mm
THREE
The Textual Structure of Performance
islliiiii..
W;^' -yry: I
/'■■^■^■^•■vr-*
The work of art aims at revealing essential or important elements, or important ideas, in a clearer and more concrete way than real objects are capable of revealing them. It succeeds in this by employing a set of interconnected parts, systematically changing the relationships between these parts.
Hippolyte Tame, The Philosophy of Art
Signification occurs when we submit the data of the world to a "coherent deformation."
Maurice Merleau-Ponty, Signs
3.1, MULTIPLE SYSTEMS AND SINGLE SYSTEMS
In the last chapter I reviewed some of the important characteristics of the performance text ("ephemeral" presence, lack of persistence, heterogeneity- of expressive media, multiplicity of codes, mmticUmenSionality) by virtue of which it is now perhaps possible to locate the performance text within the general category of semiotic texts. Up to this poult I have concentrated on the physical, visible level of performance, and will now shift my attention to the formal-systemic level, where We can locate series of theoretical entities "constructed" by the analytical-interpretive process.
The most important of these entities is the textual structure of performance, a system which emerges from the combination of the various codes of the performance text, and which assures the coherence Of the code^ relations.1 My conception of the textual structure of performance Coin-cides substantially with the concept of the "textual system" elaborated by Metz in the context of the cinematic text. It is expressed in the following definition:
The system has no physical existence; it is nothing more than a logic, a prin-
- ■■■ ;i/;;myym^y
y:\mmy,.
.ASM
't
fag-
84
The Semiotics of Performance
dple of coherence. It is the intelligibility of the text: that which must be presupposed if the text is to be comprehensible.2 (Metz 1971: ET 75-76)
Thus the basic characteristic feature of the textual structure of performance is its systematic quality. As we shall see, the performance code also belongs to the systematic level but it cannot be confused with the textual structure of performance, of which it is merely a component, because it does not possess the second characteristic of the textual structure: uniqueness. To claim that the textual structure of a given performance is "unique" is to maintain that as a particular combination of codes, it belongs, by definition, only to that performance and to that performance alone, thus guaranteeing the individuality/unrepeatability of the performance on the structural level. However,, the same cannot be said of the codes that make up the textual structure of performance: a code is, by definition, "a system which is valid for several texts/' that is, a multiple system (Metz 1971: ET 76). As we shall see in chapter 4, a particular gestural, in-tonational, or mime code may appear in different performance texts (as well as in extra-theatrica] texts). This means it can enter into different combinations with different codes. On the contrary, the textual structure of a performance such as, for example, the complete network of code relations underlying Luca Ronconi's Oresteia (1972), or, more precisely, the network underlying a single performance of the Ronconi production, belongs to this performance alone (if one concedes that it is indeed possible to "reconstitute" it), and cannot be found—in the same may, or in its entirety—in any other performance.3
According to Metz, who has the merit of having developed the distinction between "singular systems" (textual structures, in the present case) and "multiple systems" (codes), this cUstinction not only holds true for the field of performance but may constitute a general semiotic principle of great importance.
[This distinction] seems to us to be of great importance for any structural analysis. ... It comes down to saying that it is the peculiarity of certain structures that they underlie entire series of events while concerning none of these events in particular (thus the code of a language is present in every sentence, narrative codes in every narrative, the typographical code in every printed page, etc.), while other structures are linked from the very beginning to unique events which they characterize and which are so to speak by definition unfit to be used again, at least not in exactly the same way. Such is the structure of a sonnet (not the sonnet form) or of a sonata (not the sonata form). (ET 76)
Two further characteristics of the textual structure of performance must be kept in mind. The first is the fact that the textual structure of performance performs a single, unique combination, i.e., it combines in an unrepeatable way codes that are variously specific and codes that are variously nonspecific at the same time (Metz uses the word mixite in this
The Textual Structure of Performance
85
context). Obviously, this characteristic is also common to the performance text as a concrete discursive event.
