Philosophy is not the passage from a confused world to a universe of closed signification. On the contrary, philosophy begins with the awareness of a world which consumes and destroys our established signification but also renews and purifies them. To say that self-sufficient thought always refers to a thought enmeshed in language is not to say that thought is alienated or that language cuts thought off from truth and certainty. We mist understand that language is not an impediment to consciousness and that there is no difference, for consciousness, between self-transcendence and self-expression. In its live and creative state, language is the gesture of renewal and recovery which unites me with myself and others. We must learn to reflect on consciouness in the hazards of language. M. Merleau-Ponty (1973: 17)

What We Have Done

We started this investigation with an assumption drawn from our common sense experience of the world: Schools are not neutral environments. They have a power over children. This power goes far beyond the teaching of the three R's and a few other such facts. This common sense assumption is also a psycho-sociological assumption of very long standing. More than sixty years ago, John Dewey, in Democracy and Education, made of it a fundamental principle of social life and the basis of a philosophy of education. As happens in such instances, it was then realized that the simplicity of the common sense intuition hides a process that is infinitely complex. It is easier to say that schools do something, than it is to say what it is that they do, and how they go about doing it. It is certain that they do it by being an environment to which people must respond in terms of it. But who made the environment? What preserves its order? How do we get to know enough about it to understand how it works and, perhaps, to control it?

This book has been an exploration of the school environment from a special point of view, one that I consider particularly important and rather neglected until now. With notable exceptions, few authors draw the full consequences of the facts that have constituted the theoretical framework of the preceding analysis:

1) human social cooperation proceeds through language and other symbolic systems, and necessarily through it (though not solely through it); 2) language is an environment to which one reacts, and over which one has, as an individual, but minimal control (even though a language is also a matter of individual competence); 3) knowledge about human beings, being a form of social cooperation, always proceeds through language both at the time of the investigation and at the time of the reporting of the investigation; 4) language, as a social artifact, has a form that is the product of a historical process and which can be described through instances of its use but also independently from these instances; 5) linguistic form, through its use in social cooperation, participates directly in shaping the experience of social life and the expression of this experience.*

Many are those who talk about the power of language. Most of them, particularly in the American tradition, rarely move on to examine it as a concrete, material object with a structure of its own. Language is often used as an example of a socio-environmental object. It is talked about as one aspect of this environment, as one of the means for cooperation alongside many others. It is commonly recognized that one has to go through language to reach what one is looking for, whether this is a psychological effect (what is the effect of this environment on an individual who has to live with it?) or a more sociological one (what is the organization of the environment?).

In the early years of modern social science, the postulates I have adopted were most seriously reflected upon by people like Sapir, Whorf, and perhaps Benedict, Krroeber and a few others. But their insights, when they have not been rejected, have been used more regularly as arguments for the development of theories of personality structure than as a call for the study of linguistic systems as such. The cause for this is probably to be found in the common reaction of American authors to the European structuralist theories of language. While certain Saussurian phrasings undoubtedly have to be changed, while Levi-Strauss sometimes overreached himself, the American rejection has been more basic. For many sociolinguists, ethnomethodologists or cultural anthropologists, to talk of language (or culture) as, in sane ways, independent either from the people who use it, or from the social situations of its use, leads to an unacceptable form of idealism, since it seems to place the locus of the structure outside any physical support.

These are complex philosophical and theoretical issues that I do not wish to discuss here. Let it just be said that it is not necessary to understand the structuralist intuition as necessitating the idealism that is lurking behind much Saussurian rhetoric. The origin of language is to be found in the interaction of human capacities with, and in, various environments, as John Dewey argued a long time ago (1916: 113-4). But this cannot be used as an excuse not to look at it for itself. Since there is a social aspect to language that is powerful enough to give a definite content to otherwise empty genetic capacities, it must be possible to describe this order.

What I have tried to do, in the first half of the book, is to give an account of a language based order at a level of abstraction that allows for a kind of structural generalization (to distinguish it from the kind of probabilistic generalizations that are allowed by statistical correlation research). This order is of central importance to the participants. It may be considered the order they live by, the order that interests us above all when we search for what makes the school a particular kind of place with a particular power over those who participate in it. While there are, undoubtedly, other orders by which the participants live (socio-biological ones, economic ones, etc.), the symbolic order has an encompassing power that transforms the effects of all the other powers. All human performances are mediated by language, even the silent ones, for they would not be "silent" if it was not considered possible that they might also not be. Language, then, is not to be considered, in any simple sense, "one of" the various cultural artifacts that we might want to list in an ad hoc fashion (along with, for example, family structure, religion, political organization, etc.). It is the superordinate structure that shapes all interactions within any of these orders, artifacts or domains.

