Structural Linguistics and Anthropology

John Phillips

The Sick Rose

O rose, thou art sick!
The invisible worm
That flies in the night,
In the howling storm,


Has found out thy bed
Of crimson joy,
And his dark secret love
Does thy life destroy.

Whatever interpretation we put on (or "find out" in) "The Sick Rose" we can see that it will have been possible owing to analogical structures. Roses become sick because some germ or bug infects them. People become sick when some germ or bug infects them. By extension we might find that societies become sick when some germ or bug (evil intentions) infect them. Our thinking about all kinds of thing is infected too by structures and patterns that we find repeated in lots of different situations. The signified, that is, the meaning, of anything seems to come out of a pre-existing system the makes it possible and governs it. Structural analysis thus aims to "find out" the systems of thought that govern the ways we construct our world and interpret our experience. Structural analysis, however, as it was first set up, aimed to do this while remaining unaffected by social and/or cultural systems themselves. That is, they aimed for a purely scientific perspective that would not be governed or controlled by underlying structures. The most striking results in a field other than linguistics emerges with the work of the French anthropologist, Claude Levi-Strauss. He thought that linguistics was the first discipline among the humanities (or social sciences, as some parts of the humanities like to be known) to be established on purely scientific principles.

Necessary Laws

In Structural Anthropology Levi-Strauss writes: "for the first time, a social science is able to formulate necessary relationships." (SA 33). The most striking thing about that sentence is the characterization of the type of relationship structuralism is interested in. The laws it reaches for are the general ones (like scientific laws of gravity, say). Its laws must be necessary. A necessary law is a law that always applies whatever the situation is. The basic criteria for a scientific analysis are as follows: a scientific analysis must be real, it must simplify and it must explain. Simple explanations from complex data ought to be able to lead to correct predictions regarding actual situations of the type being analysed. Levi-Strauss took the example of structural linguistics and applied its method to the kind of relations that interest the anthropologist: kinship relations.

Kinship Relations

Way back in Ancient Greece Aristotle had made the famous statement "Anthropos is a political animal." He had also pointed out that kinship begins with two parts in need of each other, the man and the woman. Right up until the emergence of structural anthropology, the structures of kinship were regarded as being determined by much the same simple unit that Aristotle had suggested, that is, the family. For instance A. R. Radcliffe-Brown in the 1930ís had said that the unit of structure from which social relations are built up is the "elementary family" consisting of "a man and his wife and their child or children." The first order of relation that is built from this unit is constituted by three types of social relationship, according to Radcliffe-Brown: "that between parent and child, that between siblings, and that between husband and wife." Beyond this first order is a second order of social relationships, which connect different elementary units through a common member (uncles, aunts, cousins, etc.). From there any number, in principle uncountably many, orders can be extrapolated. It is such a simple explanation and fits so well with common sense that it lasted as a convincing, hardly questionable account of social relations for thousands of years. However, Levi-Straussís application of the principles of structural linguistics produced quite a different picture of the underlying laws that govern kinship relations.

Second Order First

On the basis of structural linguistics Levi-Strauss looked for the form of underlying relationships that could be said to apply to any group at all. That is, as a good scientist, he was looking for the "essence of human kinship." He found that what Radcliffe-Brown had called first order relationships (relationships within the elementary family group), were in fact a function of, and dependent upon, what Radcliffe-Brown had called second order relationships (those between families and involving a common member). In other words kinship relations do not derive from the family unit with Mom and Dad at its core. Rather they derive from systems that perpetuate themselves through specific forms of marriage. These specific forms, in terms of the given synchronic system, always precede the emergence of individual families. As Levi-Strauss says: "It is not the families (isolated terms) which are truly Ďelementary,í but, rather, the relations between those terms." There, he sounds just like Saussure talking about signs.