Finally, we must consider the dynamic quality of the textual structure of performance, which almost never limits itself to assembling the codes in a static manner, but combines them instead in an original way, and in the process of relating them to each other ■"restructures"' and "transforms" them.
3.2. DEGREES OF DYNAMISM IN THE TEXTUAL STRUCTURE OF PERFORMANCE
The textual structure of a performance is therefore something more than simply the sum of the codes that create it. This "something more" is actually the result of the process of restructuring, or deplacement (Metz), that it exerts on the codes.4 The distinction that I will draw in 4.6. between "performance codes" in the strict sense and "theatrical conventions," as well as my definition of the role played by the conventions, will enable me to confirm my current assertion regarding the dynamic quality of the textual structure of performance. It will also allow me to define Metz's rather ineffable deplacement of the codes within a given performance in more concrete terms as a deviation from their extra-theatrical usage, as the consequences of the action of the "theatrical conventions" of that performance.5
Leaving out the kinds of conventions that are dominant in a particular performance text, and hence their effect on the extra-theatrical codes (see 4.7.), we can already see that the degree to which the performance codes in the performance text in question deviate from their extra-theatrical use depends, quite predictably, on the extent to which the textual structure is capable of dynamically integrating these performance codes. Let us consider, for example, the theatrical productions of the historical avant-garde, and especially the movement generally known as "director's theater." From about the middle of the last century when the director's role began to assert itself as an aesthetic principle rather than simply as empirical practice, directors have explicitly aimed at creating a mise-en-scene understood as a unified work of art, created from the fusion of various expressive media (words, music, dance, painting) to form an indivisible, homogeneous whole, where the basic elements undergo a more or less radical transformation. Wagner's Gesamtkunstxoerk constitutes the most explicit and extreme theorization in this direction. As for examples of the same phenomenon on the level of praxis, I should at least mention (apart from the work of Craig and Appia, both theorists rather than practitioners) the outstanding period in Russian theater during and after the revolution (Mejerchol'd, Tairov, Evreinov, Vachtangov), the productions of Piscator and Brecht, the experiments of Gropius, Schlemmer, and Moholy-Nagy at the Bauhaus, the important theatrical innovators in France (Copeau, Dullin, Artaud, Bar-
86
The Semiotics of Performance
rault), right down to the current (but rarely exciting) practice of director's theater.6
In other cases, however, the disparity between the performance code and the extra-theatrical code may be very slight, and sometimes may not even seem to exist. Apart from the (fundamental) role played by the type of theatrical conventions used, this situation is generally linked to a (significant) reduction in the dynamic power proper to the textual structure of performance. This reduction means that a diagram of the code relations would take on the shape of a simple juxtaposition rather than a hierarchical interrelationship. Even in this case there are many examples. To begin, we could consider the widely independent status enjoyed paratactically by the partial texts that form the basis of a large number of festivals in the Western tradition, festivals that range from aristocratic and courtly celebrations of the Renaissance and baroque periods, to folk festivals past and present. Instrumental music, songs, poetry, dances, costumes, emblems, and images are blended together in the festivals. These elements originate before and continue after the festival itself, and are to a great extent autonomous and independent. In such cases, the degree of integration and restructuring of the codes in the performance tends to approach zero (without ever quite reaching zero, as we shall see later).