This encompassing position of language explains the tendency of the actions of one people in various domains to appear, in some way, homologous, "integrated" as the early cultural anthropologists said. Thus only can we deal with the common intuition that schools replicate, in some ways, the general organization of America, and contribute to the reproduction of the culture's overall structure. In their specific organization, schools are extraordinary and unique environments within American society, especially if we compare them with families, churches, the various work places, etc. We could not explain the schools' power as transmitters of cultural patterns if we assumed that only that which is directly experienced in its specific form can be learned and then reproduced.

Conversely, to deal with language as a force that exercises itself in the encounter with the world, and in the communication of experience, is to preserve the independence of both the world and experience. It may be that—as Bateson said--the world and the mind have but one structure (1979). Before we can outline this unity, we must separate. What is striking about the people we listened to is their anxious search for appropriate expression. They are not machines more or less well programmed to spout out pre-scripted platitudes. They are alive in history. This is what I tried to capture in the second part of the book. Human history is not an entropic movement towards typicality or normality. It is a negentropic movement of creation in an ordered environment. It is a struggle with this environment without which there could be no struggle. Thus do we move.

To talk of this environmental order as a "linguistic" order is partially metaphorical if we think of "language" in the narrow sense only. Some, nowadays, prefer to talk of "semiotic system" to refer to the kind of matters I have dealt with. This is a way to emphasize that much more than words can be looked at from the point of view of their symbolic power. What is more important than a terminological dispute is the impact the label has on our understanding of the ways of knowing about the human world. For me, a linguistic/semiotic approach is one that is characterized by the direct attention that is paid to the forms that expression can take in a population (the "symbols"), and the relation between these forms, and between them and what is made with them ("what it all means"). Research done in these terms is not necessarily so different from the more traditional forms of monograph, except to the extent that it takes very seriously the fact that the ethnographic enterprise is, at all stages, one that is conducted through language and, ultimately, about language.

To talk of language in this context is thus not arbitrary. Furthermore, to talk of language rather than "semiotic system" or "system of symbols" may have the merit of reminding many of the concreteness of the object of study. An investigation into the "meaning"of symbols can easily take idealistic and mystical overtones. An investigation into the relationship between the form of symbols and the various situations in which these forms can be used can, however, remain within the boundaries of science: The object of study is exterior to the person who conducts the study and is, thereby, available to independent investigation for verification.

When this is set, it becomes possible to question in a new way the kind of knowledge such an investigation yields. I borrow the phrasing of this question from Clifford Geertz who, while starting from a different point than I, found it necessary to wonder too: "What do we claim when we claim that we know the semiotic means by which persons are defined to one another? That we know words or that we know minds?" (1976: 325). What do I claim when I claim to understand the organization of the language used by teachers and students in an American high school? That I know this language only, or that I have achieved a kind of existential entrance into the very being of these men and women?

Geertz does not give a very clear answer to this question. It may be that the question is not quite answerable. It remains a question that cannot be ignored if only because the non-professional audience is not bound by the caution that might be expected from a social scientist. The following pages consist of a commentary in lieu of an answer.

Writing the Results of a Structural Analysis

Doing an ethnography is a process in which a person (the ethnographer) experiences someone else's (the natives, the informants, the participants as I decided to refer to them here) symbolic performances and, subsequently, performs other types of symbolic performances using the participants' performances as a kind of pre-text on which to build new texts according to more or less explicit coding rules. In a first stage, the ethnographer writes fieldnotes, transcribes (in the literal sense of trans-scripting) interviews, looks at video tapes and draws posturings, etc. Later, and often in many stages, sometimes by himself, and then in reaction to various readings by friends, colleagues, professional editors, etc., a final text is produced. This text reflects the cultural position of the author(s) both through its own structuring and through its constructing of an ideal audience. The final text, in all cases, is the product of a human activity. It is an artifact, a literary construct. This text is never a mirror image of the original experience, nor is it ever fully transparent to it. As Merleau-Ponty showed (1973), language always mediates our encounter with the world even when, indeed particularly when, the text seems the most transparent to what it appears to be "about." I mentioned much earlier the fundamental point that the surrealist painter Magritte made graphically when he wrote beneath a drawing of a pipe "This is not a pipe." A monograph is only a monograph. It is not the life of the people it purports to describe. It is a possible representation of this life using "ethnography" as the medium. Other media could be used, other texts could be constructed. The participants themselves often construct such texts, and so do many different types of observers. Nothing of this can be dismissed.

