Como pensam as instituições? (Douglas 1986)

Como pensam as instituições? (Douglas 1986)

DOUGLAS, Mary. 1986. How institutions think. Syracuse: Syracuse University Press.

Obs: Este livro resulta de uma coletânea de textos reunidos e editados por ocasião de um curso ministrado por Mary Douglas (6th Abrams Lectures) na Syracuse University (New York).


A theory of institutions that will amend the current un-sociological view of human cognition is needed, and a cognitive theory to supplement the weaknesses of institutional analysis is needed as well. (Douglas 1986:ix)


This is the first book I should have written after writing on African fieldwork. […] This volume is one more post hoc introduction [to “a coherent argument about the social control of cognition”]. […] I wish I could hope that this volume might be so acceptable as to break the spell, so that I could now start writing forwards instead of backwards. (Douglas 1986:ix-x)


I wrote Purity and Danger (1966) in an attempt to generalize from Africa to our own condition. My friends told me at the time that Purity and Danger was obscure, intuitive, and ill-prepared. They were right, and I have been trying ever since to understand the theoretical and logical anchoring that I would have needed to present a coherent argument about the social control of cognition. (Douglas 1986:ix)


This book begins with the hostility that greeted Emile Durkheim and the Durkheimians when they talked about institutions or social groups as if they were individuals. The very idea of a suprapersonal cognitive system stirs a deep sense of outrage. […] An individual that encompasses thinking humans is assumed to be of a nasty totalitarian sort, a highly centralized and effective dictatorship. (Douglas 1986:x)


The one scholar whose mark is most strongly on the whole area covered here is Robert Merton. To him I respectfully and affectionately dedicate the book, trusting his generosity to overlook its failings. My husband deserves a special tribute. When two problems seem insoluble, our long experience of domestic life has suggested an oblique approach. Instead of a head-on attack on each separate issue, one set of problems can be made to confront the other. This strategy, which produces new definitions of what has to be solved, gives the framework of this book. (Douglas 1986:x-xi)

INSTITUIÇÕES (não têm individualidade cognitiva, mas conferem identidade por meio de analogias, sistemas classificatórios, operações de esquecimento e rememoração e decisões de vida ou morte)

Institutions Cannot Have Minds of Their Own (Chapt.1)

Institutions Are Founded on Analogy (título do capítulo 4)

Institutions Confer Identity (título do capítulo 5)

Institutions Remember and Forget (título do capítulo 6)

Institutions Do the Classifying (título do capítulo 8)

Institutions Make Life and Death Decisions (título do capítulo 9)

COOPERAÇÃO/REJEIÇÃO e SOLIDARIEDADE/DESCONFIANÇA (o elo social é um tema passional e fundamental, mas evitado)

Writing about cooperation and solidarity means writing at the same time about rejection and mistrust. Solidarity involves individuals being ready to suffer on behalf of the larger group and their expecting other individual members to do as much for them. It is difficult to talk about these questions coolly. They touch on intimate feelings of loyalty and sacredness. Anyone who has accepted trust and demanded sacrifice or willingly given either knows the power of the social bond. Whether there is a commitment to authority or a hatred of tyranny or something between the extremes, the social bond itself is taken to be something above question. Attempts to bring it out into the light of day and to investigate it are resisted. Yet it needs to be examined. Everyone is affected directly by the quality of trust around him or her. Sometimes a gullible steadfastness allows leaders to ignore the public need. Sometimes trust is short term and fragile, dissolving easily into panic. Sometimes mistrust is so deep that cooperation is impossible. (Douglas 1986:1)

For them [Durkheim and Fleck], true solidarity is only possible to the extent that individuals share the categories of their thought. (Douglas 1986:8)

Not just any busload or haphazard crowd of people deserves the name of society: there has to be some thinking and feeling alike among members. […] Just because it is legally constituted, a group cannot be said to “behave” – still less to think or feel. […] If this is literally true, it is implicitly denied by much of social thought. (Douglas 1986:9)


[I]ndividuals in crises do not make life and death decisions on their own. Who shall be saved and who shall die is settled by institutions. Putting it even more strongly, individual ratiocination cannot solve such problems. An answer is only seen to be the right one if it sustains the institutional thinking that is already in the minds of individuals as they try to decide. (Douglas 1986:4)


This book is written precisely to encourage more probing into the relation between minds and institutions. (Douglas 1986:7)

OS PRESSUPOSTOS INDIVIDUALISTAS DA TEORIA DA AÇÃO RACIONAL (para o caso de 5 homens isolados e sem alimentos)

Only the individualists, bound by no ties to one another and imbued by no principles of solidarity, would hit upon the cannibal gamble as the proper course. (Douglas 1986:8)

O DIÁLOGO DE SURDOS (premisses=assumptions=institutions; improve our understanding = reformulate = transform)

Arguing from different premises, we can never improve our understanding unless we examine and reformulate our assumptions. (Douglas 1986:8)


Emile Durkheim had another way of thinking about the conflict between individual and society [in comparison with the rationalists/individualists]. He transferred it to warring elements within the person. For him the initial error is to deny the social origins of individual thought. Classifications, logical operations, and guiding metaphors are given to the individual by society. Above all, the sense of a priori rightness of some ideas and the nonsensicality of others are handed out as part of the social environment. He thought the reaction of outrage when entrenched judgments are challenged is a gut response directly due to commitment to a social group. In his view, the only program of research that would explain how a collective good is created would be work in epistemology. […] [But] Durkheim’s sociological epistemology ran into considerable opposition and has remained undeveloped to this day. By upgrading the role of society in organizing thought, he downgraded the role of the individual. […] He seemed to be invoking some mystic entity, the social group, and endowing it with superorganic, self-sustaining powers. For this he earned attack as a conservative social theorist. In spite of these weaknesses, his idea was still too good to be dismissed. (Douglas 1986:10)

To read The Elementary Forms of the Religious Life in isolation from the rest of Durkheim’s work is to insure misunderstanding it, for his thinking was a single arch in which each major publication was a necessary statement. He harped always on the one theme, the loss of classificatory solidarity. He deplored its irreplaceability and the crises of individual identity that follow from absence of strong, supporting, publicly shared, and privately internalized classifications. He taught that publicly standardized ideas (collective representations) constitute social order. He recognized that the hold they have upon the individual varies in strength. Calling it moral density, he tried to measure its strength and to assess the effects of its weakness. According to Durkheim, sociological method requires that individual responses be treated as psychological facts to be studied in a frame of reference of individual psychology. Only collective representations are social facts, and social facts count for more than psychological ones because the individual psyche is constituted by the socially constructed classifications. Since the mind is already colonized, we should at least try to examine the colonizing process. […] When Durkheim wrote with Marcel Mauss the essay on primitive classification (1903), what had already been a long-term conviction (that true solidarity is based on shared classifications), started to become a method. […] [W]hile everyone else was adopting institutionally prescribed postures about modernity, the loss of legitimacy, wonder and sacredness, Durkheim and Mauss proposed to analyze the extent to which the mundane classifications we use are projections of the social structure partaking in the aura of sacredness. The sacred that Weberians regretted was an unanalyzable mystique. The sacred for Durkheim and Mauss was nothing more mysterious or occult than shared classifications, deeply cherished and violently defended. That is not all: this idea of the sacred is capable of analysis. (Douglas 1986:96-7)