The Renaissance festival provides an example of this phenomenon, and is described by the most up-to-date theatrical historiography as a "set of autonomous forms of expression in their highest state of development" (Cruciani 1972: 9), or as a form of "architecture in time, catalyzing the diverse elements that appear within it, elements that are among the most distinguished in various art forms, customs, and activities" (Cruciani and Taviani 1980: 35). One of these forms of expression, or highly distinguished elements (in fact, the most important of them), is comedy, which maintains a relationship of mutual legitimation with the festival:
Classical comedy appears in the context of the festival not because it is particularly entertaining (indeed it is often experienced by witnesses as one of the most boring moments), but because it is an outstanding example of the new culture's rediscovery of antiquity. Thus on the one hand the festival, as a form of architecture in time, gives meaning to the presence of the comedy, while on the other hand the comedy gives its "blessing" to the festival, by representing a link with a body of artistic work based in an era that transcends the contemporary. (1980: 35)7
In Renaissance productions of ancient and contemporary comedy, various partial texts with their relative codes are often endowed with considerable autonomy, and are not intended to function as an illustration of the content of the texts. This can be seen, for example, in the costumes, and especially in the stage scenery, which was' allowed a very narrow range of rigidly preestablished possibilities (comic, tragic, and satirical scenes). The aim of the scenery was not to visualize the locations of the dramatic action,
The Textual Structure of Performance
87
but rather to offer an image of a more or less idealized city (Ruffini 1978a:
164).8
In our own times we can find very obvious examples of performances where the partial texts and codes are integrated with scarcely any dynamic intensity (with a consequently insignificant diplacement of the codes) in the many-faceted movement of experimental theater which began with "happenings" and pop art in the 1950s and 1960s, evolved through the 1968 experiences of street theater and improvisation, and continues today with the most recent offerings of the so-called post-avant-garde and trans-avant-garde (performance art, conceptual theater, and the like). Despite the great diversity of their poetics and techniques, all of these phenomena have two features in common. First, they all create a conglomeration of different "elements/' where incoherence and (perhaps only superficial) discontinuity are frequently emphasized in a provocative way9 (for example, the already mentioned "compartmental structure" of "happenings" and environmental performances by Kaprow, Oldenburg, and Whitman, but also the "pieces" created by Robert Wilson and subsequently rearranged in different ways for his new performances). Following the example of Duch-amp's technique of the "ready-made," the second feature that the avant-garde experiments have in common is the use of objects, images, gestures, and forms of behavior taken "as they are" from daily life and assigned new communicative intentionalities, as well as strong "aesthetic" connotations, thanks to the distancing procedure of decontextualization/ recontextualization with which they are presented. The aesthetic process of display and staging constitutes them as signifiers available for numerous new sign-correlations. To give just two examples, I will mention Rauschen-berg's assemblages in the late 1950s and Oldenburg's sensational Autobodys in 1963.
Regarding the distinctions I articulated on the dynamic quality of the textual structure of performance, it is useful to recall the opposition that Ruffini (1978a) observes between "mise-en-scene" and "placing on stage," based on the binary opposition of functionaHty/nonfunctionality. According to Ruffini, we can speak of "placing something on stage" when "the text transferred simply ends up juxtaposed with other partial texts without significant amounts of reciprocal interference" (1978a: 130). Ruifini cites the following as examples of "placing on stage": the newscast, the ceremonial, and some Renaissance comedies. On the other hand, he perceives the presence of "mise-en-scene" when "the transcriptive codes are functional-ized to serve the content of the text," as in "director's theater," or in "public speaking." Ruffini's hypothesis also makes the already mentioned connection between the extent to which the codes of the textual structure of performance are integrated and the extent to which they deviate from extra-theatrical usage. "While adherence to different norms is the basis of -placing something on stage, in the case of mise-en-scene, by contrast, there is a sense of breaking all cliches" (131-32). Obviously, the contrast that Ruffini draws is meant to be understood and used only as a theoretical polar-