The Elementary Unit of Kinship

Levi-Strauss, armed with the structural method, felt that he should be able to derive simple principles from an overabundance of apparently contradictory empirical data. Take a number of different types of social organization and you will have different systems of behaviour and attitude. There is nothing to explain the differences. The Trobriand Islanders of Melanesia tend to exhibit warm, open and friendly relations between fathers and sons but marked antagonism between nephews and maternal uncles. The Cherkess of the Caucasus, on the other hand, tend to show hostility between fathers and sons but a marked tendency for the maternal uncle to help his nephew and to give generous gifts, etc. The former is characterised by a matrilinear descent, where the latter is characterised by patrilinear descent. But that is not the end of the story. In each case there corresponds a specific kind of relation between husbands and wives and brothers and sisters too. With the Trobriand the husbands and wives are intimate and open (like the fathers and sons) yet the relation between brother and sister is governed by a rigid and harsh taboo. On the contrary, in the Caucasus (where uncles and nephews have the special relationship), it is the brother and sister who tend to have reciprocal, warm and tender relations and the husband and wife who are joined antagonistically (the wifeís brother often has to jump in to protect her from a beating by her husband). A pattern thus emerges. This is Levi-Straussís formulation:

In both groups, the relation between maternal uncle and nephew is to the relation between brother and sister as the relation between father and son is to that between husband and wife. Thus if we know one pair of relations then it is always possible to infer the other. (SA 42)

So the uncleís relationship to the nephew becomes the key to the most elementary structure of kinship and is thus the key to understanding any social organisation that is governed by kinship. (Not all are. Modern post-industrial societies seem much more to be dominated by the logic of global commodity diversification, for instance, though the kinship relation is retained on the level of myth or ideology: the image of the family helps to sell cornflakes in a world where elementary family unity is actually less and less essential.) The formula, we should note immediately, is analogical. Analogy is a kind of logic that allows for the insertion of otherwise missing terms on the basis of presently known ones. The unknown can be derived from the known. If A is to B as C is to D and we know what A, B and C are then we can infer D with no trouble. The only test of such logic is empirical. Does it predict correctly? Levi-Strauss found that in this case it did.

Uncles with Attitude

The starting point for his discovery was attitude. The key relations, as we have seen, concern the brother and sister, the husband and wife and the nephew and uncle. The attitude of the nephew and the uncle towards each other turned out to be the key for understanding the elementary unit of kinship for all social organisations that he looked into. The elementary unit is thus the structure that rests upon the four terms: brother, sister, father, son. One of the remarkable things about this is that attitude was and still is often thought of as an unimportant effect of the terms of social organization reflected in the psychological dimension, rather than a sign of the actual systems themselves. If a boy is antagonistic towards his father perhaps it is a phase he is going through or something. Anyway children are always like that, it is their response to authority. Theyíll grow out of it when they become fathers and start tormenting their own sons. This is partly because attitudes so often flare up as a result of non-institutionalised, diffuse psychological factors. However, once these are discounted then one finds that attitude plays a fundamental role in the most elementary of institutional structures--the kinship relation. In other words what we often take to be the effects of individual emotional life turn out to be determined by structural laws that govern the organisation of society generally. This is indeed a major breakthrough in the early days of structuralism. Another significant gesture was Levi-Straussís turn towards the structural analysis of myth. Here we will see him adapting not only the structural method but also acting upon the strength of his findings regarding kinship. Most important is the analogical framework, which will allow him to suggest an ultimate equation. First, though, letís look at one or two objections that arise.

The Incest Taboo--woman as symbol of exchange.

The incest taboo, which for Levi-Strauss is universal (it applies in all societies), is composed of a law: In human societies a man must always obtain a woman from another man (a daughter or a sister). The structure of the law implies an avuncular role from the very beginning. The maternal uncle is part of the structure even before, and independently of the fact of there being, any actual maternal uncle. What this means is that the relationship between the sexes is not symmetrical. You donít get women exchanging men anywhere, though you do get fictions in some societies that represent this as being the case. The woman is always that which is given in relations of exchange between men and she is thus the symbol of exchange for a system that functions only to perpetuate itself. Another analogy emerges: woman is to nature as man is to culture. If culture is regarded as the consequence of systems of social organisation then woman stands symbolically for nature in a system of oppositions that seems to rise up spontaneously (remember the sick rose):




