I have attempted to take seriously the consequences of this situation in this book by doing several things that can be done given the inherent constraints of the medium I have at my disposal (i.e. an unillustrated book of not unlimited length). These constraints prevented me from giving readers the opportunity to judge the process that led me to choose the excerpts from the fieldnotes that I have analyzed inthe body of this work. The notes are several times the length of this book and cannot possibly be published. In any event, there is no way to display the preceding step, the step that led from the field experiences to the notes. What can be done is to display much more systematically than is usually done the process that leads from fieldnotes to analytic statements. I have quoted extensively from the notes themselves (by contrast with the summarizing rewrites normally used in most published monograph. I made explicit the kind of questions that I first asked of these fieldnotes and where I was then led.

The local analyses could have led in many different directions. I could have used the conventions of the general evocation of a situation to present a kind of phenomenological account of the school. I chose to go in another direction that would allow for a more explicit account of the features of the general structure. This led me to abstract content and thus to produce something that is removed from the kinds of discourse that are, in fact, produced in the school. The point of this statement is precisely not to restate what people were saying—after all I let them talk sometimes at great length, in their own words through the interviews and documents I excerpted. The point is to gain a form of knowledge about the participants that they do not have and that may be helpful to people who are interested in the life of the participants.

I recognize that many readers are interested in a knowledge that is expressed in terms akin to those used by the participants in the school. It must be possible to relate my words with the traditional discourses that are held about schools outside of school (by philosophers of education, sociologists and anthropologists of education, politicians, parents, etc.). It is with this in mind that I now summarize the process that led me to the statements of rhetorical structure that I made in Chapter IX. First, I restate this structure. Then, in a reverse movement of specification, I relate it to the traditional language of "freedom," "democracy," "individualization," "order," "community,' etc.

We started with a story and its moral: "Teachers who are not backed by the administration have disciplinary problems." Taking into account the setting, the position of the speaker, the other morals that other speakers in the same or different settings and positions could draw from different events, we reached an initial statement of the set of the various things that could be said by all the adults about professional relationships. We saw that all participants could talk in terms of roles, actors in their roles, sets of actors-in-role, persons, and groups of persons relating to each other as persons rather than as actors in a role. These distinctions underscored the speech of all the adults whatever their position, the setting of their talk and the personal opinions. What they might stress at various times did vary, but not the ensemble of those things that could be stressed. It is in this sense that my outline of the set is already a statement of structure and not of content even though the words used to describe the various positions are still rather strongly specified (being drawn as they are from a kind of socio-psychological jargon).

The next step in the analysis started with another story, a story about teachers being angry about survey research conducted by the administration. We then moved to an examination of the kind of writing that is exchanged between administrators and teachers to see if we could find in it a rhetorical order. We moved from sentences like "There is something intriguing about a teacher surplus which now exists in our country today," (T24) to sentences like "Let me take this opportunity to thank each of you..." (T25) to "All school principals are directly responsible..."(T28), to "We shall do our utmost... to help each student..." (T29), etc. Once again, we took into account setting, position of the speaker or writer, possible purpose, medium, etc., to discover the ensemble of the possible means of expressing the struggles that were, in fact, being fought. Given the wealth of the pronominal system, I built the final statement in terms of pronominal use even though I now suspect that I could have used other grammatical categories with essentially similar results (I might, for example, have looked at the use of verbs within the framework of a modified speech act theory). I have explained elsewhere the theoretical reasons that led me to an analysis of pronominal use that is at odds with many recent approaches to the subject (Varenne, forthcoming). In particular, I showed that pronominal structure is not determined by the situation in which pronouns are used. Rather, it is the power of pronouns that they can strongly suggest that a particular situation is to be interpreted by the participants in a certain manner, and not in another. This never absolutely constrains the audience who is always free to reject the proposed identification. Any person in the audience must nevertheless construct his response in terms of the suggestion, however outrageous it may appear.