In writing about the sacred, Durkheim was trying to put his finger on how institutions do the classifying. His idea was not that sacred power flashes out as an inherent property of constitutions and kings, but the other way around. The peoples he chose to represent the elementary social forms have no constitutions, kings, or any superordinate coercive authority. To the Australians, the sacred can only draw its power from their own consensus. Its coercive strength, which arms the whole universe with punishing taboos to reinforce the individual’s wavering commitment, is based on the classifications inside the same individual’s head. It is based essentially upon the classifications pertaining to the division of labor. Thus, his theory of the sacred is not just one about disappearing civilizations but also one about moderns, since we also have a society based on the division of labor. The book on suicide (1897) and his development of the idea of anomie are Durkheim’s best demonstration that he expected us to learn about ourselves from ethnographic societies. (Douglas 1986:97)


In its early formulations, the sociology of knowledge in Germany was dogged by relativist problems and dominated by propagandist intentions. As these elements were gradually eliminated, the focus of the subject turned much more upon the relations of the individual to the social order in general. The effect of variation in the social order was (and is still) largely overlooked. All the focus was upon the interests. The usual typology of knowledge, for example, tends to explain different points of view by reference to the conflicting interests of different sections within modern industrial society. There was no attempt to compare viewpoints based on totally different types of society. […] It is clear that no disciplined comparative framework would emerge from a sociology uninterested in the range of variety among different societies. (Douglas 1986:11)


[S]ociology, though it may have started with philosophical questions and political issues, received its major impulse for development because it provided an indispensable tool for administrative purposes. So Durkheim’s intellectual program has languished. (Douglas 1986:11)

FLECK e DURKHEIM – e Goodman e Becker (complementares na luta pela cognição social)

Fleck elaborated and extended Durkheim’s approach. […] In many places Fleck went far beyond Durkheim; in others he missed Durkheim’s central synthesizing idea. Both were equally emphatic about the social basis of cognition. […] Fleck went further than Durkheim in analyzing the idea of a social group. He introduced several specialized terms: the thought collective (equivalent to Durkheim’s social group) and its thought style (equivalent to Durkheim’s collective representations), which leads perception and trains it and produces a stock of knowledge. […] For Fleck the thought style is as sovereign for the thinker as Durkheim held collective representation to be in primitive culture, but Fleck was not talking about primitives. (Douglas 1986:11-3)

Fleck was not interested in sacredness or in social evolution. Nonetheless he applied the Durkeimian idea of a sovereign thought style to modern society, even to science. This would have horrified Durkheim. As Fleck said, the Durkheimians exhibited “an excessive respect, bordering on pious reverence, for scientific facts” […]. He ridiculed their attitude as a naive obstacle to the building of a scientific epistemology. […] In dealing with the criticisms that affect them both, a good strategy is to get Durkheim and Fleck to make a common defense. Sometimes Fleck has the best answer, sometimes Durkheim. Fighting as allies, back to back, each can supplement with his strength the weakness of the other.(Douglas 1986:14)

We may be tempted to suppose with Durkheim that scientific ideas force their evidence upon our experiments. We know that this runs counter to the history of science and to the tracing of distinctive thought styles. Fleck was more up-to-date in insisting that a scientific fact does not smack the researchers between the eyes and compel assent. He showed that it took four centuries before scientific advances in other fields were important enough to establish a definitive distinction between different diseases originally clumped together as venereal [.] […] A combined Durkheim-Fleck approach to epistemology prevents either science or religion from being accorded too much privilege. Both science and religion are equally joint products of a thought world; both are improbable achievements unless we can explain how individual thinkers combine to create a collective good. (Douglas 1986:37)


For Durkheim the division of labor accounts for the big difference between modern and primitive society: to understand solidarity we should examine those elementary forms of society that do not depend on exchange of differentiated services and products. According to Durkheim, in these elementary cases individuals come to think alike by internalizing their idea of the social order and sacralizing it. The character of the sacred is to be dangerous and endangered, calling every good citizen to defend its bastions. The shared symbolic universe and the classifications of nature embody the principles of authority and coordination. In such a system problems of legitimacy are solved because individuals carry the social order around inside their heads and project it out onto nature. However, an advanced division of labor destroys this harmony between morality, society, and the physical world and replaces it with solidarity dependent on the workings of the market. Durkheim did not think that solidarity based on sacred symbolism is possible for industrial society. In modern times sacredness has been transferred to the individual. These two forms of solidarity are the basis of the main typology in Durkheim’s theory. (Douglas 1986:13)


Certainly there is a new interest in distinct styles of reasoning in the history of science. (Douglas 1986:15)


The faulty argument can be expressed as follows. Smallness of scale fosters mutual trust; mutual trust is the basis of community; most organizations, if they do not have a base in individual selective benefits, start as small, trustful communities. Then the special characteristics of community solve the problem of how the social order can ever emerge. Many maintain that after the initial birth through the community experience, the rest of social organization can be explained by complex interlocking of individual sanctions and rewards. […] Has no one writing on this subject ever lived in a village? Ever read any novels? Tried to raise funds? […] One may wonder if this is a form of inquiry or an ideology or a quasi-religious doctrine. […] For the appeal to the small, idealized, intimate community is strong in political rhetoric. […] Michael Taylor […] is also among many who believe that small communities are a form of society where rational self-interest does not dictate the outcome of decisions […]. Given only that it be small enough and stable enough, members of the community are thought freely to make contributions that they would withhold in larger and more fluid conglomerations. This formula is somewhat question-begging, because the issue is how that community gets to be stable. (Douglas 1986:24-5)


The individual cost-benefit analysis applies inexorably and enlighteningly to the smallest micro-exchanges, with them as well as us. […] It is when making threats and offers that individuals often invoke the power of fetishes, ghosts, and witches to make good their claims. The resulting cosmology is not a separate set of social controls. In Durkheim’s work the whole system of knowledge is seen to be a collective good that the community is jointly constructing. (Douglas 1986:29)


Any attempt to probe the foundations of social order brings to light the paradoxical foundations of thought […] [,] questioning how systems of knowledge come into being. There is plenty of good reason to think that rational choice theory is inadequate to explain political behavior. Something is going on in civic affairs that the theory of rational choice does not capture. According to the Durkheim-Fleck position, the mistake is to have ignored the epistemological problem. Instead of supposing that a system of knowledge springs into being naturally and easily, their approach extends skepticism about the possibility of collective action to skepticism about the possibility of shared knowledge and shared beliefs. (Douglas 1986:29-30)

REJECTION OF EMOTIONAL CAUSAL LOOPS (avoid explaining the genesis of rationality by it’s suspension)

The case for ritual stimulating the emotions is weak. Hasn’t anyone ever been bored in church? It is important to notice that this clearly goes against Durkheim’s principles of sociological method […]. Social facts must be explained by social facts. Dipping at will into the psychological level was precisely what Durkheim’s method aimed to stop. Durkheim evaded his own rules of method by making the sacred depend for its vitality on the emotional excitement of great gatherings. Fleck used the more coherent principle that trust and confidence are prerequisites of communication; he thereby avoided the inconsistency of suspending rationality in order to explain the origin of rational thought in effervescent emotions stirred up by grand-scale public rituals. It is safer to follow Durkheim’s teaching, rather than his practice, and safer to reject the functional explanation based on emotions that keep the system going. (Douglas 1986:34-5)