88
The Semiotics of Performance
ity, and concrete examples must be located along the intermediate points of an infiriite continuum. In fact I believe that no real examples of "zero-degree" integration-deviation can be found. Even in the extreme cases of "happenings" and performance art there is usually a new context to "modify" the real, everyday element I have mentioned, preventing it from being perceived in a completely automatic way. The act of ostension always makes some of the concrete traits pertinent at the expense of others, by constituting the object as a functive capable of becoming the vehicle of new meanings.10 We should also bear in mind that the Renaissance festival {another of Ruffinf s examples of "placing on stage") assembles the autonomous expressive forms that come together within it into a "different," "detached" time-space, which is the ideál time-space of celebration, thus modifying the semantic valences (and in particular the ideological valences) of the elements used with a kind of Metziah déplace-ment:
What qualifies and unifies the various forms of the festival's expression is their being positioned as celebration. This celebration is not only the superficial occasion, the honoring of some person or some moment of the year; it has instead the much weightier value that I have mentioned: it is the ideal time for the society to take a look at itself in its "eternal" dimension, it projects itself into its own Utopia . . .: it is the "well-regulated city." (Cruciani 1972:11)
Although Cruciani also insists a great deal on the autonomy of the different forms of expression that converge in the festive event as a whole, he mentions the festival several times as a "complex system" or as a "structuring'' and "formalizing" unit which does not limit itself to the paratactic assembly of elements from different sources but which—within limits-reorganizes these elements, redefining their function and meaning as they are brought together.11
3.3. PARTIAL STRUCTURES AND MACROSTRUCTURES
Up to this point I have alluded to the textual structure of performance in the singular, as if each performance has only one structure. There are nevertheless two kinds of completely different though compatible considerations that lead me to postulate a (limited) multiplicity of the textual structures of every performance text.
The first consideration is of an objective nature and concerns the already mentioned macro-textual character of the performance text, the fact that it really consists of the interweaving or juxtaposition of various partial texts, each one of which can be in turn endowed (and generally is endowed) with an already complex system of codes; and hence with a textual structure (i.e., a partial textual structure). For example, one can easily suppose that the scenographic text of a stage set in a baroque theater was organized by a perspective code, an iconographic code, a proxemic
The Textual Structure of Performance
89
code, as well as by various conventions of genre (see 4.6.2.). The homology between the performance text and the textual structure of performance is perfect in this case also. Just as the performance text is in reality a macro-text (as we have seen), the entity that I called the textual structure of performance represents in fact a structure of structures, a macrostruc-ture (naturally not in van Dijk's semantic application), i.e., "a complex system made of reciprocal mterweavings, interrelations, and inscriptions of specific systems which are in themselves already complex" (Bettetini 1977: 26).
There is nevertheless a problem with the hypothesis that several partial textual structures are manifest within the performance text and organized into a complex macrostructure. If the textual structure of a performance is "unique" as I have claimed, that is, if it actually belongs to one performance text and only to one, what is the situation of the partial textual structures that constitute it? Are these similarly unique or not? And, if they are not, can they be considered textual structures in the proper sense, since they lack a characteristic that we have seen to be constitutive of textual structures? The dilemma is not easily resolved, and perhaps cannot be resolved once and for all. Only several concrete analyses carried out according to the theoretical framework that I am in the process of delineating will succeed in resolving this and other problems that remain open at the moment. However, in a hypothetical and purely provisional way, let us imagine the possibility superficially suggested by several factors that "uniqueness" or "singularity" is not one of the properties of partial textual structures. This is equivalent to supposing, for example, that the multi-coded system underlying the gestural text of a given performance can appear basically unchanged in the gestural texts of other performances. It means, in other words, that two or more performances can present identical partial texts (in the obvious sense of being structurally identical).12
