Such a transference of ideas is further complicated by the transference of sentiments. From beginning to end in a magical rite we find the same sentiment, which gives sense and style to the ceremony, and which, in all truth, directs and orders the associations of ideas. This is the factor which explains how the law of continuity functions in magical rites.
In most applications of the law of sympathy through contiguity, it is not merely a matter of spreading a quality or state from one object or person to another. If the law, as we have formulated
it, were absolute, or if-in those magical rites where it functions-it were the sole factor involved (and then only in its intellectual form) and if we are in fact only concerned with the association of ideas, then we would be able to state at once that all the elements of a magical chain-constituted by an infinity of possible, necessary or accidental contacts-would be equally affected by the quality they were engaged in transmitting and, consequently, that all the qualities of any element in the chain, whatever they were, would be transmitted in toto to each other part. This, however, is not the case; if it were, magic would be impossible. The effects of sympathetic magic are always limited to the effects desired. On the one hand, the current of sympathy is interrupted at precise points. On the other, only a jingle transmissible quality or, at the most, only a few, are transmitted. Thus, when the magician absorbs his client's illness, he himself does not become ill. Similarly, he communicates only the everlasting nature of the powder taken from a mummy which is used to prolong life, the value of gold or diamonds, the insensitivity of a dead man's tooth. Contagion is limited to those properties which the magician detaches and abstracts from the whole.
It is even held that by their very nature the properties in question are localized in one spot. A man's good fortune, for example, is to be found in the straw of his thatched hut. From the idea of localization derives that of separability. The ancient Greeks and Romans thought that they could cure eye diseases by transmitting a lizard's sight to an afflicted individual. The lizard has his eyes put out and is then brought into contact with pebbles which are used as amulets. In this way its sight, cut off at the roots, could be made to go, in its entirety, wherever the magician wished. Separation and abstraction are expressed, in this example, by these rites; but this care is not always necessary.
Those limitations which are placed on the theoretical effects of the law are the real condition of its application. The same requirements which make the rite work and lead to the
association of ideas also determine their selection and limitation. Thus in all cases where the abstract notion of magical contiguity functions, the association of ideas is accompanied by transfer of sentiments, by phenomena of abstraction and exclusive attention, and by direction of intent: phenomena which take place in the consciousness, but which are objectivized in the same way as the association of ideas themselves.
The second law, the law of similarity, has a less direct expression than the first as far as ideas of sympathy are concerned. We think Frazer was right when he, along with Sidney Hartland, reserved the term sympathy proper for phenomena including contagion, and called this other category, which we shall now deal with, 'mimetic sympathy'. This law of similarity has two principal formulas which it is important to distinguish: like produces like, similia similibus evocantur; and like acts upon like, and, in particular, cures like, similia similibus curantur.
As far as the first of these formulae is concerned, it amounts to saying that similarity equals contiguity. The image is to the object as the part is to the whole. In other words, a simple object, outside all direct contact and all communication, is able to represent the whole. This is the formula which is apparently used in black magic. However, it is not simply the idea of an image which is at work here. The similitude which comes into play is, in fact, quite conventional; there is nothing resembling a portrait. The only thing the image and the victim have in common is the convention which associates the two. The image, the doll or the drawing is a very schematic representation, a poorly executed ideogram. Any resemblance is purely theoretical or abstract. The law of similarity, therefore, when it is working, like the preceding law presupposes the existence of phenomena of abstraction and attention. Assimilation does not derive from any illusion. Moreover, these images are dispensable. The mere mention of a name-even thinking it, the slightest rudiment of mental assimilation-is sufficient for an arbitrarily chosen
substance-bird, animal, branch, cord, bow, needle, ring-to represent the victim. The image is, therefore, defined only through its function which is simply to produce the person. The basic thing is that the function of representation should be fulfilled. From this it follows that the object to which this function is attributed may be changed for another during the course of a ceremony or that the function itself may be divided between several objects. If one wishes to blind an enemy by passing one of his hairs through the eye of a needle, which has sewn up three shrouds, and then sticking holes in the eye of a toad with the same needle, the hair and the toad are both used in turn as volt. As Victor Henry has remarked on the subject of a Brahmanical magical rite, a single lizard may, at various points in the rite, represent the curse, the person uttering the curse and the evil contained in the curse.
