From the above given descriptions of shamanizing it can be seen that the forms of shamanizing should be regarded as any other ethnographical complexes — they consist of variable elements and it may be supposed that each of them has a certain function in the complex. Therefore, our analysis, in order to be complete, must enter into further details.
First of all, it should be pointed out that the basis of all performances consists in theories concerning the characteristics of the spirits and the possibility of the regulation of their activity. In order to understand the fact that, for instance, among the Transbaikalian Reindeer Tungus the shamans, in great performances, are usually dealing with the spirits of the upper world, and that the Manchu shamans have very little, even almost nothing to do with the spirits of the upper world, we have not only to refer to their respective conceptions of the spirits, but also to find out how the functions of the shamans are distributed among these different groups. Here it should be noted that the principal spirit of the shamans spirits s'aman'i da yachan the complex of the Reindeer Tungus of Transbaikalia is transferred to the group of spirits inhabiting the upper world, while the Manchus have no such principal spirit, and all shamanistic spirits are located in this world — chiefly on the tops of mountains. Among these Tungus the shaman's spirit is only one of their great spirits of the upper world, and this spirit is naturally closely connected with the other spirits of that world. It is not a simply mastered spirit, but is almost a protector, while other spirits are mastered. Thus the Tungus shaman has nothing to do with it. In fact, the function of dealing with the spirits of the upper world, such as all classes of endur'i has been taken up by the p'oyon saman (who, as shown, is not at all a shaman, but a priest), by common experienced people and by representatives of religions, such as Buddhism and Taoism. Moreover, among the Manchus, a great number of diseases, such as, for instance, smallpox, chicken-pox, measles, and others, which especially affect children, are considered as a special group of infectious diseases controlled by the group of spirits of the upper world (Chinese complex of n'angn'ang). A great number of other pathological conditions are considered by the Manchus as diseases, which sometimes ought to be treated by specialists — Chinese doctors. Naturally, all of them are exempted from the potential scope of shaman's interference. Among the Reindeer Tungus of Transbaikalia the same cases are ascribed to the activity of the spirits which may be managed by the shamans. Since the existing hypotheses and theories as to the regulation of diseases are different, the competence of the shamans is also different, and consequently there is a difference in the cases attended by the shamans of these two ethnical groups.
There is one more condition which produces difference, namely, the great number of methods of producing extasy and carrying out the performances. The Manchu shamans borrowed their methods from the Chinese and the great Buddhistic priests, while the shamans of the Reindeer Tungus of Transbaikalia have these elements from a different source — the Buriats, Yakuts, and other neighbours who possess distinct complexes.
We have already seen that the paraphernalia used by the Manchus and Tungus of Transbaikalia are also different, being received and adapted for different purposes. For instance, an elaborated complex of iron implements is found only among the Manchus. Practically these paraphernalia have an influence on the process of shamanizing. It is thus evident that in these two instances we have two distinct complexes of shamanistic performance, which are conditioned by the existing complexes of spirits, theories concerning the treatment, technical methods of shamanizing, and different paraphernalia.
When we take two very distinct complexes, we can easily see how their structures can be understood as a result of the historic formation and readaptation of elements into a complex with a definite function; but when the complexes are more or less alike, it is more difficult to see the functional mechanism and the internal equilibrium of the complexes, which is necessary in order to understand both the functional mechanism and the equilibrium of the complex.
The difficulties of this case can be seen, for instance, in the complex of the Manchus and that of the Tungus of Manchuria, the latter being much nearer to the Manchu complex than the complex of the Reindeer Tungus of Manchuria, but at the same time distinct enough for revealing some characteristics of functional mechanism and an original system of the equilibrium of the complex.
Among the Birarchen the theory of spirits accepts the class of endur'i which cannot be dealt with by the shamans almost in the same way as among the Manchus. It is therefore natural that the shamanizing to the upper world is not practised. This function, however, is not performed by a p'oyun saman, but merely by experienced people, among the Birarchen the complex of spirits of the upper world, as active factors in human life, are of a lesser importance as compared with the same spirits of the Manchu complex. The complex of spirits responsible for diseases of children, or, better to say, regulating these diseases -smallpox, chicken-pox, measles, etc. — are regarded in the same way as among the Manchus, but the group is increased with another series of complex spirits ain'i burkan which produces a great number of diseases (apparently typhus, paratyphus influenza, etc.) which cannot be treated by the shamans. Hence, we have still a greater number of diseases, (as compared with the Reindeer Tungus of Transbaikalia) which are exempted from the shaman's competence. Some methods of shamanizing are similar in the Manchu and Birarchen complexes, as for instance, the properties of the shaman when he introduces a fire-spirit. It is very likely that the source is the same — the Chinese complex. On the other hand, the shamanistic paraphernalia in general are quite different, and this fact implies certain differences in the technique of performance, e.g. the absence of the belt with iron conical trinkets in the Birarchen costume excludes the necessity of making rhythmic movements with the back during the performance, the sounding effects being produced by the brass mirrors chiefly fixed on the frontal part of the coat. Thus, the Birarchen costume differs from the Manchu costume, whence there are differences in the technique of performance.