Religion does not explain. Religion has to be explained. We cannot allow Durkheim and Fleck and their friends to brush the main problem aside without more justification. Like everyone else, they must spell out the logical steps of their case or accept the charge of mysticism and appeal to the irrational. (Douglas 1986:36-7)


Merton originally cited the Hopi rain rite as a case of a ritual that performs the latent social function of rousing emotions that’ support solidarity. The dance does not produce rain for the parched desert, but it serves a latent social function. Following the same argument with the same illustration, Elster attributes the Hopi Rain Dance to the Trobrianders, living in fertile, well-watered islands. We suspect that if he had attributed the Trobrianders’ ocean-fishing magic to the land-locked Hopi, it would not have mattered. The anthropology does not matter. It is not even interesting enough to be read. In this debate, it serves only as a stalking horse for more serious quarry, whatever that may be. (Douglas 1986:42-3)


Philosophers of science go to great trouble to learn the terminology and theories of relativity and quantum physics. Yet they pay scant attention to the social group that is the carrier of a thought style. […] By classing discoveries in physics or biology as the main object of their research, philosophers of science have already adopted an implicit theory of knowledge. It is even one that has been tried and rejected elsewhere, the idea of a passive perceiver. (Douglas 1986:43)


How a system of knowledge gets off the ground is the same as the problem of how any collective good is created. In Durkheim’s view the collective foundation of knowledge is the question that has to be dealt with first. According to his theory, the elementary social bond is only formed when individuals entrench in their minds a model of the social order. He and Ludwik Fleck invited trouble when they wrote of society behaving as if it were a mind writ large. It is more in the spirit of Durkheim to reverse the direction and to think of the individual mind furnished as society writ small. The entrenching of an idea is a social process. This is compatible with the prevailing notion in the philosophy of science that a theory is entrenched by its coherence with other theories. But the burden of the argument is that the whole process of entrenching a theory is as much social as it is cognitive. Conversely, the entrenching of an institution is essentially an intellectual process as much as an economic and political one. (Douglas 1986:45)

ANALOGIA: a fórmula da legitimidade social

A focus on the most elementary forms of society brings to light the source of legitimacy that will never appear in the balancing of individual interests. To acquire legitimacy, every kind of institution needs a formula that founds its rightness in reason and in nature. Half of our task is to demonstrate this cognitive process at the foundation of the social order. The other half of our task is to demonstrate that the individual’s most elementary cognitive process depends on social institutions. (Douglas 1986:45)

The favorite analogy generalizes everyone’s preferred convention. (Douglas 1986:50)

How does one constructed analogy win over another? How does a system of knowledge get into orbit? How does one good idea compete with another? This is a central issue in the history of science. (Douglas 1986:57)

Individuals, as they pick and choose among the analogies from nature those they will give credence to, are also picking and choosing at the same time their allies and opponents and the pattern of their future relations. Constituting their version of nature, they are monitoring the constitution of their society. In short, they are constructing a machine for thinking and decision-making on their own behalf. (Douglas 1986:63)

CONVENÇÃO e LEGITIMIDADE: a instável instituição mínima (Lewis)

Minimally, an institution is only a convention. David Lewis’ definition is helpful: a convention, arises when all parties have a common interest in there being a rule to insure coordination, none has a conflicting interest, and none will deviate lest the desired coordination is lost […]. Thus, by definition, a convention is to that extent self-policing. (Douglas 1986:46)

We want conventions about pedestrian crossings to exist, but we will violate them ourselves if we can do so with impunity. Enough impatient pedestrians to create a critical mass will march across and hold up the cars in defiance of traffic lights. The conditions for stable conventions to arise are much more stringent than it might seem. Communities do not grow up into little institutions and these do not grow into big ones by any continuous process. For a convention to turn into a legitimate social institution it needs a parallel cognitive convention to sustain it. (Douglas 1986:46)


In the rest of this volume, institution will be used in the sense of legitimized social grouping. The institution in question may be a family, a game, or a ceremony. The legitimating authority may be personal, such as a father, doctor, judge, referee, or maitre d’hotel. Or it may be diffused, for example, based by common assent on some general founding principle. What is excluded from the idea of institution in these pages is any purely instrumental or provisional practical arrangement that is recognized as such. Here, it is assumed that most established institutions, if challenged, are able to rest their claims to legitimacy on their fit with the nature of the universe. A convention is institutionalized when, in reply to the question, “Why do you do it like this?” although the first answer may be framed in terms of mutual convenience, in response to further questioning the final answer refers to the way the planets are fixed in the sky or the way that plants or humans or animals naturally behave. (Douglas 1986:46-7)


It is at this time fashionable to say that social institutions encode information. They are credited with making routine decisions, solving routine problems, and doing a lot of regular thinking on behalf of individuals. This recent work is very pertinent. However, we find that there are many ways of talking about institutions as organizers of information. […] Human rationality is inherently bounded. Institutional organization is now widely treated as a way of solving problems arising from bounded rationality. Using Oliver Williamson’s analysis as a point of departure, Andrew Schotter (1981) has rewritten the description of institutions in information theoretic terms. In this sense, information is not a more or less available commodity; it is whatever is newsworthy. The more that an item of behavior is predictable, the less information it carries. The focus of study has shifted from the flow of information (which is rather like a flow of commodities, in Williamson’s sense) to studying the amount of information carried by a particular item seen against the background of standard expectations. This analysis, based on E. E. Shannon’s model of information, treats institutional structures as forms of informational complexity. Past experience is encapsulated in an institution’s rules so that it acts as a guide to what to expect from the future. The more fully the institutions encode expectations, the more they put uncertainty under control, with the further effect that behavior tends to conform to the institutional matrix: if this degree of coordination is achieved, disorder and confusion disappear. Schotter presents institutions as entropy-minimizing devices. They start with rules of thumb and norms; eventually they can end by storing all the useful information. When everything is institutionalized, no history or other storage devices are necessary: “The institution tells all” […]. […] This is fine and highly congenial to a Durkheimian analysis. The one snag is that it does not say how institutions ever start and get enough stability to do all of that. (Douglas 1986:47-8)

A NATURALIZAÇÃO DAS CLASSIFICAÇÕES SOCIAIS (a coerência como princípio analógico estabilizador da ordem social

Equilibrium cannot be assumed; it must be demonstrated and with a different demonstration for each type of society. Schotter reminds us that disorder is more probable than order. Before it can perform its entropy-reducing work, the incipient institution needs some stabilizing principle to stop its premature demise. That stabilizing principle is the naturalization of social classifications. There needs to be an analogy by which the formal structure of a crucial set of social relations is found in the physical world, or in the supernatural world, or in eternity, anywhere, so long as it is not seen as a socially contrived arrangement. When the analogy is applied back and forth from one set social relations to another and from these back to nature, its recurring formal structure becomes easily recognized and endowed with self-validating truth. […] Ultimately, the whole system is grounded on nature, on the preeminence of the right hand over the left, of the east over the west, of the north over the south, and so on. The institutions lock into the structure of an analogy from the body. (Douglas 1986:48-9)

[T]he social convention […] needs a naturalizing principle to confer the spark of legitimacy on what they want to do. The analogy from nature goes as follows: as natural progenitor (say wolf for lion) is to natural offspring (cubs, whelps), so live father is to live son and dead father to dead son. Extending backwards, it can justify the same relation invoked between dead father’s father’s father with dead father’s father and dead father, according to the scale of the living persons ready to be involved in the legitimated social arrangements. […] Thus the institutions survive the stage of being fragile conventions: they are founded in nature and therefore, in reason. Being naturalized, they are part of the order of the universe and so are ready to stand as the grounds of argument. Two examples […] of these naturalized principles of social organization [are:] […] the foundation of a primitive state on the analogy between the relation of female and male with the relation of left and right[;] […] [and] the foundation of a lineage on the analogy of the relation of genitor to offspring. Many more such analogies that confer natural status on social relations abound in anthropological literature. (Douglas 1986:52)