We could go on. These are famous and obvious. It is structuralism that provides the means to analyse these systems so thoroughly. A number of scholars in the last thirty years or so have very thoroughly interrogated both Levi-Straussís findings as well as his own preconceptions. This remains important when reading his work. But we must remember the principles upon which he started. This is not an empirical discovery. It is based not on experience alone but on structural analysis: the analysis of the relations between terms rather than the terms themselves. We are analysing a symbolic system rather than actual relations between particular men and women. Sure, the system determines to a large extent the way those relations might go and provides insight into them (and has considerable support in cultural history). But as a symbolic system it remains open to unpredictable change, as we have observed, and it is the nature both of the unpredictable element (i.e., it cannot be subjected to calculation) and the ways in which it can help effect desired changes that have been the concern of a great many of the critical discourses of the past thirty years. If woman really is a sign whose meaning is the system that governs her status in exchange then her signified must remain open to a future that has in no way been completely determined (like the signified of any sign). In other words the feminine is the name for that which cannot be named. This formulation will come back to haunt us.


The Structural Analysis of Myth

Levi-Straussís method is encapsulated in a short yet justly famous article called "The Structural Study of Myth," which was also published in the Structural Anthropology collection. It is worth returning to this article because both the strengths and weaknesses of the structuralist project are rather beautifully revealed there. He provides a concrete example to start with by analysing the famous Oedipus myth of the ancient Greeks. We will have cause to return to this myth, but for now letís follow Levi-Strauss as he lays its meaning bare. His purpose is to illustrate his technique, of course, so no definitive analysis is intended--that will be reserved for the group of myths he is actually going to go on to analyse in full.

On the model of language, he supposes that myth is formed of separable components or segments, minimal units of signifying material. In the same way that a verbal sign is composed of phonemes (a written sign by graphemes) a myth will be composed of mythemes. He then points out that, like language, there is a non-temporal (synchronic) aspect just as there is a temporal one (on the level of parole). It is the work of the analyst to establish the structural conditions that underlie the myth--those that organise the mythemes. The events as they unfold in sequence in the story are treated as if they belong in columns so that repetitions of similar events or functions can be classed together independently of the chronological sequence. He uses an easy analogy:

Say, for instance, we were confronted with a sequence of the type: 1, 2, 4, 7, 8, 2, 3, 4, 6, 8, 1, 4, 5, 7, 8, 1, 2, 5, 7, 3, 4, 5, 6, 8 . . . , the assignment being to put all the 1ís together, all the 2ís, the 3ís, etc.; the result is a chart:










































He then demonstrates the operation by trying it on the Oedipus myth. The significant events, as he takes them, are as follows:

1.†††††††† Cadmos seeks his sister Europa, ravished by Zeus.

2.†††††††† Cadmos kills a dragon.

3.†††††††† The Spartoi fight and kill one another.

4.†††††††† Labdacos, Laiosí father comes in here and his name (meaning "lame") will be significant.

5.†††††††† Oedipus kills his father, Laios.

6.†††††††† Significantly Laios means "left-sided" (or lop-sided, limp on one side).

7.†††††††† Oedipus goes on to kill the Sphinx.

8.†††††††† Oedipus means "swollen foot."

9.†††††††† Then Oedipus marries his mother.

10.†††††† Later Oedipusí son Eteocles kills his brother Polynices.

11.†††††† And finally Antigone buries her brother, Polynices, against the prohibition of the state.


 The chart then looks like this:

Cadmos seeks his sister Europa, ravished by Zeus






Cadmos kills the dragon†††††††††††††††††††††††



The Spartoi kill one another






Labdacos = lame (?)


Oedipus kills his father


Laios††††††††††††††† ††††††††††† = left-sided (?)



Oedipus kills†††† ††††††††††† the sphinx





Oedipus = swollen-foot (?)

Oedipus marries his mother, Jocasta





Eteocles kills his brother



Antigone buries her brother, Polynices, de- spite prohibition





Each column is now regarded as being a unit and can be read from left to right as if it was an orchestral score (the music analogy will continue to be popular among structural analysts). In this way, Levi-Strauss tells us, we have the means of understanding the myth. This is his analysis of the first three columns:

The first column has as its common feature the overrating of blood relations. The second column expresses the same thing, but inverted: underrating of blood relations. The third column refers to monsters being slain.