This assumption seems easy to verify in performances belonging to genres that are strongly coded and hence highly standardized. Let us take, for example, the gestural codes in Sicilian puppet theater (Pasqualino's 3978 study distinguishes four types of movement made by these puppets), noting that the gestures remain almost unchanged from one performance to the next, at least within the various schools and traditions (it is widely known that the movements of puppets in the Palermo tradition are not the same as the movements of puppets in Naples or Catania). Or we could consider various forms of repertory theater, also rigidly defined, from tnudra to Indian classical dance, and from the over-coded kinesis of nineteenth<entury opera to academic ballet. As for the scenographic text, we can find famous examples of structural invariability in the fixed scenery of early Renaissance theater, and, in a completely different area, in the setting for Noh performances (platform/rear plarforrn/bridge, three pines, musicians). But examples of code groups with multiple manifestations can also readily be found outside traditional theater. Let us take,
90
The Semiotics of Performance
for example, the "hypernaturalistic" techniques of physical expression used by Ryszard Cieslak, Jerzy Grotowski's leading actor. We can see that these techniques recur without major changes along with their relative code system (as also happens in the case of other actors: Z. Molik, Z. Cynkutis, R. Mirecka, and others) in Akropolis (1962), The Constant Prince (1965), and Apocalypsis cum figuris (1968). This is also true of Robert Wilson's iconography from Deafman's Glance (1971) to Einstein (1976) and Edison (1979), This iconography continues to be based throughout on the same visual encyclopedia and on unchanging image systems (surrealism + hyperrealism, all filtered through stereotypes of the "average" American imaginary). Similarly, in the case of Luca Ronconi's theater, the actors' "alienated-epic" delivery (not to mention other blatant repetitions) reproduces in an almost completely unvarying manner through several performances (such as Marisa Fabbri in the already mentioned production of the Oresteia, and in The Bacchantes [1978]), an entire system of codes underlying the verbal, paralinguistic, and mimetic texts, relating to the use of voice, i.e., to tone, accent, volume, timbre, pauses, and rhythms.13
At this point it should be clear that the phenomenon under discussion is only an aspect of the broader phenomenon of theatrical interiextuality that will be examined in 5.4. It should be equally obvious that partial texts with common partial textual structures provide a particularly striking example of a class of performances that can be regarded in their entirety as a single performance text: a genre, an artistic movement, the work of an author, and so on. As we will see in the next chapter, the possibility of intertextual recurrences is guaranteed by a type of convention that I will describe as "particular. " This kind of convention possesses among other traits the ability to activate a specific competence in the addressee, with resulting expectations and predictions.
3.4, MULTIPLE INTERPRETATIONS, MULTIPLE STRUCTURES
The second order of considerations that prompts me to postulate multiple textual structures in a given performance is completely different from the first, but does not contradict it. The textual structure of performance, like the codes that it combines, does not constitute an empirical object, materially discernible on the surface of the performance text and concretely preexisting the analytic process. To quote Metz again, the textual structure of performance represents instead ''an ideal object constructed through analysis." The analytic process elaborates this theoretical construct on the basis of explicit and implied elements in the discursive event by bringing to light the codes and other productive conventions of the performance text, and by actualizing the network of their relationships. Hence even If the textual structure of performance is clearly constructed through 'the revelation of
7259
The Textual Structure of Performance
91
properties and elements immanent in the text, it is essentially a pragmatic entity. But we must now admit that different readings or analyses of the same performance (carried out from various viewpoints, with different presuppositions, on the basis of uneven intertextual and encyclopedic competencies, and with varying contextual aspects if we are dealing with repeat performances of a single production) can attribute to that performance textual structures that are at least partially different. This multiplicity, and hence the partial, provisional, and relative quality of the structures,14 is the result of the partial, provisional, and relative multiplicity of analytic-interpretive approaches.
