The thick description of Clifford Geertz: An anti-scientific Approach to Anthropology

: In this article, I analyse the ‘thick description’ notion developed by Clifford Geertz to explore its scope and limitations. It is not my intention to refute or consider invalid the approach developed by Clifford Geertz; on the contrary, it is on his basis that I propose some corrections that are loose in his epistemological system. Some of his notions can be clarified, and others are wrong. I will discuss his (i) ontology, (ii) individualistic method, (iii) arbitrary notion of truth and (iv) the development of what a ‘thick description is’. To analyse this concept in detail, I evaluate how it is operationalised in some of the author's texts. The analysis shows its limited explanatory capacity due to its idealistic approach to the concept of culture, which overlooks other approaches in anthropology. Finally, I conceptualise thick description by adding materialistic properties to it.


Introduction
It is not the consciousness of men that determines their existence, but their social existence that determines their consciousness (Marx, 1977, Preface). Clifford Geertz was one of the most influential anthropologists of the 20 th century, and with his figure, interpretive or literary anthropology became the trend in the discipline. His ontology poorly follows Dilthey (1989) and Weber (1978), considering that social facts are subject to interpretation. According to hermeneutics, society is a text (Liakos, 2019). Thick description is developed in his classic The interpretation of cultures . Therefore, my analysis will be focused on that text. Still, I will expand it to other theoretical and empirical works, such as From the native's point of view… (Geertz, 1974) or Ritual and social change… (Geertz, 1957), among others.
I know that stating a premise as strong as "anti-scientific" is, to say at least, problematic. Still, I assume that is because of the hatred that the so-called humanist disciplines took regarding science, as Reyna (1994Reyna ( , 1997 clearly stated in two articles almost thirty years ago. Some of Geertz's notions can be clarified, and others are wrong. I will discuss his (i) ontology, (ii) individualistic method, (iii) arbitrary notion of truth, and (iv) the development of what a 'thick description is.' (i), (ii) and (iv) are scattered in his works, and they lack a connection between them to call it a theory. Point (iii) happens randomly and can only be related to intuitionism without any scientific consensus.
Some words cannot be used freely: the intuition in Geertz is related to some of his intellectual influences, in this case, Schutz and Husserl. Studying why human beings make sense of their acts is not futile by any means. Still, it is curious the family resemblance that Clifford Geertz shares with other Schutzenthusiasts: the Austrian school of economics that had a reciprocal relationship with the German philosopher (Kurrild-Klitgaard, 2001). I have no interest in digging into the ideology of these scholars, but rather their epistemology, especially Ludwig Von Mises. If we read Human action (Von Mises, 1949), we will find a family resemblance with Clifford Geertz: both scholars intend to build a 'different' science, that is, one that is away from the 'regular' one. Mises will call this 'praxeology', while  will say that social sciences can only advance by sophisticated interpretations. There are essential differences between these two authors, but both will rely on ideas, concepts, and meanings to develop their ideas.
Nevertheless, Geertz has his main philosophical influences from two distinctive scholars: Max Weber and Ludwig Wittgenstein. Weber will be used mainly for describing what culture is. Still, his method, even his empirical examples of ideal types as seen in Economy and society (Weber, 1978), are abandoned and replaced for an arbitrary interpretation of social life, as I will develop further in this article. Wittgenstein's influence is explicit: a concept's meaning only has sense if we consider the context in which it is produced. Even if we give detailed attention to what  says about his influences, his approach is not far from hermeneutics. In fact, he is trying to draw an interpretive theory of culture, which is a logical consequence of a constructivist approach. This will be clear in the following sections.
Before starting, I would like to clarify some concepts. A critique of the interpretive approach is commonplace to treat a scholar as a 'positivist' in a pejorative manner. I have no relation with that tradition other than considering that social life can be studied with experiments, testable hypotheses, and measurable assumptions; in any case, my position is methodologically positivist (Riley, 2007), something I do not find a sin. In any case, this assumption is reductionist, to say at least, but 'positivism' effectively attacks anyone against the ontological turn in social sciences (Inside Higher Ed., 2010). Far from the positivist tradition, I find myself in the systemic approach developed by Mario Bunge (1979-89) in his treatises; therefore, I do not consider the interpretive tradition as science and, again, this will drive more clarifications.
