Levi-Strauss and the Structural Analysis of Myths

According to Levi-Strauss, there is no fundamental, categorical differ­ence between myth and history, and many narratives we encounter move between these two. Yet, myth and history are different, as the first is a closed narrative that will not change in its basic structure no matter who tells the myth and how it is told.

According to Claude Levi-Strauss, the aim of a myth is ‘to reach by the shortest possible means a general understanding of the universe—and not only a general but a total understanding’ (Levi-Strauss 1979, 17). This is the reason why myths are foundational. The understanding of the world, of the personalities that populate that world, of their relations and interactions—in short, the message myths convey—is meant to be all-encompassing and total, hence, excluding and even tabooing any rival understanding.

Levi-Strauss further argues that myths always reflect a binomial struc­ture of understanding the world and that this is reflected in the narrative figures, the metaphors, and the narration of the myth. He exemplifies this with a myth from western Canada in which a skate convinces the South Wind not to blow every day but only every second day. The binomial structure is not only apparent in the encounter of two, the fish and the wind, but also in the shape of the skate, which is large when seen from above or below and thin when seen from the side, and in the switch from the wind blowing every day to every other day.

Levi-Strauss’ elaboration of structuralism as a theoretical interpreta­tion of the world and as method of analysis makes him the revolution- iser of social sciences, notably of ethnography, his original field of study. However, his innovative approach to analysing so-called primitive societies reflected epistemological and ontological advances of the early twentieth century, like Georges Dumezil’s comparative linguistic approach to the analysis of myths (Segal 1996). The central epistemological proposition of structuralism is that human activity and thought follow a systematic logic; that this systematic logic can be identified in repetitions and parallelisms; and that one can distinguish two levels of structural development, one apparent (syntagmatic) and one symbolically underlying (paradigmatic). All three propositions had been developed before Levi-Strauss systema­tised them for the analysis of Amerindian myths. Levi-Strauss’ originality lay in consistently and unconditionally applying these three methodologi­cal innovations to ethnography and there particularly to all domains of human life, whether family relationships or myths.

Ferdinand de Saussure, whose ‘general linguistics’ will be discussed in more detail below, and his disciple Roman Jakobson particularly influ­enced Levi-Strauss’ mythology through their analytical proposition of two axis, the syntagmatic and paradigmatic axes, which both create the poetic dimension of a text through their interplay in the narration. Levi- Strauss was particularly interested in the proposition that the interplay of the two axes creates emotional, musical, and subliminal effects, which are not apparent in the flow of the narration itself or in its subject. Like other structuralists of the time, Levi-Strauss argued that texts (written or oral) do not represent any superfluous element; anything in the text is there by some inherent necessity, even if this necessity escapes the superficial look of the observer.

Levi-Strauss succinctly observes: ‘Mythical stories are, or seem, arbi­trary, meaningless, absurd...’ (Levi-Strauss 1979, 3).

Myths have to be believed to be true. They therefore appear absurd only to those who do not believe in them; commonly, they therefore include also a subordinate tale detailing the sanctions that threaten who­ever doubts the truthfulness of the tale. From the inside, myths are right and, obviously, sensible stories even though the audience might well know and understand that not all has factually happened in exactly that way. Doniger relates that Sudanese storytellers will begin with this formula (Doniger 2011, 2):

I’m going to tell a story

(Audience) Right!

It’s a lie.

Right!

But not everything is false!

Right!

The audience commonly knows and acknowledges that myths are not necessarily factually true. Whether myths are factually correct or not is therefore not a criterion by which their meaning can be understood or even their quality as myth could be debated. This means that myths can­not be ‘falsified’, and that is one of the major reasons why myths persist even if historical or social science research exposes their errors, flaws, and lies.

If myths are therefore not ‘falsifiable’ because they cannot be compared to a real world, how can myths be analysed and how can we know their impact on how societies are organised and identify themselves? The start­ing point of analysing myths is to ask what it is in them that is believed.

In order to identify what exactly is important in myths and how they persist and work, the analysis has to deconstruct the myth and recon­struct its context of meaning.

This poses an epistemological problem, for the object of analysis is partly constructed by the observation itself and, vice-versa, the observer is part of the object of analysis. As Patrick Jackson in his recent study of the ontology and epistemology of international relations formulated, the perspective to adopt is one of ‘world-mind monism’ in assuming that myths and our (in this case, my) interpretation of them are inher­ent to the same socially constructed mental world (Jackson 2011, 115). This approach is commonly subsumed under the notion of ‘reflexivist’ approaches and explicitly rejects the causal-mechanism modelling of so- called positivist approaches to political science and international rela­tions. However, as Jackson points out, the simple rejection of positivism and the claim that reflexivity of the observer is an epistemological neces­sity is not sufficient to disarm the argument about a lack of proof and scientificity. This needs to be done by setting out clearly the methodol­ogy of such a reflexivist approach (Jackson 2011, 186). Levi-Strauss’ structuralism is, by its systematic and rigorous design, a methodology that does exactly this.