Here we have again two complexes which are different and which, at the same time, possess a great number of common elements. When we compare two pairs; namely, the Manchu -Reindeer Tungus, and the Manchu — Birarchen, we can see that the difference in the theory of spirits in the first pair shows a difference of the performance complexes, but in the second pair this is not so, for in both cases the spirits of the upper world are exempted. It must be also noted that the exclusion of the spirits of the upper world in the Manchu and Birarchen complexes may historically be due to two different conditions, namely, the shamans may have been gradually eliminated from the interference with the upper world, or the spirits of the upper world may have been introduced much earlier than the shamans, and thus were not included in the groups of spirits dealt with by the shamans; or it may have happened later, as it would be, if these groups should now adopt, for instance, Christianity.
The difference of the trinkets on the shaman's costume in all three instances is such, that the performances themselves differ: in the Manchu complex there is an intensive side-movement with the back, in the Reindeer Tungus costume — a play of differently selected bells in the back and front parts of the coat, in the Birarchen complex — a slight jumping to produce a rhythmic noise of all brass-mirrors. These are facts of great interest, for, in spite of the difference, the same aesthetic and hypnotic effect of the performance is reached. One can clearly see what the function is of the trinkets, brass-mirrors and bells, although all of them may be differently explained, and justified by different theories.
When we analyse element by element, and complex by complex, the facts concerning performances, we will obtain the same picture of geographical and ethnical distribution of types of performance and definite complexes observed in different ethnical groups, as we did in the case of the analysis of the shamanistic paraphernalia. However, there will not be a perfect correlation. I shall give here some instances.
(1) The performance of great shamanizing, when shaman goes to the lower world, is carried out among the Goldi chiefly for bringing souls of dead people to the place of settled life, while among the Manchus it is done chiefly for finding out the cause of a sickness which has been ascribed to some deceased person, and among the Reindeer Tungus of Transbaikalia it is usually carried out merely to maintain relations with the dead clansmen and to assist them if they need something, among the Manchus the care of souls of clansmen is a duty of the p'oyun saman; the carrying of souls, among both Manchus and Reindeer Tungus of Transbaikalia, is not a duty of the shaman. A special use of this form of shamansing among the Goldi seems to be a local phenomenon, also practised by their neighbours the Udehe and the Oroci - but unknown among the Tungus of Manchuria I do not mean to say that this performance has never been practised by the Tungus, of Manchuria, but it is not practised now. The caring for the souls is much more typical of the Goldi complex — perhaps it is one of the manifestations of Chinese influence — than of any other complex, including the Manchu one. Therefore, it may be supposed that the shamanizing to the lower world among the Goldi is of secondary origin. However, there is no other direct evidence to support this supposition.
(2) The performance of shamanizing to the upper world spirits, as found among the Reindeer Tungus of Transbaikalia, the Nomad Tungus (a slightly different complex!), and the Reindeer Tungus of Manchuria (also a different complex in so far as this statement is based on oral communication), but it is unknown among the Manchus and Northern Tungus groups of Manchuria. As stated, this form of performance might not exist all among these groups. Considering the fact that the complex of the upper world is especially well represented and shamanizing according to this complex is practised among the Buriats and Yakuts, and that there are some elements (wooden staff-horse, counting by nines, terms, etc.), which cannot be suspected of being Tungus inventions, but are used especially by Buriats, we may suppose that this complex of performance was borrowed by the Tungus from the Buriats, or was inspired by the Buriat complex. It borrowed, it was greatly modified, and perfectly adapted to the Barguzin and Nerchinsk Tungus complexes. The complex of the upper world performance is almost a direct imitation of the Buriat complex, and that of the Reindeer Tungus of Manchuria seems to be either an imitation of the Yakut form, or was strongly inspired by the Yakut complex. The above shown three forms of the performance that exist today may thus be of different origin, but in all cases are well adapted to the existing conception of triple spirit-world and to the possibility of influencing the spirits.
It will be useless to proceed further with the analysis of the various forms of performances which, as a matter of fact, reveal no new aspects as to the character of ethnical and geographical distribution of complexes. As shown by the above demonstrated cases, the analysis of performances is more difficult than that of the paraphernalia, for the chance of parallelism in shamanizing is greater, than in the imitation of forms. The same theories and the same methods, which as such may be independent of the performance as a whole, may be responsible for the creation of similar forms of performance. Thus, although we may carefully mark down all data and reconstruct the complexes, these methods of dealing with the problem will throw no new light on the general character of shamanism, or on the spreading of particular complexes and elements. Our hypothetic complexes might become quite artificial, and as an instrument for further dealing be quite dangerous constructions. However, there are some aspects in the performances of all types and in all groups which can be further analysed and generalized: (1) the technique of performance; (2) the ritual; (3) psychological conditions of the performance; and (4) the social aspects of the performance.