So, what is science, and what is a theory? Again, as mentioned before, since my approach is Bunge's sistemism, science definition in this system has ten different properties, and its futile to develop them here (see Bunge, 1991), and the same goes for the definition of what a theory is, but this can be summarised in a system of hypotheses. It is sufficient that a few of them are true or approximately true to lead to appropriate consequences, provided that the false ingredients are not used in the deduction or are practically harmless (Bunge, 2000). Specifically, if we talk about anthropology, there are opposite traditions to the interpretive -even if the latter is mainstream nowadays -such as mathematical anthropology (Read, 2019 ), or ecological anthropology (Moran, 2018), only to name two examples. Further explanations about the difference between Weber and Geertz will be developed in this article, as the confusion on their approaches is a place hard to avoid.

Essential problems with 'thick description'
What is a thick description? If we conceptualise this notion, scattered in The interpretation of cultures  would be: the (i) constructions of other people's constructions; (ii) sorting out the structures of signification; (iii) determining their social ground and import, (iv) microscopic and (vi) aiming to generalise. These are the essential points of this method pointed out by Geertz, although he refers to it as a theory.
One of the most significant issues with this approach concerns two intrinsically related assumptions on Geertz: truth and interpretation. With the basis of (ii) we do not know what 'sorting' a meaning is. We can only infer it. According to , we can only perceive the social code or not. There needs to be a specification on how to do it. It could be developing a literature review when the author states that theoretical ideas are adopted from related studies (Geertz,1973), but that refers to explaining the phenomena, needing to understand what we are about to study. We can only say that this can be solved by extended fieldwork, but we must know history, politics, or what? Everything needs to be clarified. This vagueness leaves (ii) unsolved, or, as Geertz (1973, p. 24) said: "You either grasp an interpretation or you do not, see the point of it or you do not, accept it or you do not". In that case, point (iv) could be helpful for further clarification, but how can that be possible if we do not know what we are studying? Geertz never states whether he studies individuals, groups, or their relationships. The problem is unsolved with a surprising claim: Geertz goes 'back and forth', according to his 1974 lecture, From the native's point of view...
Stating that culture is a web of meaning leaves an epistemological assumption that we can only develop interpretations about the senses that individuals give to their life: culture is a text, and we can only interpret it. Points (ii) and (iii) connect dangerously. If we do not know what sorting a structure of signification is, how can we be sure we are doing it right? Geertz adopts -conscious or unconscious -an irrationalist gnoseology: how can I know that my work is correct? Because I have the intuition of that. And what if it is wrong? We never know, and it is just a matter of 'sorting' things out.
Let's analyse some examples. The well-known cockfight in Bali would represent a drama, a theatre play, that evokes masculinity in Mali's male population. Logically following his psychoanalytic framework, a vital masculine impulse is reflected on the cocks: the metaphor is obscene but has no real logical connection. How is an entire population able to condense their emotions in a cockfight? That essentialism could be said about any sport and some populations chosen randomly. What about politics, interests, speculation, treason or other 'winks' -in Geertzian jargon -that we, maybe, cannot see?
Suppose in a cockfight I must ally with my kingroup, which is a struggle for status rather than money. In that case, we have the same discovery that Marcel Mauss or Franz Boas (1896) did with the Potlatch (extensively analysed in Codere, 1956), and they did not claim to use a 'thick description' when culture is a 'semiotic concept'. These researchers worked when the discipline was in its beginnings.  did it more than fifty years later.
I am not the first critic of this problematic side. Crapanzano (1986) points out that assuming that the Balinese people can be taken as entities as if a community was speaking for itself has no logical point, and it is nothing but a random association that the Geertz does relating citizens with their animals. But I consider that this critique has minimum flaws: when  is saying that the Balinese did not see him and his wife, but actually they were being studied, there is no contradiction: the author is only using a metaphor (Crapanzano, 1986). Reynoso (1995) has the same arguments: the cockfight in Bali only addresses reductionist interpretations of men and their animals by rhetoric.