By using formal analogies that entrench an abstract structure of social conventions in an abstract structure imposed upon nature, institutions grow past the initial difficulties of collective action. […] We should now consider how analogies from nature are found and, above all, how they are agreed upon. This points back to the logically prior question of how individuals ever agree that any two things are similar or dissimilar. Where does sameness reside? The answer has to be that sameness is conferred on the mixed bundle of items that count as members of a category; their sameness is conferred and fixed by institutions. (Douglas 1986:53)

In the work of trying to understand, disorder and incoherence are more probable. Whenever a high degree of logic and complexity is found, it is a matter for surprise and needs to be explained. […] A truly complex ordering is the result of sustained effort. Some inducement must exist to explain why the effort is made. […] Let us assume that, in the absence of heavy demand (meaning, in the absence of inducements for specialized concentration), classification will meet minimum needs by taking the path of least effort. That path will quickly lead to a loose collection of social analogies drawn upon nature, and there it will peacefully come to rest. (Douglas 1986:56)

Once a social system has been founded in reason and nature, we can see how cognitive energy is saved by tracing the career of a successful theory. First, on the principle of cognitive coherence, a theory that is going to gain a permanent place in the public repertoire of what is known will need to interlock with the procedures that guarantee other kinds of theories. At the foundation of any large cognitive enterprise are some basic formulae, equations in common use, and rules of thumb. In science such shared techniques of validating spread across different subdisciplines. For example, the mathematics of seepage is used in mineralogy and in ophthalmology. So also the Nuer use the same formula for marriage and blood debts. The anchoring of a set of theories in one field imparts authority to a set elsewhere, if it can be anchored by the same procedures. This is just as true for social forms of validation as for scientific ones. […] Most rediscovered theories turn out not to have built originally on the current cognitive infrastructure and so to have missed savings in energy. Often when a new scientific discovery has been rejected and left to lie inert until later, it is precisely an idea which lacked formulaic interlocking with normal procedures of validation. The best chance of success is to confront the major public concerns and to exploit the major analogies on which the socio-cognitive system rests. (Douglas 1986:76-7)

A new discovery has to be compatible with political and philosophical assumptions if it is to get off the ground in the first place, to say nothing of being remembered afterwards. (Douglas 1986:80)

One well-instituted tool can easily ruin the career of a theory that cannot use it. One well-connected unifying method can drive out an idea that does not depend upon its accredited formula. (Douglas 1986:89)

Only one term sums up all the qualities that enable a speculation to become established and then to escape oblivion; that is the principle of coherence. To employ the same interlocking methodology that holds other clumps of scientific activity together is essential. With this secure, much else will be added; individual researchers will know how to ratify their private claims and how to attract collaborators to collective action; they will know what can safely be overlooked and what must be remembered. […] The principle of coherence is not satisfied by purely cognitive and technological fit. It must also be founded on accepted analogies with nature. This means that it needs to be compatible with the prevailing political values, which are themselves naturalized. […] Inevitably, if it seems that an analogy does match nature, it is because the analogy is already in use for grounding dominant political assumptions. It is not nature that makes the match, but society. (Douglas 1986:90)

If Rivers had a great success for his colonial model of psychic control and if Bartlett neglected the project of identifying social pressures on the cognition of modern man, both the success of the one and the diversion of the other’s intent can be explained by the power of a dominant naturalizing metaphor. The metaphor of evolutionary progress in nature was so congenial that any research based on it could claim the benefits of general coherence. (Douglas 1986:90)

A thinker who classifies the phenomena to be examined according to known and visible institutions saves himself the trouble of justifying the classification. It is already the normal conceptual scheme for those who live in and think through similar institutions. (Douglas 1986:94)

Any institution that is going to keep its shape needs to gain legitimacy by distinctive grounding in nature and in reason: then it affords to its members a set of analogies with which to explore the world and with which to justify the naturalness and reasonableness of the instituted rules, and it can keep its identifiable continuing form. […] Any institution then starts to control the memory of its members; it causes them to forget experiences incompatible with its righteous image, and it brings to their minds events which sustain the view of nature that is complementary to itself. It provides the categories of their thought, sets the terms for self-knowledge, and fixes identities. All of this is not enough. It must secure the social edifice by sacralizing the principles of justice. […] This is Durkheim’s doctrine of the sacred. All the other controls exerted by institutions are invisible, but not the sacred. According to Durkheim, the sacred is to be recognized by these three characteristics. First, it is dangerous. If the sacred is profaned, terrible things will happen; the world will break up and the profaner will be crushed. Second, any attack on the sacred rouses emotions to its defense. Third, it is invoked explicitly. There are sacred words and names, sacred places, books, flags, and totems. Such symbols make the sacred tangible, but they in no way limit its range. Entrenched in nature, the sacred flashes out from salient points to defend all the classifications and theories that uphold the institutions. Fot Durkheim, the sacred is essentially an artifact of society. It is a necessary set of conventions resting on a particular division of labor which, of course produces the needful energy for that kind of system […]. The sacred makes a fulcrum on which nature and society come into equilibrium, each reflecting the other and each sustaining the known. (Douglas 1986:112-3)

INTERVENÇÃO (e não representação)

Fleck insisted that the development of knowledge depends on how the knowledge is expected to intervene in practical life. Thinking has more to do with intervening than with representing (Hacking 1983). The same applies to ancestors: they are known by their interventions. (Douglas 1986:50)

SÍNTESE DO ARGUMENTO ATÉ AQUI: o dispositivo cognitivo que fundamenta a instituição é a analogia naturalizante

It is well said that individuals suffer from the bounding of their rationality, and it is true that by making organizations they extend the limits of their capacity for handling information. We have shown how institutions need to be established by a cognitive device. Mutual convenience in multiple transactions does not create enough certainty about the other person’s strategies. It does not justify the necessary trust. The cognitive device grounds the institution at once in nature and in reason by discovering that the institutions’ formal structure corresponds to formal structures in non-human realms. (Douglas 1986:55)


First, for discourse to be possible at all, the basic categories have to be agreed on. Nothing else but institutions can define sameness. Similarity is an institution. Elements get assigned to sets where institutions find their own analogies in nature. (Douglas 1986:55)

To make a fresh start from the side of cognition, consider how the most elementary logical idea itself depends on social interaction. This is the idea of similarity or resemblance. When several things are recognized as members of the same class, what constitutes their sameness? […] Comparison of cultures makes it clear that no superficial sameness of properties explains how items get assigned to classes. Everything depends on which properties are selected. (Douglas 1986:58)

Institutions bestow sameness. Socially based analogies assign disparate items to classes and load them with moral and political content. (Douglas 1986:63)


On the one hand, the emotional energy for creating a set of analogies comes from social concerns. On the other hand, there is a tension between the incentives for individual minds to spend their time and energy on difficult problems and the temptation to sit back and let founding analogies of the surrounding society take over. (Douglas 1986:55)