Having said this, I must immediately add that although analyses and readings of the performance text, like those of any other text, can be conceived of in a theoretically lmlirnited or at least mdefinite number, I do not think that the same can be hypothesized, for the textual structures that can be attributed to a given performance. These must be considered finite, and indeed may be very few in number. To admit an infinity of structures in a text might lead, I think, to the capitulation of the entire semiotic enterprise, as well as probably constituting a contradiction in adjecto (insofar as it dissolves the very notion of the "text"). For my own part, I propose to distinguish between (a) pertinent interpretations, which, though varied, are capable of assigning a textual structure to the performance in question, assuring the cohesion of its code relations, and (b) non-pertinent interpretations, which do not intend (or are unable) to organize the formal systemic elements (codes) in the performance text in question into a coherent structure. It should be clear in this case that /non-pertinent/ does not mean /improper/ or /illegitimate/. I use the term in a purely objective way, excluding all negative connotations. On the other hand, I use /pertinent/ to refer to readings that assign a textual structure to the performance because (and to the extent that) these readings seem appropriate from the semiotic viewpoint from -which I have decided to investigate theatrical performance in this volume.15
Even in the case of the reception of the performance text, as in the case of other discursive occurrences, everything turns on the dialectic between fidelity and freedom, between the textual restraints that direct and guide the receiver and the receiver's discretionary possibilities. According to Eco:
The comprehension of an aesthetic text is based on a dialectic between an acceptance and repudiation of the sender's codes—on the one hand—and the introduction.and control.of the addressee's codes. . . . On the one hand, [the reader] is challenged by the ambiguity of the object; on the other hand, he is regulated by its contextual organization. (1975: 342-43)
When I speak of a limited (or at least always finite) multiplicity of pertinent interpretations 1 mean to emphasize this dialectic, attributing equal importance to both poles, without giving weight to one of them
92
The Semiotics of Performance
to the determent of the other or vice versa, as would happen in these hypothetical cases: (i) unlimited multiplicity, and (ii) a single reading or structure.
Going beyond these considerations of a general, pre-aesthetic type, however, the reasons for the limited multiplicity of the textual structures that can be associated with a theatrical performance can finally be articulated, anticipating some of the concepts that I will analyze in greater depth below. As we shall see in chapter 4, the performance text constitutes a type of aesthetic text. Therefore, in referring to the modifications in the use of extra-theatrical cultural codes and to the "invention" of new sign-functions implemented by performance, we can postulate the existence of a performance idiolect as an example of the "aesthetic idiolect" which Eco defines in order "to designate the rule governing all deviations at work in the text, the diagram which makes all deviations mutually functional" {1975: 339). We have thus found another, quite specific reason for the limited multiplicity of the textual structures linked to a performance: to qualify as such they take into account the rules of the idiolect governing the performance text in question. (This is another property of the textual structure of performance that must be added to the Hst in 3.1.). Completing the distinction proposed above, I would state more precisely that a reading/analysis/interpretation of a performance text is pertinent if, and only if, it is capable of assigning a textual structure to thai performance, and hence, of taking into account the rule(s) of the idiolect(s) governing it.
We might ask: what kinds of relationships exist between the textual structure of performance and the performance idiolect? These relationships are undoubtedly very close, and at the present stage of research I am not in a position to define them with greater precision. I could, however, offer some suppositions. For example, we could assume that we are dealing with the same theoretical entity considered from two different viewpoints or from two different aspects. More precisely, we might conceive of the performance idiolect as the semantic counterpart of the pragmatic entity constituted by the textual structure of performance, or as the invariable among a certain number of pertinent interpretive variables. In this case, the relationship would be similar to the relationship postulated between macrostructures and the topic in textual linguistics (Manetti and Violi
As another hypothesis, we might consider the idiolect as a pragmatic entity "produced" by analysis, consequently creating one idiolect for each textual structure that analysis attributes to theatrical performance.16 The difference between the idiolect and the textual structure of performance can now be articulated not as an opposition between the semantic and the pragmatic but rather as an opposition between deviation and norm, which Ecb dairns is fundamental in his comments on the category of the idiolect. On the basis of Eco's definition, if the performance-idiolect can indeed be conceptualized as the diagram of the code-deviations vrithin the performance text, it is legitimate to suppose that it does not contain
The Textual Structure of Performance
93
all the codes and conventions that constitute the textual structure of performance but only some of these, specifically the codes and conventions that represent and produce aesthetic deviations. I will use the term "distinctive" to describe these performance codes and theatrical conventions in chapter 4. Yet this second formulation is also entirely provisional and unsatisfying, not only because the classification of the codes and conventions of the performance text that I will propose later, and upon which such a hypothesis is founded, cannot be considered definitive, but especially because the deviational character applies exclusively to "distinctive" conventions and codes only on the aesthetic level (concerning the subversion of institutionalized norms of genre).17 It also becomes proper to both of the other kinds of codes and conventions ("general" and "particular") if we take into account the transgression of norms that regulate the everyday use of gestures, oral delivery, and the like.