The dogmatic interpretive approach hits a threshold of epistemological vacuum. Let us assume that social science progresses by 'the sophistication of interpretations' and, in the end, would lead us to generalisation. It would be helpful to know the meaning of 'generalisation' in this case. Is that an explanation? Would that explain poverty in a tribe? No, it would be only beneficial to analyse how individuals perceive poverty. Still, as there are no objective criteria to know if my description is correct, I will have to be guided by a hunch -"[...] you grasp it or not" (Geertz, 1973, p.24). There is a solution for this, but it is worth being clear about its vagueness. Two of them could be doing extensive fieldwork or studying specialised literature about the subject. On the contrary,  is telling us, basically, that we have to stay alert for any event that could happen in the society that we are studying: we better grasp the meaningfulness of it, or we will fail in our anthropological task.
Then, what is culture for Clifford Geertz? If we conceptualise its essential properties, the definition would be as follows: (i) a semiotic concept with (ii) webs of significance that are (iii) public and (iv) fictional. If culture is a 'semiotic concept' because we can only interpret it, we might as well leave aside work and reproduction to explain why a society can reproduce or perish. In that sense, assuring food would be a text. The lack of explanatory power is astonishing. It is 'public', as everyone can interpret it. Moving to generalisations, is this related to individuals, societies, or both? Geertz (1974, p. 43) defines it as: " [...] hopping back and forth between the whole conceived through the parts which actualise it and the parts conceived through the whole which motivates them, we seek to turn them, by a sort of perpetual intellectual motion, into explications of one another". Supposedly, this is part of a hermeneutic circle, and we must understand their "[...] symbol system" (Geertz, 1974, p. 45).
Defining culture as fiction  only because it is part of human creation could be interesting at first sight. Still, we can easily remember that Dilthey had the same thought in a detailed work by Helle (1995): nearly a century later,  is saying the same, forgetting that labour, reproduction, and technology are part of the culture: his idealist approach puts aside these three randomly named factors that are materialistic. 1 The scholar's assumption begs the question: culture from who? Is this product of social reality? Is it in the individual's Facundo Acta Scientiarum. Human and Social Sciences, v. 44, e65730, 2022 mind? We only see that culture is 'public'. Even Dilthey (1989) clearly stated that the science of individuals had to consider their psychophysical reality, which formed history and society. He was a methodological individualist but explicit in his conception. There is a materialist anthropologist that covered the meanings of the natives. In his famous work, The making of great men… (Godelier & Maurice, 1986), Godelier explains social domination among the Baruya and does not leave aside their symbolic practices. The meaning individuals give to their practices can even be found in classics such as Sahlin's Stone age economics (2013) 2 .
Apparently, culture is everywhere, so there are no clear paths to start. Nevertheless, there are clues. Following his study of religion in Ritual and social change… (Geertz, 1957), the transformation of how a funeral takes place is related to institutional changes. Before this -relatively simple -conclusion, we are stricken with gossip in a tribe: an unnecessary novel when the causal relation is easy to see. That is not an isolated fact. In the famous From the native's point of view… (Geertz, 1974), 29 pages of 31 are mere attacks with no indicated names and several literary tools for filling the text, something with no sense when the conclusion is simple as stating that other ways of interpreting the world have to pass through our bias. Nothing new in philosophy, but even in Anthropology: Boas (1896) said it before, and he had a coherent theory, as it is detailed in Smith (2015, Chapters 1 and 2) Attacking or making assumptions with no reference is common in Clifford Geertz. In Local knowledge… (Geertz, 1983), he builds a strawman pointing out that, supposedly, culture was not recognised as measurable or attributable to causes only because of hermeneutics, historicism and psychoanalysis (Geertz, 1983). What is happening in other sciences regarding culture? We do not know, and the author deceives us. Eight years before, Benditt (1975) noticed at least seven usages of the term 'interest' in political science directly linked with people's meaning of their activities. On the same page, the author claims that social sciences -in plural -turned to study particular cases offering no evidence of such a statement. The same happens in The interpretation of cultures , and there is no point in discussing every strawman that Geertz does.