However much they try to insulate their work, scientists are never completely free of their own contemporary society’s pressures, which are necessary for creative effort. Scientific theory is the result of a struggle between the classifications being developed for professional purposes by a group of scientists and the classifications being operated in a wider social environment. Both are emotionally charged. Both kinds of classification depend on social interaction. One (that of the scientists) makes a determined effort to specialize and refine its concepts so as to make them fit for use in a discourse that differs from though it is contained within the entrenched ideas of the larger, encompassing social group. (Douglas 1986:56)

[T]he scientific formulae that emerge always carry the marks of their social origins. (Douglas 1986:56)


A foreign culture may work without having a good scientific classification. The senses in which it may be said to work are political, economic, social, ecological. For the intermeshing of practical purposes, folk classification makes a world that is reliably intelligible and predictable enough to live in. The objectives of folk classification are quite different from those of scientific classification; the latter is developed to express specialized theory generated in specialized institutions, which also have their own foundational ideas and are also grounded in nature. Each group of scientists is able to resist the temptation to rest upon the founding analogies of the outside society only to the extent that it is insulated from it. […] But this archaic religious classification and many other contemporary ones known to anthropologists owe their divisions much more to their capacity to model the interactions of the members of society than to a disinterested curiosity about the workings of nature. There is a fundamental shift to a scientific classification from a socially inspired one. The striving for objectivity is precisely an attempt not to allow socially inspired classifications to overwhelm the inquiry. There can be no smooth transition from the socially inspired to the scientific classification. The first cannot develop into the second by pressing deeper and deeper beneath the surface of things in the quest for knowledge, because the quest for knowledge is not one of its objectives (Levi-Strauss 1962). (Douglas 1986:58-9)


Somewhere the argument is flawed. How can the ability to discriminate between shades of yellow, or to make other judgments of nearness or distance, or of other quality differences, ever lead to putting items into classes? To recognize a class of things is to polarize and to exclude. It involves drawing boundaries, a very different activity from grading. To move from recognizing degrees of difference to creating a similarity class is a big jump. The one activity can never of itself lead toward the other, any more than institutions can evolve toward a complete organizing of information by beginning from spontaneous self-policing conventions. (Douglas 1986:60)

A theory of the world would need to start with dividing, not with grading. (Douglas 1986:62)

O DENTRO E O FORA (a máquina de guerra de Melanie Klein)

In Melanie Klein’s account of an infant’s first attempts to find order in the world, the dominant preoccupation is […] the problem of inductive rightness. It [the baby] needs to pick out of the crowd of present sensations some practical basis for projecting forward (to use Nelson Goodman’s term), a version of the world that works (Goodman 1983). The baby has no habits to rely on, and there is no existing version to be remade. […] Matching samples will not lead to discriminating kinds. According to Klein, the urgent thing is to know which painful and pleasant experiences come from inside and which from outside. The first basis of projectible kinds is the difference between self and not-self (Klein, 1975). (Douglas 1986:62)

The questions it [the infant] asks resemble military intelligence. It needs to know whether the source of milk, if external, is one breast or several, and if several, how to distinguish allies from enemies? Is this the good breast or the bad breast? Is it for me or against me? The earliest social interaction lays the basis for polarizing the world into classes. Survival depends on having enough emotional energy to carry this elementary classificatory enterprise through all the hard work needed to build a coherent, workable world. Social interaction supplies the element missing in the natural history account of the beginnings of classification. (Douglas 1986:62-3)


The institution works as such when it acquires a third support from the harnessed moral energy of its members. More of this in the last chapter. All three processes [intellectual, social, and moral energy] are simultaneously at work. (Douglas 1986:63)

Information theory draws our attention particularly to divergent patterns. It assumes that for any given pattern a prior buildup of energy is needed. A pattern of given complexity, once stabilized, uses less energy than was required to bring it into existence. For example, heat under a pan of water takes time before the water begins to swirl and bubble. If more energy is pumped in, it has to be used up by new patterns of complexity. So if the heat under the pan is increased, the water will swirl around in a more and more complex pattern. There has to be some way of dissipating any energy that is in excess of what is necessary to maintain the pattern (Prigogine 1980). Over and above a certain point, the extra input of energy will not be able to be absorbed by increasing complexity, and there will be a radical change in the whole pattern. For example, the water will turn into steam. To write of institutions as complex patterns of information […], and to think of the relative efficiency of their channels of communication […], should lead to considering the amount of energy used for making a particular kind of institution and how it is deployed in a more complex or less complex pattern. And from here it should lead to assessing the volume of transactions that it is capable of handling. Otherwise, information theory in political science is mere academic window dressing, a new favorite metaphor to replace the outdated functionalist metaphor of the 1950s. (Douglas 1986:112)


At this stage we can start to trace the effects of turning individual thought over to an automatic pilot. First, there is a saving of energy from institutional coding and inertia. […] For example, the common English word, man, with its archaic plural, men, has stood out against the onward sweep of plural endings in s. […] Thanks to the weight of institutional inertia, shifting images are held steady enough for communication to be possible. (Douglas 1986:63)


Feminist theory in anthropology has had a lot to say about these equations as justifying the subjection of women (Strathern 1980). Even when the feminine gender is associated with the more esteemed side, it still can be used to justify the women carrying the heaviest physical burdens. (Douglas 1986:64)

O PROBLEMA DE HUME (critérios de julgamentos de sistemas de justiça)

David Hume’s teaching that justice is an artificial virtue gives a lot of trouble. The idea that justice is a necessary social construct is exactly parallel to Durkheim’s idea of the sacred, but Hume clearly refers to us, ourselves [not aborigines]. He brings our idea of the sacred under scrutiny. Our defensive reaction against Hume is exactly what Durkheim would predict. We cannot allow our precepts of justice to depend on artifice. Such teaching is immoral, a threat to our social system with all its values and classifications. Justice is the point that seals legitimacy. […] For this very reason, it is difficult to think about it impartially. In spite of a wide belief in the modern loss of mystery [e.g. Weber], the idea of justice still remains to this day obstinately mystified and recalcitrant to analysis. If we are ever to think against the pressure of our institutions, this is the hardest place to try, where the resistance is strongest. On this subject anthropologists have a privileged position for they record many diverse social forms each venerating its particular idea of justice. […] Hume’s idea of the artificial virtues is integral to his skeptical program (1739, 1751). It was part of his attack on all theories of innate ideas, whether of causality, natural law, or private property. His radical constructivism makes him exactly the anthropologists’ philosopher. When it is a matter of finding logical structures in nature, Hume says that all we ever see there are frequencies, and from these we form habits and expectations. When it is a matter of natural justice, all we can ever know is that we need regulated interactions; to meet the need we develop principles. Accordingly, the idea of justice is not a natural response as to an emotion or to an appetite. As an intellectual system, it has a kind of second-order naturalness because it is a necessary condition for human society. Fabricated precisely for the purpose of justifying and stabilizing institutions, it is founded on conventions in exactly the sense quoted above from David Lewis (1969). Thus, no single element of justice has innate rightness: for being right it depends upon its generality, its schematic coherence, and its fit with other accepted general principles. Justice is a more or less satisfactory intellectual system designed to secure the coordination of a particular set of institutions. […] If this turns out to be logically unassailable and yet unacceptable to philosophers who are otherwise strong on logic we shall chalk it up as another instance of the power of the sacred to rouse an emotional defense. […] Hume’s approach does not allow us to refuse the name of justice to a system merely because it does not accord with our own. Philosophers can hardly dismiss all civilizations antecedent to our own as defective in moral judgment without seeming to be biased. […] When Hercules Poirot caught the Countess Rossakoff with stolen jewels, she denied any intuitive rightness of private property: “And what I feel is, why not? Why should one person own a thing more than another?” (Christie 1935). The trouble with trying to defend an immutable principle of justice is that not everyone sees the self-evident thing. Rules that now seem to us moderns as monstrously unjust did not strike our forebears as wrong. Slavery and the subjugation of women are vulnerable to the same arguments that Hume used against the intuitive right to property. (Douglas 1986:113-4)