The analysis of latent content is not arbitrary but based on structural principles. Levi-Strauss himself condemned the arbitrary interpretation of myths, notably those referring to psychoanalytical explanations such as repressed fears, as ‘too easy’ and unhelpful as they are based on ‘clever dialectics’ rather than systematic analysis (Levi-Strauss 1955, 429). The first principle is the claim that all elements, what Levi-Strauss calls with a self-created neologism mythemes, which make up a narration, entertain a relation to each other.

Saussure formulated linguistics as social science, namely that the exami­nation of language without consideration of its ‘social side’ is inconceivable:

But what is language? It is not to be confused with human speech (lan­guage), of which it is only a definite part, though certainly an essential one. It is both a social product of the faculty of speech and a collection of neces­sary conventions that have been adopted by a social body to permit indi­viduals to exercise that faculty (de Saussure 2011, 9).

For Saussure, the particularity of human language is not that it is oral speech; any other organ could have replaced language, for example, hand symbols. The particularity of language is that it is an oral system of symbols, which are socially agreed upon: ‘a system of distinct signs corresponding to distinct ideas.’ (de Saussure 2011, 9). Language only becomes more than sounds by the fact that ‘an auditory image becomes associated with a concept’ (de Saussure 2011, 9). These concepts are cre­ated and modified outside the individual as ‘[the concept] exists only by virtue of a sort of contract signed by the members of the community’ (de Saussure 2011, 12).

The social nature of language determines the relationships between the concept (which he calls signifiie) and the sound-image with which it is asso­ciated (which he calls signifiant). These relationships can be constructed through semantics (for example, prepositions), syntactically (for example, the difference between a car track and a track car), or grammatically (for example, infinite and definite verbs). Saussure further distinguishes syn­tagmatic terms from paradigmatic terms, the former being terms that link the preceding and succeeding term, the latter being terms that can be replaced, as such, by others.

He therefore conceives of two types of relational patterns that make words carry meaning and transform language into speech (parole): in a linear form, syntagmatic terms diachronically initiate meaning of one term after the other; in a deep form, paradigmatic terms imply meanings by referring to other terms that could be used or that should be thought synchronically with this term.

Levi-Strauss draws on Saussure’s linguistics and develops them further by arguing that every longer language fact—narratives—can be broken down into mythemes, which are also ordered following a specific struc­ture. Signifiant and signifiie, that is, expression and meaning, are not only related to each other within one sentence but also within one narrative unit. He illustrates this by taking apart the Oedipus myth and rearrang­ing the m ythemes according to their similarities, and he ends up with four columns of similar mythemes for which he then, in a second step, formu­lates binary relationships (Levi-Strauss 1955, 428-444).

These relationships between groups of mythemes follow the logic enun­ciated by the mythemes. For instance, the logic of the relationship of the group of mythemes in the Oedipus myth that deal with the killing of a kin are logically related to the group of mythemes which deal with the venera­tion or love of a kin. Neither group of mythemes appears arbitrarily in a myth but by structural necessity: it is, in Levi-Strauss’ terms, impossible to have a myth speaking of patricide (to stay within the Oedipus myth) with­out also talking of mother love. In fact, the killing of a man only becomes patricide because there is also the love of a woman who is simultaneously mother of the killer and wife of the man killed (Barthes 1966, 5).

It is here, in this argument of a necessary logical relationship between elements, that Levi-Strauss’ structuralism has been most effectively criti­cised as he, indeed, does not venture into the epistemological question about the origins of the necessary logic. The critique has been formulated from two perspectives. On the one hand, Levi-Strauss’ structuralism has been severely criticised from a mind-world dualist perspective as interpret­ing something into texts which is not there or which, in the end, cannot be proven to be there. The debate between Levi-Strauss and Propp already points in this direction, as Propp rejected the use Levi-Strauss made of his structural analysis, at least as the debate is summarised by Alan Dundes, who writes: ‘Propp is concerned with empirically observable sequential structures whereas Levi-Strauss is interested in underlying paradigms’ (Dundes 1997, 43).

The accusation is that any assumption of a paradigmatic structure underlying a narration is nothing but an interpretation. This statement only makes sense if Propp is assumed to be ‘objectively’ looking at tales from an outsider perspective, and the statement also only gains polemi­cal weight if ‘being interested in underlying paradigms’ implies not being interested in the apparent story. The critique does not forcibly denounce that narratives have underlying and subliminal meanings, but it consid­ers that the narrative cannot provide material to grasp this implicit sense. Arguing that words are sound and fury, these critics maintain that research­ers lack objective principles and rules to analyse subliminal meanings, and therefore cannot produce scientific knowledge.