Interpretation as an explanation: a problem unsolved
If we define explanation as discovering the mechanism of why x causes y, an interpretive approach is not an issue in this case. We can quickly turn to Weber's classic work explaining the development of capitalism because of religion (Weber & Kalberg, 2013), even if his hypothesis is debatable (Sparhawk, 1976). And at first sight, it is not a problem with Clifford Geertz either because he states that theoretical ideas are not created in a vacuum; they are: "[...] refined in the process, applied to new interpretive problems. If they cease being useful with respect to such problems, they tend to stop being used and are more or less abandoned. If they continue being useful, throwing up new understandings, they are further elaborated and go on being used" (Geertz, 1973, p. 27).
This definition is interesting because it proposes a dialogue with previous research and puts the investigator in a position where it is mandatory to be aware of what was written about the phenomena he is about to study. Nevertheless, this is only interpreting interpretations. Apart from the observations mentioned above, Geertz does not develop a theory per se: how does that social interpretation explain something? Just by adding discovery to the specialised literature? The author mocks about a 'general theory' -once again, a strawman ) -but he never specifies how we can generalise. According to him: it "[...] is not to codify abstract regularities but to make thick description possible, not to generalise across cases but to generalise within them" ( Geertz, 1973, p. 26). How does this happen? If Anthropology is interpreting symbols, then there is nothing to do about the low life expectancy rate of native populations worldwide. The explanation would be institutional, in any case, but the symbolic aspect of it will tell us something about consequences, not causes. That is the outcome of a semiotic -discursive -concept of culture.  conceived the thick description as a theory. If we follow Bhattacherjee (2012), a thick description could be a logically connected system of constructs and prepositions; the problem is that any link we can find is vague. In the first place, if culture is a semiotic concept, then everything is subordinated to what people think about their activities: the world is a group of ideas. How, why and when these notions were produced are not a problem for . At least it is not explicit, so we can only trust people and their meanings. As we can only guess the 'social ground' of facts, our assumptions can be wrong, so generalising them could be pointless.
Another level of the thick description is ethnography: in fact, that is what anthropologists do, according to . This position leaves theory with no ethnography as invalid labour -at least is implied -and the list of cultural anthropologists that equals anthropology and ethnography is exhaustive. 3 This may be a result of the lack of systematisation in , but it is evidently negative: the consequence is the impossibility of constructing theories about culture. This is the exact opposite of what he does in chapters 11 and 12: to analyse the meaning of politics for Indonesians,  examines history in detail and evaluates empirically how the Weberian concept of 'legitimacy' operates. In fact, in this exhaustive analysis of how a state is built, its political, cultural and legal objectives were not mentioned in the thick description conceptualisation. Instead, we only have a vague relation with this if it is in the structures of meaning. They have more or less 'social ground and import'. How can we justify what we are studying? Only because the ethnographic authority says so. There are several ways of choosing cases in social sciences that are far from the scholar's taste. Goertz (2017), Gerring (2008), and several authors developed methods for measuring why a case is relevant.
In the end, what is Geertz doing? He claims that 'interpreting', 'analysing', and in relation with the study object in Local knowledge… (Geertz, 1983). What is the meaning of that? What is the object 'itself' that he is pretending to know? It is pointless trying to interpret what the author is saying, which is counterproductive. By not having an objective point where analysis is supposed to begin or end, mistakes are bound to be made. Geertz argues that there is an involution in agriculture -taking Bali as a single case -but the problem is that this is not true in Indonesia (White, 1983). Even the concept is not valid in itself (see the works of Alexander and Alexander, 1978;Kahn, 1985;White, 2018). Criticisms of Geertz's arbitrary interpretations of life in Bali are summarised in Gordon (1992). Reyna (1994) highlights the bold gamble of literary anthropology in explicitly speaking out against science to falling back on mere interpretations that, in the end, are nothing more than gossip between individuals. The semantic vagueness and logical inconsistencies can be read in Reynoso (1995).