In the same way as the law of contiguity, the law of similarity can be applied not only to persons and their souls but also to objects and modes of objects, in their possible and real aspects and their moral and material ones. The idea of the image-as it becomes more extensive-assumes the nature of a symbol. Rain, thunder, sun, fever and unborn children may be symbolically represented by poppy flowers, an army by a doll, a village union by a pot of water, love by a knot, etc., and they are created through their representations. The fusion of images is complete for these cases as well as the earlier ones, since it is not an imaginary wind, but the real wind which is found enclosed in a bottle or goatskin, tied in knots or encircled by rings.
On top of this application, the law results in a whole series of interpretations which are quite remarkable. In the determination of symbols, in their utilization, we have the same phenomena of exclusive attention and abstraction without which we should be unable to conceive of the application of the laws of similarity-as exemplified in the images used in black magic-nor the functioning of the law of continuity. Of the objects chosen as
symbols the magicians are concerned with but a single quality, for example in clay-its coolness, weight, its leaden colour, its hardness or softness. The needs and propensities of the rites not only determine the choice of symbols and the use to which they are put, but also limit the consequences of assimilation, which theoretically, like the series of contiguous associations, ought to be limitless. Moreover, all the qualities of the symbol are never transmitted or symbolized. The magician believes himself in control, to be able to channel at will the effects of his actions. He is able, for example, to restrict the effects of funerary symbols to sleep or blindness. The magician who brings rain is content with a shower, since he fears a deluge. The man who is assimilated to a frog which has had its eyes gouged out does not magically turn into a frog.
This apparently arbitrary business of abstraction and interpretation does not result in an infinite multiplication of possible symbolic structures. On the contrary, we have noticed that the existing scope for lively imaginations in any one magical system seems curiously limited. For one object we have to be content with a single symbol, or at most but a few. More surprisingly, there are only a few objects which can be expressed symbolically. The magical imagination has been uninventive to such an extent that the small number of symbols which have been thought up have been put to manifold uses: magical knots are required to represent love, rain, wind, curing, war, language and a thousand other things. The poverty of the symbolic system is not the creation of single individuals whose dreams, psychologically, would be very free. The individual finds himself confronted by rites and traditional ideas which he is never tempted to refurbish because he has faith in tradition; without tradition there can be no beliefs nor rites. For this reason it is natural that traditional symbols should be meagre.
The second formula of the law of similarity-that like acts on like, similia similibus curantur-differs from the first in that, even in
its expression, the actors take into account those phenomena of abstraction and attention which always condition, as we pointed out, the application of the other rites. While the first type deals only with general evocations, these rites involve an effect being produced in a well-defined direction. The course of the action is then determined by the rite. Take, for example, the legend of the curing of Iphiclos. One day his father, Phylax, brandished a blood-stained knife at him while they were castrating goats. He was made sterile through the sympathetic effect of this action and failed to have children. When the diviner Melampos was consulted he made Iphiclos drink wine mixed with some rust from the knife which had been recovered from the tree where Phylax had hidden it. This was repeated over ten days. The knife was capable of exacerbating Iphiclos's condition, and at the same time Iphiclos's qualities could pass into the knife through sympathy. Melampos permitted only the latter to take effect, and limited it to the illness in question. In this way the king's sterility was absorbed by the sterilizing power of the instrument. The same thing occurred in India when Brahmans cured dropsy through ablutions. Here the patient was not made to take an overdose of liquid, but the water with which he came into contact absorbed the liquids which were making him suffer.
While these facts can be grouped under the law of similarity, deriving as they do from the abstract notion of mimetic sympathy, of attractio similium, their special features cause them to be placed in a class apart. This is more than a corollary of the law; it is a kind of concurrent notion which may equal it in importance owing to the number of rites which it dominates in each ceremony.
Before leaving the exposition of the second form of our law of similarity, we find ourselves face to face with the law of opposition. When like is found to cure like, what we have in fact is the opposite. The sterilizing knife produces fertility, water produces the absence of dropsy, etc. A complete formula for these rites
would be: like drives out like in order to produce the opposite. Conversely, as far as the first series of facts involving mimetic sympathy is concerned, when like evokes like it drives away the opposite: when I cause rain to fall by pouring water on the ground I am causing the disappearance of drought. In this way the abstract notion of similarity is inseparable from the abstract notion of contrariety. The two forms of similarity could thus be brought together in one formula, 'opposite drives away opposite', or in other words could be included in the law of opposition.
But this law of opposition has always existed separately in the minds of the magicians. Sympathy may be equated with antipathy, but the two are clearly distinguished from each other. A proof of this is that in antiquity there were books called . Whole systems of ritual-involving pharmaceutical magic and counter-spells-can be classed under the idea of antipathy. Magic has always speculated on polarity and opposition: good and bad fortune, cold and hot, water and fire, freedom and coercion. A large number of things have been grouped together in opposites and this opposition has been taken advantage of. We therefore consider the idea of contrast as a distinct idea in magic.