I will continue my discussion of these questions in greater detail in the next chapter. For the moment, I will return to the textual structure of performance, and will add, in conclusion, returning to Eco's remarks on the aesthetic idiolect, that its critical characterization "is not as easy as its theoretical postulation." Taking the analogy further, we could say that the textual structure of performance is a "regulatory hypothesis," and attribute an aspect of "infinite approximation" to the analytic interpretive process that attempts to characterize it (Eco 339). It is important in both cases to emphasize the pragmatic characteristic of this theoretical construct, its partial and provisional nature as a working tool.
3.5. ANALYSIS/READING/CRITICISM
Up to this point I have used the terms /analysis/ and /reading/ (as well as /interpretation/) interchangeably to designate various approaches to the performance text. It is time to differentiate them, attributing a more precise meaning to each. From now on, the term /analysis/ will designate scientific analysis, or the explanatory, descriptive approach of the scholar, and especially of the semiotician. The term /reading/ on the other hand will indicate ordinary or naive analysis, meaning the audience member's reception during the course of theatrical communication. Though referring above all to an analytical approach in the narrow sense, the theoretical hypotheses formulated up to this moment (and the hypotheses that, will follow) also concern at least in part the ordinary fruition of performance. We must in fact remember that: (a) (textual) "analysis" counts "reading" among its goals; one of its aims is to reveal the rules and mechanisms of theatrical reception through appropriate models, discovering the modalities of textual cooperation "foreseen" by the performance text and inscribed within it;18 (b) analysis implies that the subject who engages in it has passed through a previous phase of reading.19
Thus, generally speaking, the pertinent/non-pertinent distinction intro-
94
The Semiotics of Performance
duced in 3.4. functions both for srientific analysis and ordinary reception. Yet in practice it seems very problematic (and also not very useful) to submit the mdefinite, if not unlimited, number of levels on which audience readings can be stratified to a close examination of this kind, because of their largely implied and unexpressed character. The spectator's interpretation is almost never committed to writing, except in completely atypical cases, and very often is not even verbalized. As Barthes has commented referring to the literary experience, reading is an unspoken and immediate "conferring of sense" (1966: 48).