At this stage, we reached the point of furthest formalization. I made clear that the rhetorical nodes were not items of content but rather specific ways that could serve to frame any content as a special kind of content with a special signification and with a special rhetorical power over the audience. In other words, depending on the mode chosen, the same content could be made to refer to radically different aspects of the people's overall experience. Furthermore, the ensemble of the modes limited what the people could in fact express of this experience since it dealt with all possiblities and thus prevented a fully original intuition from expressing itself in any easy fashion. The ensemble of the rhetorical positions as they could be formally summarized using the pronominal system for its symbolization, was as follows:


This (remember Magritte and his pipe!) "is" not the structure of the rhetorical system, not in any single manner. It is an abstract, symbolic representation of the four positions the participants recognized through their discourse. The usefulness of this representation does not lie in its verisimilitude, nor in its evocative power. Even at the time when I first summarized my analysis, I had to provide a secondary labelling system which, to an American audience, is probably more evocative:

/I/   is also  "the personal mode"  
/we/   is also "the group mode"  
/it/   is also "the generic mode"  
/they/   is also "the collective mode"  

I was then able to show that: to use the

to use the "personal mode" (is to talk about) /Persons/
to use the "group mode" (is to talk about) /persons-in-relation-to-persons/
to use the "generic mode" (is to talk about) /roles/
to use the "collective mode" (is to talk about) /actor-in-role sets/

We now have three ways of writing about the making of the same distinctions in various discourses. I suggested still other ways of writing these distinctions: - to talk of /persons/ (is to talk of) a human biological entity in his/ her unique self, i.e. an "individual"

to talk of /persons-in-relations-to-persons/ (is to talk of) "friendship" groups
to talk of /roles/ (is to talk of) administrative or life style categories
to talk of /actor-in-role sets/ (is to talk of) occupational groups, cliques,

This approach allows us to build a theory of cultural signification that is not (rhetorically) idealistic as all such theories have to appear when they talk of core cultural symbols as realizing themselves in actual texts. An American symbol like 'role" or "person" is not an a priori thing that expresses itself in a third person singular or a proper name statement. It is rather that the notion of role is an elaboration on a more fundamental symbolic structure. To talk about "roles" allows statements about human action to be phrased in abstraction from any live human beings at the very same time as they are made to apply to a whole class of human beings treated separately one from the other.

This aspect of the participants' symbolic competence, that is, their capacity to create such statements, is what is at issue here. It is not necessary that they "believe" in what they are saying, nor that they be fully conscious. When they become conscious, scale participants may, in fact, be truly bothered by what they may be made to utter, or by the practical consequences of rhetorically instigated actions. But the capacity remains.

What is interesting in all this, and what reveals /I/ is also /We/ is also /it/ is also /they/ is also - to use the - to use the - to use the the usefulness of formalization in such an analysis is what is happening in the process of de-formalization. Everyday content is being brought back. It can be seen that, first, contents from different areas are being brought back together (e.g. occupational groups and cliques) which begin to suggest patterns of correspondences across domains, and that, second, diversity is again made possible. The formal statement can help clarify the transformational relationships between the various domains of everyday life. They can also help us see the relationship between everyday life and the general statements of philosophy, political or educational thought that we, as participants of our special environment within the general society, are particularly interested in, statements about individualism, freedom, the proper delivery of education, etc.

From /I/ to 'Individualism'

It is well known that there is a great gap between, on the one hand, the language of the philosophers of education, of the practitioners of the various disciplinary approaches to the study of education, and even of the people who are the closest to the school when they move away from everyday concerns, and, on the other hand, the language that these people use when they are dealing with these everyday concerns.* It is also well known that the sacred discourse presents itself as relevant to what happens in the routine of school life, or at least to what should happen there. It is a discourse about "ideals" but * It should be noted that most philosophers of education have taught, and that most teachers have taken courses in the philosophy of education. The sacred language is thus one that is shared by both groups—even though one group may end up using certain aspects of this language more than the other group. the people who utter them, generally, do not think that this makes them unreal, on the contrary perhaps. The very fact that participants can use such discourse, even if it is only on certain occasions, suggest that the discourses have some relevance.