Given that equality as a natural right or as a universal principle of justice is still the most prominent difference between Western and many other systems of justice, it is not enough simply to dismiss all of the latter as obviously unjust. (Douglas 1986:116)

Yet, however vehemently we assert our own principles of justice, they are still the principles that have emerged over the last two hundred years, along with the emergence of an economic system based on individual contract. Turning itself from a horizontal pattern of integration to a vertical one, which depends on drawing independent individuals up from bottom to top, the whole information system has to be transformed. When the perturbation has reached a certain point, the dissipative structures can no longer hold the pattern. First, the founding analogies need revision. Louis Dumont has traced the eighteenth-century effort to refocus its ideology away from organic metaphors. He shows that Mandeville’s parable of the independent industrious individual bees was a landmark in the turning away of Western thought from hierarchical models of society toward justifying individualism […]. […] When the analogy with nature has been changed, the system of justice also needs revision. Now it has to promote the vertical movement of individuals instead of containing them within their horizontal layers. The result has been the sacralization of a society based on an extravagant use of energy unprecedent in the history of the world. (Douglas 1986:118-9)

Without appeal to religion, intuitionism, or innate ideas, it is very hard to defend a substantive principle of justice as universally right. (Douglas 1986:117)

In other words, this feeling is ultimately incommunicable. (Douglas 1986:119)

Rudolph Otto’s justification of religious truth: if the reader has never had a mystic experience, if he has never felt the Mysterium Tremendum, if he is stranger to the sense o£ the numinous, then, says Otto the Lutheran theologian, nothing I can say will convince him: the feeling is incommunicable. (Douglas 1986:119)

According to Hume’s theory, the need for a concept of justice would only arise in certain circumstances. (Douglas 1986:117)

According to Hume, the artificial virtues are to be known by their internal coherence within an abstract system that harmonizes everyday interactions in a particular society. (Douglas 1986:119)

On Hume’s principles we can say that one system is more just than another. We can say it on two counts, one logical and one practical. According to his teaching, a system of justice is devised expressly for providing coherent principles on which social behavior can be organized. So we can compare systems of justice in respect of their coherence. This is the regular task of historical jurisprudence. Judicial reform is often justified on grounds of incoherence among the principles being used. According to Hume, arbitrariness defeats the essential purpose of justice. We can compare the amount of arbitrary rules. So there is no problem on this issue. On the practical count, we can start by asking how well a system of justice actually performs the task of providing abstract principles for regulating behavior. It could be too arcane, too complex, and too ramifying to be understood. […] Or, on another kind of practical test, is the system of justice efficient? Are the courts too remote from the centers of population? Jurists make these and other comparisons of systems of justice all the time. In doing so they are not obliged to apply the validating principles of their own institutions, not at all. The tests of coherence and non-arbitrariness, complexity and practicality, are not subjective preferences. It is as straightforward to study human systems of justice objectively as it is to measure the length of human feet from heel to toe. Systems can be compared as systems. The one thing that it is not possible to do is to pick a particular virtue, say kindness to animals or to the aged, or equality, and find a way of proving that it is always and ineluctably right and best. […] [R]ecognizing the social origin of ideas of justice does not commit us to refraining from judging between systems. They can be judged better or worse according to the good sense we can make of their assumptions. (Douglas 1986:129-1)


The aim of revision is to get the distortions to match the mood of the present times. (Douglas 1986:69)

MEMÓRIA PÚBLICA (a condição de nosso pensamento)

Public memory is the storage system for the social order. Thinking about it is as close as we can get to reflecting on the conditions of our own thought. (Douglas 1986:70)

As Merton’s example shows [multiple discoveries], competitive social systems are weaker on memory than ascriptive ones. This must be so because the competition drives out some players and brings upstarts to the top, and with each change of dynasty, public memory necessarily gets rearranged. By contrast, complex hierarchical society will need to recall many reference points in the past. […] Coherence and complexity in public memory will tend to correspond to coherence and complexity at the social level. This is what Halbwachs taught. The converse follows: the more the social units are simple and isolated, the simpler and more fragmentary the public memory will be, with fewer benchmarks and fewer levels of ascent to the beginning of time (Rayner 1982). […] The competitive society celebrates its heroes, the hierarchy celebrates its patriarchs, and the sect its martyrs. (Douglas 1986:80)

Weak or strong, memory is sustained by institutional structures. (Douglas 1986:81)


Scientists’ thoughts are held in the grip of the exacting institution of science, as ours are held in other institutions. They cannot reflect calmly upon it, and nor can we. We need a technique for standing aside from our own society, developing Merton’s little cybernetic model into a big one with several compartments that deal with the passions inherent in different forms of social organization and that display the control which socially reinforced motivations have upon individual vision. (Douglas 1986:75)


A theory about how the world should be run will survive competition if it is more than a theory, for example, if it can intervene to support individual strategies to create a collective good. (Douglas 1986:73)


Certain things always need to be forgotten for any cognitive system to work. There is no way of paying full attention to everything. (Douglas 1986:76)


A sociological theory of rejection can be more securely based than a sociological theory of value because of the public nature of penalties and prohibitions which follow on negative attitudes. The same is true for our problem. The thinkability of the social order is beset with infinite regress. Institutional influences become apparent through a focus on unthinkables and unmemorables, events that we can note at the same time as we observe them slipping beyond recall. (Douglas 1986:76)

O PROBLEMA DA ORIGINALIDADE NA CIÊNCIA (economia energética por analogia estrutural; competição por recursos escassos=originalidade)

The strategies to validate scientists’ claims use originality as a main criterion for prizes and positions. The belief in a first discoverer is nothing without the prizes and renown. The custom of naming immediately gives a major advantage to claimed originality and a disadvantage to the fact of rediscovery. What seems dysfunctional when enraged scientists make a public display of their vanity may be counted as the cost of keeping the race open to the swift. But competition is always costly in human terms. (Douglas 1986:77)


When it is recognized that a majority could prefer A to B, and B to C, but C to A, confidence in the will of something called “the majority” is eroded. (Douglas 1986:79)


[P]sychologists are institutionally incapable of remembering that humans are social beings. As soon as they know it, they forget it. They often remind one another of how artificial the parameters are that they have set around their subject matter. Famous psychologists keep upbraiding their fellows for despising or ignoring institutional factors in cognition. The literature of the social sciences is sprinkled with rediscoveries of that very idea. (Douglas 1986:81)

James Coleman is another who was prominent in making efforts in the 1950s to treat qualities of the social situation as selective principles for acceptable information. […] Coleman anticipated that the new approach would focus on the fate of information transmitted through more integrated and less integrated social networks […]. However, network analysis has proceeded without bringing the parallel and necessary analysis of attitudes and values to the same heights of sophistication, and no systematic synthesizing theory has been developed. (Douglas 1986:82)