In fact, just as similarity cannot exist without opposition, opposition does not exist without similarity. In an Atharvanic ritual, rain is made to stop by invoking its opposite, the sun, with the aid of a wood known as arka, which means light, lightning or sun. In this rite of opposition we can already see the mechanisms of sympathy, properly so called. A further proof that they are not incompatible is that the magician uses the same piece of wood directly to stop storms, thunder and lightning. In both cases the material used in the rite is the same; only their treatment varies slightly. In the one case fire is exposed, in the other glowing coals are buried. This simple ritual modification expresses the will which directs the rite. We can state, therefore,
that opposite drives out its opposite by evoking its equivalent.
Thus the separate formulas covered by the law of similarity can all be exactly fitted into the formula of opposition. From the point of view of the ritual schema used in our study of sacrifice, it can be said that symbolic structures are present in three schematic forms, each corresponding to one of three formulas: like produces like; like acts on like; opposite acts on opposite. They differ only in the ordering of their elements. In the first case, we think primarily of the absence of a state; in the second, we are dealing first with the presence of a state; in the third, we are dealing with the presence of a state opposite to that which is desired. In the first, we think in terms of the absence of rain, which has to be produced through a symbol; in the second, we think of falling rain which is made to stop through a symbol; in the third, rain is conjured up and then brought to a stop by evoking its opposite through a symbol. In this way abstract notions of similarity and opposition may both be encompassed by the more general idea of traditional symbolism.
In the same way the laws of similarity and contiguity tend to merge into each other. Frazer has stated the case well, and he could easily have produced proof of it. Rites of similarity normally involve contact: contact between the witch and her clothing, the magician and his wand, a weapon and a wound. The sympathetic effects of a substance are only transmitted by absorption, infusion, touch, etc. Conversely, these contracts are usually only vehicles for qualities of symbolic origin. In black magic, practised with the aid of a person's hair, the hair is a link between the desired destruction and the victim to be destroyed. In an infinity of similar cases we do not find any clear pattern of ideas and rites, but an interweaving of elements. The actions become so complicated that it is only with great difficulty that they can be neatly ordered into one or other of our two categories. In fact, whole series of black magic ritual contain contiguities, similarities and neutralizing oppositions alongside pure
similarities, without the magicians bothering about them or really having in mind anything except the final objective of their rite.
Let us now consider the two laws, ignoring their complexities for a moment. First of all, we find that sympathetic (or mimetic) actions performed at a distance are not always thought to be working on their own. There is the idea of euvia which leave the body, magical images which travel about, lines linking the magician and his field of action, ropes, chains. Even the magician's soul leaves his body to perform the act he has just produced. The Malleus maleficarum mentions a witch who dips her broom in a pond to bring on rain and then flies away into the air to search for it. Several Ojibway pictograms depict the priest-magician, after the ritual, holding his arms up to heaven, piercing the vault of the sky and drawing clouds towards him. It is this kind of thing which makes us imagine that similarity is the same as contiguity. On the other hand, contiguity may also be the same as similarity, and with reason: the law holds only when the individuals involved, the substances in contact, in fact the whole ritual ensemble, contain the same circulating essence, which renders them identical. That is why all our abstract, impersonal representations of similarity, contiguity and opposition-even if they may once have been separately conceived-have become naturally confused and confusing. They are obviously three aspects of one idea; and it is this idea we must now set to work and clarify.
Several magicians who concerned themselves with the meaning of their ritual have shown themselves perfectly aware of this confusion. The alchemists had a general principle, which appears to have summed up perfectly their theoretical reflections. They always prefixed it to their schemes: 'Each one is the whole and the whole is in each one'. Here, chosen at random, is a passage which nicely expresses this principle: 'Each one is the whole and it is through it that the whole is formed. One is the
whole and if each one did not contain the whole, the whole could not be formed'. This whole, which is contained in everything, is the world. And we are sometimes told that the world is conceived as a unique animal, whose parts, however disparate they may seem, are inextricably associated. Everything has something in common with everything else and everything is connected with everything else. This kind of magical pantheism provides a synthesis for our different laws. The alchemists, however, never insisted on this formula, except in so far as it provided their studies with a metaphysical or philosophical commentary, of which we have only the remnants today. However, they did insist on one formula which was juxtaposed to the other: Natura naturam vincit, etc. It is 'nature', by definition, which is found both in the object and in its parts. Here we have the basis of the law of contiguity, and it is this which is found in all members of a single species and forms the basis of the law of similarity. It is the same thing which enables an object to act on an opposite object of the same kind, and is, therefore, the basis of our law of opposition.