It is extremely difficult to "see" something like "individualism" in sentences like "pass me the book." This difficulty should not prevent us from looking, if only because the participants themselves are involved in the task of relating the sacred discourse to everyday talk. It is so important to them that they are always interested in what various "experts" have to say about this relationship. We only have to consider the disputes about 'individualized instruction' to realize how immediate the concern of "individualism" is to many and what is involved in the making of the arguments for or against the pedagogy. Questions like the following have direct impact at the most concrete levels: "Is it more individualistic to expect all students to move at the same pace through a curriculum than it is to arrange a system of checks to ensure that something has been learned no matter at what speed, in what order and with whom?" As I have shown, the way such questions are embedded in the school society, with its many other kinds of divisions, makes them unanswerable in a way that would stop all debate. The fact remains that the questions can be asked, and that debates around them are possible. This suggests that the participants do perceive correspondences between everyday life and the sacred language.

We have looked at a few sacred texts produced by the participants themselves (e.g., the "School Philosophy" (T29), the introduction to the "Program of Studies" (T63), the Desideratum (T96)). We saw that, beyond the obvious vocabulary differences that marked the texts as "sacred," there were direct correspondences between their rhetorical structure and the structure that we abstracted through our look at routine speech. I have no doubt that, should we spend much more time on more thoughtful examples of such texts, the same kind of structure would be found to operate. Structural homologies are demonstrated through the fact that it is possible, as one rewrites the various statements one has collected in more and more formal terms, to arrive at the same point from many different starting points. It is not that the abstract statement "is" the structure. I prefer a solution to the problem inspired by Bourdieu's work (1977): the formal statement is an account of the constant historical reproduction of situations in practical life. The formal statement of structure is a heuristic tool for us to perceive more starkly what used to be called the "integration" of a culture, and to account for the fact (as intuited by the observer and the natives) that, in some ways, every domain of a culture is stamped by something of very general value within this culture.

The process is less the search for a hidden, transcendental "deep" structure that would "realize" itself in speech, than it is an attempt to discipline analysis in such a manner that conclusions can be sharply drawn. What the process yields has several advantages:

1) It makes the analysis more susceptible to detailed empirical criticism and falsification;
2) It solves puzzles otherwise difficult to deal with (e.g. the relationship of everyday language to sacred languages);
3) It suggests the cultural source of many of the conflicts that make life difficult;
4) It makes American culture appear more complex than it is generally made to appear in general statements of overall pattern.

Let us see how some of the above works around the cause celebre of "individualized instruction." A big difficulty for the proponents was that their opponents could argue their own case in the very same terms as they used: The opponents could fight the new curriculum in the name of individualism. Furthermore, the very routine of the school was itself so extremely individualized that any innovation that pretended to make it even more so could easily be shown not to make much difference. There were also other, even more fundamental, difficulties: Individualism, the /I/ mode, is but one of the ways that one can use to deal legitimately with students. Any student "is" not an individual in any simple manner. He is also (in the sense that he can also be talked about in terms of) a member of a class with certain characteristics common to all members of this class (e.g. age, academic competence, sex, etc.) with certain rights, privileges, duties, responsibilities, capacities and incapacities that transcend the student's own personality. The adults must take all this into account when they deal with him. A student "is" also (in the sense that he can be talked about in terms of) a member of a group of friends formed by his own activity but which now has an independent reality. For the participants, all these realities are immediately "here," phenomenologically. Dealing with all of them at the same time is what is not easily done in a way that will allow any observing participant to see the attempt as appropriate. Actions that one might take to foster one aspect of the whole are generally dangerous and possibly unacceptable in terms of another aspect of this whole. Total individualization of instruction, critics can say, may lead to very young children graduating from high school and then going on to college before they are "mature" enough. In this context, "maturity" is made to refer to an age specific property that is not directly related to the person's individuality. Other people (or the same people in other settings or with other interlocutors) may also argue that adolescents cannot ever truly be treated as individuals since they are still fully dominated by their "peer group," the group of their friends that dissolves their individuality.

All this that can be said about the students can also be said about the teachers and the different ways one can deal with them and their activity. They, too, are individuals entitled to their own opinions, they, too, are members of role sets with different interests from those that have different general characteristics from them. And they, too, are members of various friendship groups. We saw at great length the kind of problems this created for the people who were in charge of administering the teachers, and for the teachers themselves in their ambiguous relationships with their students.

