464 BUREAU OF AMERICAN ETHNOLOGY [BuU. 175 



"survivor," so that, psychoanalytically at least, even these "single sui- 

 cides" belong, in a sense, to (real or social) "death clusters." 



This insight enables us to reconcile Durkheim's (1897) theory of 

 the motivating force of social isolation in suicide with the psycho- 

 analytic theory of clustering suicides. It is sufficient to suggest that, 

 from the psychological point of vieic, the physical death of a love 

 object and the psychosocial withdrawal of one's associates have exactly 

 the same effect and suicide-motivating force. 



As regards clustered suicides, the fact that the successful first suicide 

 of a given cluster is sometimes followed by a bungled second suicide 

 is of special importance for the understanding of the basic importance 

 of the Mohave tendency to commit suicides in clusters. Indeed, on 

 closer scrutiny these bungled reactive suicides do not suggest that the 

 "model" suicide lacks compellingness. On the contrary, they suggest 

 that the "model" is so compelling that even the subjectively inade- 

 quately motivated survivor — who does not really want to die — feels 

 impelled to make at least a suicidal gesture, be it only by clumsily 

 attempting to cut his own throat (Case 117) or, more simply, by con- 

 templating suicide, without actually doing anything about it (Case 

 115). Moreover, it is quite noteworthy that the imitative suicide 

 usually tries to kill himself by the same means as his model. This is 

 apparently recognized even by the Mohave themselves, since they 

 specified that Pa:hay (Case 117) used a knife only because — unlike 

 his brother (Case 116) — he had no gun at his disposal. A more in- 

 direct proof of this thesis is the fact that the "vicarious suicide" of 

 victims of witchcraft leads up to the vicarious suicide of the witch, 

 who incites his victims' relatives to kill him (Devereux, 1937 c; and 

 pt. 7, pp. 387-426) . Likewise, both the first and the second of a pair of 

 twins who do not wish to live cause themselves to die by identical 

 means, i.e., by making themselves ill (Devereux, 1941; and pt. 7, 

 pp. 348-356). 



The last argument showing the motivating force of the "model" 

 suicide, as distinct from that of genuine subjective reasons, is that 

 none of the six "single" suicides was bungled. By contrast, tliree of 

 the clustering suicides (Cases 117, 121 and 122), and all four of the 

 recorded funeral suicides were bungled. Moreover, two of the six 

 single suicides were committed by rather gruesome means, such as 

 self -suffocation by eating earth (Case 123) or hanging oneself from 

 a low beam (Case 124), whereas all successful clustering suicides were 

 committed by simple, rapid, and reliable means. It is also note- 

 worthy that the "spectacular" funeral suicides by fire were invariably 

 bungled, and that the would-be suicides knew in advance that they 

 would be saved in ample time. 