The alchemists did not confine themselves to the field of abstract considerations and it is this fact which proves to us that these ideas really worked in magic. They understood by , by nature, the idea of a hidden essence and their magical water which produced gold. The idea behind these formulas-one which the alchemists never tried to conceal-involves a substance which acts on other substances by virtue of its properties, whatever their mode of action. This action is a sympathetic one or may be produced between sympathetic substances. It can be expressed in the following way: like acts on like; and we should add, along with our alchemists, that like attracts like, or that like dominates like . They say that this is because you cannot act on the whole with the whole. Since nature
is disguised in forms, there has to be a convenient relationship between the ???, that is, the forms of the objects which act upon one another. Thus, when they say 'nature triumphs over nature' they mean that there are objects which have a relationship of such close dependence that they are fatally attracted to each other. It is from this point of view that the nature of the destructive element is to be envisaged. In fact, it is an element which dissociates things, which uses its influence to destroy unstable components, and as a result brings about new phenomena and new forms, attracting to itself those identical and stable elements which they contain.
Have we here a general notion of magic, rather than merely a particular notion associated with one branch of Greek magic? Most probably the alchemists did not invent it. It is found in philosophy and we see it applied in medicine. It also seems to have functioned in Hindu medicine. Whatever the case, it matters little whether the idea was ever consciously expressed elsewhere. It is clear that these abstract representations of similarity, contiguity and opposition are inseparable from ideas of things, natures or properties which are transmissible from one being or object to another. This is all we wish to deduce from the facts. It is also true that these properties and forms resemble different rungs of a ladder, which must be scaled before one can act on nature, and that the inventions of the magician are not made freely and that his methods of action are essentially limited.
2 Concrete impersonal representations Magical thinking cannot, therefore, thrive on abstraction. We have clearly seen that when the alchemists spoke of nature in general, they were referring to a very special kind of nature. For them it was not a pure idea, covering all the laws of sympathy, but a very distinct representation of effective properties. This brings up the topic of those concrete personal representations which are known as properties or qualities. Magical rites can be explained much less clearly
through the application of abstract laws than through the transfer of properties whose actions and reactions are known before-hand. Rites of contiguity, by definition, involve the simple transmission of properties. A child who does not speak receives the talkativeness of a parrot; a person with toothache is given the hardness of rodents' teeth. Rites of opposition are not more than struggles between properties of a similar kind, appertaining to different species. Fire is the correct opposite of water and for this reason it can drive away rain. Finally, rites of similarity are such only because they can be reduced, in a manner of speaking, to the sole and absorbing contemplation of a single property: a magician's fire reproduces the sun, because the sun is fire.
But this idea of properties is both a very clear one and a very obscure one at the same time-a fact which applies to all magical and religious ideas. In magic and religion the individual does not reason, or if he does his reasoning is unconscious. Just as he has no need to reflect on the structure of his rite in order to practise it, or to understand the nature of his prayers and sacrifice, so he has no need to justify his ritual logically, nor does he worry about the whys and wherefores of the properties he employs, caring very little to justify in a rational manner the choice and use of his materials. We are sometimes able to retrace the secret pathway of his ideas, but he himself is usually incapable of it. In his mind he has only the vaguest idea of a possible action, for which tradition furnishes him with a ready-made means, yet he has an extraordinarily precise idea of the end he wishes to achieve. When he recommends that a woman in child-labour should not let flies buzz around her for fear she should have a daughter, it is believed that flies are endowed with sexual properties whose effects must be warded off. When the cream jug is thrown out of the room in order to bring good weather, the jug is endowed with properties of a certain kind. However, there is no attempt to trace back the chain of associated ideas by which the originators of these rites arrived at their notions.
These kinds of representations are perhaps the most important concrete impersonal representations in magic. The widespread use of amulets attests their extension. A good deal of magical ritual is concerned with manufacturing amulets which, once they have been ritually medicated, can be used without rite. Moreover, a certain type of amulet is made of substances and elements whose appropriation may not have necessitated any ritual. This is true of precious stones, diamonds, pearls, etc., to which magical properties are attributed. However, whether an amulet derives its virtues from ritual or from intrinsic qualities of the material itself, it is fairly certain that when it is used the owner clearly considers only its permanent attribute.