Psychologists […] are so committed to the assumption that individual psychic development is restricted by social conventions that they see all conventional and institutional constraints as wrongful. […] For psychologists, the idea that stabilizing factors could be useful for cognitive and emotional development is unthinkable. […] [I]t is professionally impossible in psychology to establish the notion that institutional constraints can be beneficial to the individual. The notion can be scouted, but it cannot enter the memorable corpus of facts. (Douglas 1986:82-3)

BARTLETT, RIVERS, DURKHEIM: casos exemplares de suas próprias descobertas sobre os pressupostos do pensamento

In his earlier book, Psychology and Primitive Culture (1923), Bartlett had taught emphatically that the individual is always a social individual and that social influences selectively control cognition and emotion. He was already drawing heavily on Rivers’ work and comparing something he and Rivers called “primitive comradeship” with the “collective conscience” of the writers of the L’Annee Sociologique. He described how in primitive society conflict is averted by instituted separation – a pregnant idea – and how curiosity is brought under institutional control. […] One reason why this interest in institutional control on thinking never became more than a speculation lies undoubtedly in certain current evolutionary assumptions. Both Bartlett and Rivers thought (along with Durkheim) that social control of the free ranging curiosity of individuals was stronger in primitive society. The primitive individual was altogether less of an individual and more of an automaton obeying group cues. This evolutionary assumption was quite congenial to the period of colonial empire and provided the latter with its naturalizing analogies. It was self-evident that modern man had lost his natural sensitivity to group signals, just as the human race had lost the sense of smell so useful in lower animal orders. (Douglas 1986:86)


[A]n institution cannot have purposes. […] Only individuals can intend, plan consciously, and contrive oblique strategies. […] Institutions systematically direct individual memory and channel our perceptions into forms compatible with the relations they authorize. They fix processes that are essentially dynamic, they hide their influence, and they rouse our emotions to a standardized pitch on standardized issues. Add to all this that they endow themselves with rightness and send their mutual corroboration cascading through all the levels of our information system. No wonder they easily recruit us into joining their narcissistic self-contemplation. Any problems we try to think about are automatically transformed into their own organizational problems. The solutions they proffer only come from the limited range of their experience. If the institution is one that depends on participation, it will reply to our frantic question: “More participation!” If it is one that depends on authority, it will only reply: “More authority!” Institutions have the pathetic megalomania of the computer whose whole vision of the world is its own program. For us, the hope of intellectual independence is to resist, and the necessary first step in resistance is to discover how the institutional grip is laid upon our mind. (Douglas 1986:92)

The high triumph of institutional thinking is to make the institutions completely invisible. When all the great thinkers of a period agree that the present day is like no other period, and that a great gulf divides us now from our past, we get a first glimpse of a shared classification. Since all social relations can be analyzed as market transactions, the pervasiveness of the market successfully feeds us the conviction that we have escaped from the old non-market institutional controls into a dangerous, new liberty. When we also believe that we are the first generation uncontrolled by the idea of the sacred, and the first to come face to face with one another as real individuals, and that in consequence we are the first to achieve full self-consciousness, there is incontestably a collective representation. Recognizing this, Durkheim would have to concede that primitive solidarity based on shared classification is not completely lost. (Douglas 1986:98-9)

How can we possibly think of ourselves in society except by using the classifications established in our institutions? (Douglas 1986:99)

At the same time as institutions produce labels, there is a feedback of Robert Merton’s self-fulfilling kind. The labels stabilize the flux of social life and even create to some extent the realities to which they apply. […] People have always been labeling each other, with the same consequences – labels stick. […] As fast as new medical categories (hitherto unimagined) were invented, or new criminal or sexual or moral categories, new kinds of people spontaneously came forward in hordes to accept the labels and to live accordingly. The responsiveness to new labels suggests extraordinary readiness to fall into new slots and to let selfhood be redefined. […] It is a […] dynamic process by which new names are uttered and forthwith new creatures corresponding to them emerge. […] [P]eople are not merely re-labeled and newly made prominent, still behaving as they would behave whether so labeled or not. The new people behave differently than they ever did before. (Douglas 1986:100)

[I]nstitutions survive by harnessing all information processes to the task of establishing themselves. The instituted community blocks personal curiosity, organizes public memory, and heroically imposes certainty on uncertainty. In marking its own boundaries it affects all lower level thinking, so that persons realize their own identities and classify each other through community affiliation. Since it uses the division of labor as a source of metaphors to affirm itself, the community’s self-knowledge and knowledge of the world must undergo change when the organization of work changes. When it reaches a new level of economic activity new forms of classification must be designed. But individual persons do not control the classifying. It is a cognitive process that involves them in the same way as they are involved in the strategies and payoffs of the economic scene or in the constitution of language. Individual persons make choices within the classifications. Something else governs their choices, some need of easier communication, a call for a new focus for precision. The change will be a response to the vision of a new kind of community. (Douglas 1986:102)

Something happens to the insides of our heads when a different kind of organization had made obsolete the old classifications […]. The change is not a deliberate or conscious choice. Institutions veil their influence, so that we hardly notice any change. (Douglas 1986:103)

The individual tends to leave the important decisions to his institutions while busying himself with tactics and details. (Douglas 1986:111)

The thing to be explained is how institutions ever start to stabilize. To become stable means settling into some recognizable shape. (Douglas 1986:111)

The most profound decisions about justice are not made by individuals as such, but by individuals thinking within and on behalf of institutions. The only way that a system of justice exists is by its everyday fulfillment of institutional needs. If this be conceded, it would appear that the rational-choice philosophers fail to focus on the point at which rational choice is exercised. Choosing rationally, on this argument, is not choosing intermittently among crises or private preferences, but choosing continuously among social institutions. It follows that moral philosophy is an impossible enterprise if it does not start with the constraints on institutional thinking. So let no one take comfort in the thought that primitives think through their institutions while moderns take the big decisions individually. That very thought is an example of letting institutions do the thinking. (Douglas 1986:124)


The social theory of Max Weber and that of Durkheim illustrate respectively the mixed advantages of leaving institutions to do their own classifying (Weber), and the difficulties of inspecting how they do it (Durkheim). […] Both Durkheim and Weber focused their inquiry on rationality and specifically on the relation between ideas and institutions. For both the main interest was the emergence of individualism as a philosophical principle. In Durkheim’s case the task was to explain the general question of individual commitment to the social order – the issue of solidarity, which is the same as collective action. He found the answer in shared classification. Durkheim’s work on the social origin of classification affords an independent method of self-inspection. It provides a technique for analysis that could be made proof against institutional distortion. For Weber, the task was to explain the prevalence of particular ideas and ideals at a particular stage of institutional development. These remarks already show that Durkheim had placed his inquiry at a higher level of abstraction. (Douglas 1986:93)

Weber‘s sociological golden dawn is a counterpart of Frazer‘s mythological golden bough and of River‘s colonial model of the psyche (1920). If they spoke in chorus, it was because the same institutions were doing their thinking. […] As a contemporary, Durkheim fell into all these institutional traps. He started from the same basic distinction between primitives and moderns and also regarded them as using different mental procedures. It would be stupid to suggest that he did not also subscribe, also with mixed feelings, to the idea of a vanished golden dawn of mankind. The saving grace for him was not to be interested in reconstructing the various phases of evolution that led from the beginning to now. Thus his theory is less heavily loaded with the institutionally given presuppositions. His evolutionary model only has two stages: the primitive stage of mechanical solidarity that is based on shared classifications and the modern stage of organic solidarity based on economic specialization and exchange. […] [W]e are left with two forms of social commitment, one classificatory and one economic. Even Durkheim did not believe that classificatory solidarity was uniquely associated with undeveloped stages of the division of labor, for he devoted much attention to standardized ideas of right and wrong in modern society. (Douglas 1986:95-6)