Now it seems that the analysis has, like the study of kinship relations, a predictive power, that ought to be able to anticipate significance on the fourth column. The myth will provide its own context. One of the disputed aspects of the Oedipus myth concerns the fact that all the surnames in Oedipusí father-line have connotations that suggest difficulties in walking straight and standing upright. In mythological terms this relates to the third column in the following way:

Column three refers to monsters. The dragon is a chthonian being [chthon = "earth"; chthonian = "of the underworld"] which has to be killed in order that mankind be born from the Earth; the sphinx is a monster unwilling to permit men to live. The last unit relates to the first one, which has to do with the autochthonous origin of mankind. Since the monsters are overcome by men, we may thus say that the common feature of the third column is denial of the autochthonous origin of man. (215).

By autochthonous the anthropologist is referring to the belief that mankind is born from the Earth (literally, a native rooted in the Earth). Attacking and killing the chthonic monsters signifies a denial of mankindís autochthonous birth. The units in the fourth column, however, signify an affirmation of the belief in autochthonous birth, because of the persistent connotations of lameness and difficulty in walking upright. As anthropologists explain, in myth, chthonic people are generally represented as being lame. So the third and fourth columns represent a basic contradiction: (3) mankind is not autochthonous : (4) mankind is autochthonous. We know from the study of kinship what will happen next: (4) is to (3) as (1) is to (2). Thus two clear contradictions are rendered identical in analogy. As Levi-Strauss puts it:

The overrating of blood relations is to the underrating of blood relations as the attempt to escape autochthony is to the impossibility to succeed in it. (216)

This kind of correlation serves a purpose, according to the mythologist. It serves as a kind of tool that resolves contradictions on an ideological level. It displaces the original problem (born from one--the earth--or born from two--man and woman) onto a derivative one (born from different or born from the same). The myth thus maintains a cosmological "truth," according to which mankind is regarded as being rooted like a plant, say, but against all the evidence. The structural analysis of myth is, therefore, one example of the tendency among scientistic discourses to provide explanations of religion. This fact will get us into deep water as we proceed so keep it in mind. We will ultimately have cause to subject it to a similar analysis (but then when will it end?).


The Algorithm of Myth

At the end of the same article Levi-Strauss brings the structural analysis of myth to a speculative conclusion by devising a formula, as follows:

Fx (a): Fy (b) ~ Fx (b): F a-1 (y)

The symbols (a) and (b) are values and Fx and Fy functions. If, in the example of the Oedipus myth, (a) and (b) stand for types of kinship relation, then Fx stands for overvaluation while Fy stands for undervaluation. In the second part of the formula something interesting happens. In the second half of the equation (b) takes the place of (a) in the guise of the anti-autochthonous assertion and y, which now stands for the belief in autochthony, becomes a value in itself. (a), in turn, has become a function that surreptitiously links the kinship relation to the autochthonous belief. So the first two columns relate to each other in the same way as the third and fourth relate to each other. At the same time the fourth and the first relate to each other in the same way as the third and second (the autocthonous belief is to overvaluing kinship relations as the anti-autochthonous assertion is to undervaluing kinship relations). The contradiction is squared, thus preserving the mythical belief intact. We should now be able to see the extent of Levi-Straussís ambition in bringing scientific analysis to mythology. On one side you have the abstract simplicity of the algorhithm, on the other the complex ideological sleight-of-hand of the myth. The relation is equivalent to a distinction that Levi-Strauss draws out in his introduction to Pensťe Sauvage (The Savage Mind). The distinction is between the maker of mythic narratives (on the model of the witch-doctor), on the one hand, and the technician of modern civilised cultures, on the other. He calls the creator of stories a bricoleur and dubs his craft bricolage in strict opposition to technical science. He says it is, rather, a "science of the concrete." We now need to explore one further distinction in the structuralist understanding of language use.

Claude Levi-Strauss. Structural Anthropology Volumes 1&2. Harmondsworth: Penguin, 1968. 

My discussion of Roman Jakobson's distinction between metaphor and metonymy on the paradigmatic and syntagmatic axes, respectively, can be found here: aphasia.htm