasadam. Prasadam is vegetarian. Thus, a design could be formulated with a meat-eating group, a vegetarian non-prasadam group, a prasadam group, and a prasadam group that also chants a few rounds per day of the maha mantra. In this way guna theory and Vedic practices can be further tested.

 

 

1