O PROGRAMA DE DURKHEIM (fit+coherence)

Durkheim’s program of research starts from the possibility that either there is a good fit or a bad fit between the public and the private classifications. If the fit is bad, it can be for two different reasons: the individual may reject the public classifications and refuse to let them have any hold upon his own judgments; or the individual may accept the worth of the public classifications, but know that he or she is incapable of meeting the expected standards. Lastly, the public classifications may be relatively coherent or in a state of incoherence. (Douglas 1986:97-8)

What constitutes deviance cannot be asserted until the dimensions of conformity have been delineated. To assess degrees of conformity among ourselves, we must make the same meticulous count of categories; tracing the way the physical world is turned into a projection of the social world. It is the same for us as for the Eskimos and the Australians; we must use the same method of constructing the north and the south, the right and the left, all loaded with the patterns of dominance, congregation and dispersal, for ourselves as well as for the Chinese and the Zuni Indians. (Douglas 1986:98)


Hacking is drawing a distinction between the effect of description on inanimate objects and the effect of names on humans. […] However, the contrast is not so clear […]. The real difference may be that life outside of human society transforms itself away from the labels in self-defense, while that within human society transforms itself towards them in hope of relief or expecting advantage. (Douglas 1986:101)

The interaction […] goes round, from people making institutions to institutions making classifications, to classifications entailing actions, to actions calling for names, and to people and other living creatures responding to the naming, positively and negatively. […] Having accepted that persons classify, we can also recognize that their personal classifying has some degree of autonomy. (Douglas 1986:101-2)

This is how the names get changed and how the people and things are rejigged to fit the new categories. First the people are tempted out of their niches by new possibilities of exercising or evading control. Then they make new kinds of institutions, and the institutions make new labels, and the label makes new kinds of people. (Douglas 1986:108)


Large-scale industrial processes are their own institutions. They cannot be embedded in the patterns of local, community control. (Douglas 1986:108)


The comparison of classifications as an index of other things that are happening in our own society provides a small, provisional ladder of escape from the circle of self-reference. We can look at our own classifications just as well as we can look at our own skin and blood under a microscope. We can recognize regularities appearing in whole arrays of classificatory work, just as well as grammarians can study regularities in syntax and phonetic shifts. There is nothing self-contradictory or absurd in taking a systematic look at the classifications we make of ourselves. The logical difficulties start when we try to develop value-free ideas about the good society. And yet these difficulties must be met if we are not to leave the whole inquiry in a stew of philosophical relativism. It is not at all the purpose of this book to teach that because institutions do so much of our thinking there can be no comparisons between different versions of the world, still less to teach that all versions are equally right or wrong. (Douglas 1986:109)


[T]he functioning of a society depends on some degree of coherence and […] an abstract summary of the interlocking principles on which it works promotes coordination. Once formulated the artifice acquires venerability. Durkheim could explain why […] justice seems to have been there forever. It had to have existed long before humans came into the world; so it appears old and immutable as one of nature’s fixtures, above challenge. (Douglas 1986:120)

O CALCANHAR DE AQUILES DE DOUGLAS (o mesmo de DURKHEIM): a naturalização implícita da ideia moderna de natureza

At this point the question of moral relativism has merged into questions about what is real and what illusionary in the world. I hope there is no need to get into the argument about realism. What has been said above does not throw into doubt that there are objective tests of right and wrong versions of the world and how it works. For example, imagine a system of justice that punished people for what they are alleged to have done in other people’s dreams. It would not be difficult to show that such a system draws the lines of responsibility according to a wrong version of reality and a wrong version of human accountability – so much so that it could not be organized coherently on any practical issue. The way that humans are, the facts that they walk upright and cannot be in two places at once, are incorporated as part of any system of justice. Some experience and study of the conditions of life have gone into the background of the thinking. All that is being argued here and throughout this book is that this cumulative experience of the world should explicitly incorporate the social nature of cognition and judgment. (Douglas 1986:121-2)


The preferred assumption, which implies that humans are not essentially social beings, is strong enough to prevent us seeing how they actually behave. What happens when law is abrogated? Does nature take over? […] Hume himself supposed that in a famine each would seize what he needed to survive, throwing concepts of private property to the winds. Part of his demonstration of their artificiality was to show that criteria of justice would be suspended when it is a matter of starvation. Other philosophers agree. But starving people do not rise up and seize the food that is there. Sheer force is not all that stops them from looting the stores. Within the family or village in such a crisis who starves and dies or who eats and lives is neither quite random nor dependent on force. Strongest and most numerous do not always take all when the tragic crisis arrives. History shows that famine does not automatically revoke conventions. It does not usher in something like a natural law of equal rights. By adopting such an assumption we naturalize our own ideas of equity; it is as if we assume that when nature takes over, she does what we knew we ought to have done all along, that is, to distribute equally. Crisis behavior depends on what patterns of justice have been internalized, what institutions have been legitimated. (Douglas 1986:122)

To give out the food as quickly as possible, existing channels of distribution would be the most efficient and most acceptable to the famine-stricken country. But no! As soon as the local people are brought into the relief scheme, the food gets diverted. The poorest are always the most vulnerable in a famine. But the food does not reach them. Hoarding, stealing, exploiting, recrimination, and self-righteous indignation are part of the grim story of famine relief. (Douglas 1986:122)


When individuals disagree on elementary justice, their most insoluble conflict is between institutions based on incompatible principles. The more severe the conflict, the more useful to understand the institutions that are doing most of the thinking. Exhortation will not help. Passing laws against discrimination will not help. It did not help African women for the League of Nations to pass resolutions against polygamy or female clitoridectomy. Preaching against wife battering and child abuse is not more likely to be effective than preaching against alcohol and drug abuse, racism, or sexism. Only changing institutions can help. We should address them, not individuals, and address them continuously, not only in crises. […] Between institutions of the same kind, based on the same analogies from nature, and sealed with the same ideas of justice, diplomacy has a chance. But diplomacy between different kinds of institutions will generally fail. Warnings will be misread. Appeals to nature and reason, compelling to one party, will seem childish or fraudulent to the other. (Douglas 1986:125-6)


Once it were conceded that legitimated institutions make the big decisions, much else would be changed. […] Instead of moral philosophy starting from a notion of the human subject as a sovereign agent for whom free choice is the essential condition, Sandel suggests that the human agent is essentially one who needs to discover (not choose) his ends, and that the community affords the means of self-discovery. Instead of being centered on the conditions of choice, a different kind of moral philosophy would be centered on the conditions of self-knowledge. (Douglas 1986:126-7)


Only by deliberate bias and by an extraordinarily disciplined effort has it been possible to erect a theory of human behavior whose formal account of reasoning only considers the self-regarding motives, and a theory that has no possible way of including community-mindedness or altruism, still less heroism, except as an aberration. The Durkheim-Fleck program points to a way of return. For better or worse, individuals really do share their thoughts and they do to some extent harmonize their preferences, and they have no other way to make the big decisions except within the scope of institutions they build. (Douglas 1986:128)

Tags :