Constructive conceptualisation
As I stated before, the thick description is not a theory; it is instead a method to understand social facts in terms of meaning, i.e., the interpretation that individuals give to their acts. The method does not explain social reality but is a valuable instrument for understanding what happens in realities we cannot observe. We could see a glimpse of this kind of development when Geertz worked on how politics were perceived in the construction of a state and how different notions of what public affairs were ended up in conflict: the only reason why we could understand that is that the author used history for explaining what was happening in Indonesia.
Going back to Geertz's conceptualisation, we can see some negative aspects that can be easily corrected. First, if we are analysing 'constructions of other people's constructions', the method would inevitably be idealistic. The interpretations of the individuals that we are studying are always in a social system that makes them possible: it is only by knowing the economic, political and social issues that we can understand the meaning. If kinship is relevant in a specific social system, we cannot ignore it, as well as the economic and political organisation. It is curious that he does it when he explains why a political change happens, but this is never stated in his theory.
If this is correctly developed, we will understand that a phenomenon's economic or historical ground is more or less relevant if we know its story. Geertz always starts working with epoché, describing facts that, in the end, have a relation with history. We can see this clearly in two of his significant works about Indonesia: Agricultural involution: the processes of ecological change in Indonesia (Geertz, 1963) and Negara (Geertz,1980). The problem is that his concept of culture, only referring to the meaning that the actors give to their practices, puts what we study in symbols, not in what produces them. Therefore, it is useless to explain why non-western societies have different economies, kinships, and so on, invalidating not only classical anthropology but contemporary Marxist anthropology.
The thick description, in the end, is understanding meanings as a result of considering the historical and political history of the community we are studying: this is what Geerz does, invalidating his own supposed theory. Suppose we state the contrary on his premises. In that case, we are saying that anthropology is not allowed to study the history of non-western institutions, doing comparative studies of tribal economics, for example, but only the meanings that their actors give to them.

Conclusion: the consequences of thick description
It would seem reasonable to appeal to the following argument: Clifford Geertz successfully applies his method and, in short, is critical of his philosophical arguments, so any critique of thick description fails to develop against the author's ideas in later works. Let's examine thai in detail. Geertz's ethnographic and theoretical works fall into an idealism that hides power relations, his historical approach is poor, and his theory of knowledge is vague. Concerning Negara... (Geertz, 1980), Barth (1983) points to the ahistoricism of the work, while Howe (1991) stresses that material causes are ignored.
In his influential Religion as a cultural system , there is no clear definition of religion, so I had to conceptualise it taking the spread pieces along the work: so, religion is (i) a system of symbols; (ii) anyimical states and motivations; (iii) an order of existence; (iv) efficiency; and (v) has realism. The properties are absurd in their obviousness. Any symbolist, Turner (1970), for example, can approach a religious 'system' and that definition would be tacit. For , on the other hand, ideology is nothing more or less than the preservation of identity. This is common for someone who referred to systems for much of his career. Still, in Available light... (Geertz, 2000), a philosophical work, argues that, as an ethnographer, he was doing ethnography, not systems.
What, then, is there to Clifford Geertz's philosophy? A hermeneutic chaos that is summarised synthetically in After the fact... (Geertz, 1995): in this book, the author claims that it is futile to inquire into his theory and methodology, as only a little will be found. This is consistent with the position adopted in Works and lives... (Geertz, 1988), where what matters for the anthropologist is what the actors say, the rhetorical force of this performativity. Magically, material causes, the study of systems, models, or dialogue with other disciplines is nullified, calling for the endogamic reproduction of anthropology. Clifford Geertz's interpretative approach is no more than hermeneutic intuitionism laden with flowery adjectives, with only the possibility of making any conceptual sense of it if one curiously seeks to interpret the author's theory. His work clearly sparked discussions among renowned scholars, and as a result, remarkably specialised literature has been written. The same happened with psychoanalysis, phrenology or the Austrian school's business cycle theory. The books on these topics are countless: well, this does not make such approaches valid. Fifty years after attempting to develop an interpretative theory of culture, we, fortunately, have the tools to put it aside, at least as far as science is concerned.