The fact that this variety of possible modes is available does not mean that any one speaker is obliged to stress them all equally at all points in his talk. We saw repeatedly that participants, in the actual discourse they produce in response to the other acts and discourses that are produced around them, generally only stress one possible way of handling interaction. Conversely, on those occasions when the major emphasis is put on one aspect of the structure, the de-emphasized aspects remain present, albeit often in pejorative form. "Individulization" can also be said to foster "disorder," "egoism," or competition that is crippling or elitist. "Communalization" can also be said to produce mindless "conformism," to stultify individual "growth" and/or "independent enterprise." To every positive term in the symbolic system there is a negative one of the same structural value that can be used to express one's point of view without making one's statement rhetorically irregular.

At this point, all this can only be indicative. It does strongly suggest that a structural analysis which, at one stage, can produce the lines of an apparently simple structure, can also help us recapture a more complex picture of the life of the people. These people are not "individualists" or "conformists." They are, rather, people who have to deal with a complex set of positions of which individualism is but one pole that is there only to the extent that there are other poles around it that are not "individualism." Teachers, philosophers of education and curriculum specialists talk about individuals, but they must also talk about the need to "socialize" children by "making them relate with other children" (in friendship groups --not in cliques...) and to prepare them for jobs. All these topics are of major importance to the participants and, as a whole, American education cannot avoid dealing with any of them.

From American Rhetoric to American Culture

So much has been written about American culture that nothing very new can be said about it. We all know that there is some truth to the saying that America is democratic, individualistic, inner-directed, conformist, other-directed, one-dimensional, etc. (Spindler and Spindler, 1983). What is necessary now is to look at all these apparently contradictory statements from a different perspective to see how they could be true at the same time. I do not believe Riesman (1961) seriously considered the possibility that Americans could be both inner- and outer-directed. Kluckhohn and Strodtbeck, in their pioneering work on value-orientation (1961), do not seem to have considered the possibility that their scales told more about the culture of those who designed them than they did about those who were answering their questions. When Slater (1970) complains about the dangers of extreme individualism and the need to live in consensual communities, he does not consider that the utopian communities he clamors for are communities of individuals who have freely and independently chosen to live in them. No serious sociologist seems to have considered the possiblity that it might be worthwhile to return to the kind of feudal and hierarchical familism that shocked Banfield when he encountered it in Southern Italy (1958). No social scientist or philosopher is considering that there may be an ultimate wisdom in the Hindu theories of transcendent social orders to which it should be the goal of individuals to try and fit at their appointed place (Dumont, 1977, J. Varenne, 1976).

What is American, then, is partially bounded by what is absent from America. It consists in what is there, as a whole, and at their appointed places. There is no doubt, as we saw in many different ways, that individualism is extremely "present" in American culture--even though many moralists can justifiably say that Americans are not individualistic enough and that too many institutions, like schools, do too little to preserve, or foster, the needed individualism. The evaluation of the moralist and the evaluation of the observer proceeds in different ways. For the observer what is interesting is that the moralist should be so interested in the preservation of individualism, that there can be differences of opinion about the extent of the presence of individualism, that there can be so many institutional mechanisms designed to foster individualism, and that so many people can talk about it--sometimes even to mourn this very search for individualism! It is of little concern that much of that talk is, from the point of view of the moralist, empty or misguided. It is of no importance that the mechanisms do not lead to what it was hoped they would. They are there, and they are not powerless. Individualism may be a myth, an ideal, a value, it is also an enormous amount of physical energy realized in mountains of paper, reels of film, hours of speeches, admonitions, eulogies and apologies. It is a specific way of talking about social interaction, about what one does and what other aredoing.

The same can be said about the ideals or values of community, appropriate and efficacious job performance, etc. However unhappy about it many Americans may be, an enormous amount of energy is also spent ensuring that jobs will be specifically described so that efficient performance is made completely independent from the personality of the persons who may, in fact, occupy the jobs. One has but to notice that all major discrimination suits are argued in terms of the appearance that certain job specifications are written in ways that are not independent from personal qualities that are irrelevant to the performance of the job. The specification of jobs independently from the personality of those who occupy the "role" (the /it/ mode) is an aspect of the American environment, and one that the people who participate in it, whether they have been born within it or not, whether they "believe" in its ultimate worth or not, have to deal with even when they try to create something new, even when they try to be "original."