To complete the picture of the differences existing between the two types of approach to theatrical performance represented by analysis and reading, we must finally consider the following circumstances. (1) Conditions of reception. Reading occurs during the course of a performance which imposes on it its rigid and unalterable temporality. Analysis on the other hand can benefit from repeated viewing by the spectator on various occasions, and it can also take advantage of literary and iconographic documentation on the performance. In optimal circumstances, which today are increasingly frequent, this would also include viewing a video recording. (2) Receptive competence. Reading is based on a largely nontheoretical theatrical competence, which is unexpressed and generally unsophisticated. Analysis, by contrast, presupposes by definition a theoretical competence (of a passive type, like that of the common spectator—according to a distinction that I will make at a later point) which is. explicit and conscious, and which performs among other things the task of rationalizing the spectator's implied competence.20
Lastly, along with (textual) analysis and reading we must also include a critical level located on an intermediate plane with respect to the analysis and reading, but which, like analysis, belongs to the wider field of specialized reception.21 In opposition to the increasingly widespread opinion of other scholars, I would in fact argue that within the field of specialized reception we must make a sharp cUstinction between a normative approach, proper to the author as well as the critic, i.e., individuals who are, according to Garroni, "concerned with artistic objects . . . producing them or in-tervenirig mdirectly in preferential choices that condition their production" {1972: 329) and an analytical approach in the strict sense, i.e., the descriptive/ explanatory approach proper to the critic, and especially to the semiotician, or the approach of individuals who "are concerned with artistic objects in order to analyze their semiotic structure"; in other terms, to explain their discursive function on the basis of adequate and explicit models. Naturally, if the basic distinction between the analyst-scientist and the critic is formally rigorous, this, for Garroni,
does not mean that [the distinction] is so absolute in concrete terms as to allow a clear and rigid separation between what constitutes the activity of the analyst and what constitutes the activity of the "critic" and producer, cutting one off from the other in a definitive way. (329)
The Textual Structure of Performance
95
In particular, it seems that the critic is located in an intermediate position between the plane of artistic production and the plane of the scientific analysis of that production, leaning at some moments toward the former (the "artisf'-critic, Oscar Wilde) and at other moments toward the latter (the "scienu^f-critic). Yet regardless of how many oscillations occur in practice, and regardless of how many redefinitions are offered in theory (for example, by invoking more-explicit scientific protocols), the critical approach maintains a distinct identity and specificity, linked to the normative-dogmatic moment of choice (choice of sense, a la Barthes),22 on the one hand, and/or attribution of aesthetic value on the other, which is constitu-tively proper to criticism allowing it to be seen as formally completely distinct from the scientific-analytic approach.23
It should be clear by now that this study is concerned with the analytical approach to performance, while aiming at the construction of a textual model that would also allow us to shed light on the psychological and intellectual processes of ordinary theatrical reception. The introduction of the concepts of the performance text and the textual structure of performance enables us at this point to make sufficiently clear what constitutes both the operating space and the main objectives of a textual analysis of performance. These objectives can be summed up in the following way. Given a performance whose textual "absence" can be compensated for to some degree by using available documentation, one must decipher the codes and conventions underlying it in order to identify its complex textual structure, passing though a preliminary phase of description/transcription of the discursive surface.24
As we shall see in the next chapter, since performance can never be restored in its entirety to preexisting codes, and actually institutes its own signifying conventions ex novo in many cases, the construction of the textual structure of performance—the central, characteristic focus of the analytical approach—implies much more than the usual task of decoding, requiring abductions and other "interpretive hypotheses that function as tentative forms of a new coding" (Eco 1975: 183). In fact, all the modalities of inference seem to be at work in the analysis of the performance text, as of course in every other aesthetic text, where the well-known traits of ambiguity and self-reflexivity usually induce a particularly strong "abductional tension" in the interpreter (Eco 342).
At this point it will be clear that any type of approach to the performance text, whether a reading, an analysis, or a critical examination, is conceived of as an interpretive activity, endowed with a creative and a hypothetical characteristic, inherent in every process of consciousness and in every production of meaning, according to Peirce (1931-1958). The differences lie then in the procedures that are used, in the kinds of competence brought into play, and above all in the goals pursued. Above all, the critical interpretation of performance usually departs from a normative-dogmatic conception of theater, from a poetics of theater. In fact, even when based on more-general and appropriate paradigms, theatrical
96
The Semiotics of Performance
criticism tends toward a choice/production of determined meanings that can be attributed to the work in question for the purpose of its aesthetic evaluation.25 On the other hand, the analytic or scientific interpretation of the performance text uses a theoretical model which aims at the highest degree of generality (the performance text) in order to take into account the semiotic function of performance, describing the logic according to which it "generates" meanings and produces communication.
Having outlined their differences, I must add that these two types of interpretive procedures can be mutually corroborated. Above all, from the point of view of our present needs, textual analysis can and must take advantage of cues offered by critical readings, especially if these readings offer "examples of excellent textual cooperation" (Eco 1979: 184).26