The individual, the community, the role, as symbols, are not mutually exclusive. The talk about their mutual exclusivity is but one of the forms that discourses about them can take. It is not irrelevant that so many can talk as if the various symbols were mutually exclusive. It is revealing of a problem, produced by the symbolic system which, precisely because it is symbolic, 1) must be used in communication, 2) cannot possibly deal with all the complexity of life. Other discourses are also possible that underline the fact that the various poles of the system are, in fact, in balance. It is easy to dismiss such discourses the product of unrealistic, naive optimism. It is interesting, however, that it is eminently possible to say, as a philosopher of education once said, in a class:

The good life consists of persons in community with love as the supreme value based upon the sanctity of persons as unique centers of being, always standing in relation of mutual understanding and acceptance.*

In this statement, community and individualism are made mutually interdependent. This is an old idea, an old hope perhaps. It is the idea that inspired the Founding Fathers as they tried to balance the rights of individuals and the social and legal conditions that allowed for the preservation of these rights, and then evolved a peculiar form of federalism that protected semi-independent States. The idea of the close relationship of individualism to community life is one on which, later, various philosophers and social scientists tried to build theories of human action that often appeared particularly convincing in American environments.**

The presence of the rhetoric and of the symbolic forms that signal which position one is assuming at any particular time cannot be denied. But, one may ask, is this rhetorical system a "culture"? An answer to this question depends, obviously, on one's definition of the word "culture." The one that I am using here is the one that has been the most fruitful in anthropological research, the one that is closely related to the definitions given by the Boasians and the pragmatists in the United States, the Durkheimians in France, etc. All these definitions, however much they may differ in their detail, stress that there is a power in the social environments that bends people into specifiable directions that are different from * As recorded by a student in a class given at Teachers College by Phil Phenix in 1980. ** I am thinking here of writers like Dewey (1922), G.H. Mead (1967) or Royce (1968) the directions taken by even closely related populations. The next step, in the progression of these theories is the more controversial. It produces very different statements that can appear, for a while at least, mutually exclusive. This step consists in arguing, first, that a central aspect of the constitution of the environment is its symbolic aspect and, second, that, symbols being what they are, they "integrate" the behavior of large numbers of people in many radically different situations.

The position just outlined is dependent on the intuition that cultures are, indeed, integrated, that one is recognizably in America whether one is in school, at home, in church or on the job. As Schneider has recently written, one can show that "ultimate values, collective representations... permeate the total society and are not confined to religion, to ritual, to magic or myth alone" (1976: 208). To talk of rhetoric has helped us particularly to fulfill this program by showing how the many different environments that can be found in a school reveal the presence of a united generating structure, and how this structure is, indeed, related to major concerns of Americans as these can be observed in the vocabulary of values and ideals.

To close, I would like to cane back to the question that we asked at the beginning of this conclusion: What do I claim when I claim that I understand the semiotic means which Americans use to converse about the world? Do I claim that I know Americans? America? American culture? The words which some people use? It should be clear by now that I do not claim to know Americans in their plurality. I do not "know" many Americans, if I know any. Single Americans are always more than what is suggested to them by their culture. To this extent, I am very sceptical of any work that pretends to tell who or what Americans are, in the plural. What can be done are descriptions of what people have to deal with. I do not know Americans, but I have a good idea about what shapes their expression of what they have to deal with. What this is, is, by all account, their "culture." It is

- a rhetoric;
- words to be used;
- words that suggest a world of social relationships;
- words that fail to handle experience;
- words that open new vistas.

In other words (sic), I claim to know Americans in the words that they—and I--must use.


* This is, of course, a direct echo of the Whorfian hypothesis. It differs from many early interpretations of the hypothesis in that it emphasizes the social aspect of the postulates. While linguistic form can mold personality structure, it is not necessary to investigate this one or to assume anything about limited perception in order to demonstrate the limitation of experience (through the structuring of the environments) and the limitation of expression (through the structuring of the communicative tools at the speaker's disposal. This allows any audience to misunderstand the speaker "appropriately"--in that any speech suggests alternative interpretations of itself). Recent work on Whorf suggests that this is precisely what he had in mind (Silverstein, 1976).