Ethnography?

This essay focuses on the core of ethnographic research—participant observation—to argue that it is a potentially revolutionary praxis because it forces us to question our theoretical presuppositions about the world, produce knowledge that is new, was confined to the margins, or was silenced. It is argued that participant observation is not merely a method of anthropology but is a form of production of knowledge through being and action; it is praxis, the process by which theory is dialectically produced and realized in action. Four core aspects of participation observation are discussed as long duration (long-term engagement), revealing social relations of a group of people (understanding a group of people and their social processes), holism (studying all aspects of social life, marking its fundamental democracy), and the dialectical relationship between intimacy and estrangement (befriending strangers). Though the risks and limits of participant observation are outlined, as are the tensions between activism and anthropology, it is argued that engaging in participant observation is a profoundly political act, one that can enable us to challenge hegemonic conceptions of the world, challenge authority, and better act in the world.

"That's enough about ethnography!" says Tim Ingold (2014). It was a provocation to those who value ethnography, but it seems to me that the substance of the debates that have ensued, in the Cultural Anthropology Forum and in this volume of Hau, shows more agreement than disagreement with what is special about the process of our fieldwork and writing. In this essay I seek to clarify why ethnographic research carried out by anthropologists is important beyond the confines of our own discipline, why how we do it has the potential to contribute new knowledge about the world, why-as I will argue here-participant observation is a potentially revolutionary praxis.
I am entirely in agreement with Tim Ingold and Signe Howell that when scholars in other disciplines-whether it is in geography, sociology, or education-talk about ethnographic research, it does not always come with the same potential as an anthropologist's fieldwork, even when it is more than a set of open-ended interviews, even when they claim to have spent a year doing it. There are, however, brilliant exceptions. Two books that attracted me to anthropology were written by scholars trained in other disciplines: Paul Willis's Learning to labour (1978) and James Scott's Weapons of the weak (1985). I will return to these books, but it is true that ethnography is often used so loosely in other disciplines (or by marketing firms for that matter) that one cannot assume much more than attention to qualitative research when the word is evoked.
I also empathize with Ingold's sentiments that there is a danger that anthropology can become too introverted, too obsessed with naval gazing in "the theater of its operations" and with mystification. There are self-imposed threats that can reduce what we do to mere qualitative data collection or producing case studies. This situation can arise from our failure to recognize the significance of our praxis, our diluting the rigor of our fieldwork, and our own apathy.
This nonrecognition, dilution, and apathy are compounded by forces from within our own universities and funding councils, forces that today prioritize very corporate notions of time-space efficiency in research, writing, and its assessment. In this era of metric measures and "value for money, " I know from personal experience that it is not uncommon for eSRC Grants Assessment Panels to raise questions about the cost of field visits if you can contact someone on the phone or through a computer, to question whether it is necessary to go and speak to a person face-to-face! It is undeniably evermore difficult to continue to make the case for research that prioritizes living for a year and a half with one group of people in an open-ended study, beginning with the premise that we can't possibly know what we will find or even what the right questions will be. Moreover, with language learning centers being forced to close, it is difficult to argue for the time (forget cost) necessary to learn the often-obscure languages of the people we work with. And, even if we tell our doctoral students that we expect them to eventually discard their proposals because of the new knowledge they will acquire in the process of fieldwork, that fieldwork itself will produce not only new research questions but also new fields of enquiry that they could not possibly conceive would be important from the corridors of our departments, we have still compromised by ensuring they produce prefieldwork proposals that in form resemble those of other disciplines-with research themes, questions, and sometimes hypotheses. In the compulsion to work beyond our disciplines and work with activists, practitioners, and in interdisciplinary teams, we are often under pressure to produce fast results and quick-fix solutions. It is easy to give in, to compromise, to not fight our ground.
My own fieldwork with Maoist-inspired Naxalite revolutionary guerrillas in India has taught me that self-criticism should be a crucial part of any struggle and that it is important to recognize the limitations of our allies. To that extent, Ingold is correct to call us up on where we may be faltering and to critically analyze distant others who claim to be close to us. but rather than waste our time disparaging over what they are doing in the name of ethnography, we should take courage from the fact that we are recognized, that others are trying to emulate what they think we do (even if it is not the same), that our potential is celebrated; that bbC Radio 4 even have an annual ethnography Prize. These are our friends not our foes.
We do need to be ever clearer about what we do, why we do it, and why it is important. And to that extent Ingold has helped us by generating this debate. Rita Astuti puts it the most clearly. "We do fieldwork. We write ethnography. " In this spirit, rather than fuss around about whether we need more or less ethnography, dillydallying hither and thither about whether it is too corrupted or should be resurrected, my proposition is that we keep our eye on the ball and focus on the strength that does in fact unite us. both Tim Ingold and Signe Howell have identified what this is. I would like to take their arguments further and propose that we are the inheritors, practitioners, and proponents of a potentially revolutionary praxis; and that praxis is participant observation.
Participant observation can be revolutionary praxis for at least two reasons. The first is that through living with and being a part of other people's lives as fully as possible, participant observation makes us question our fundamental assumptions and preexisting theories about the world; it enables us to discover new ways of thinking about, seeing, and acting in the world. It does so by being inherently democratic not only because of its pedagogy of a two-way process of exchange between educator and educated but also because it ensures that we explore all aspects of the lives of the people we are working with, recognize their interconnections. For instance, we can't just produce a study of voting behavior or labor relations or market activity, but must explore the interrelations between all the different aspects of life-politics, economics, religion, and kinship that matter to those we are studying-to understand any one issue. Second, a point to which I will return in my concluding section, is that by taking seriously the lives of others, participant observation enables us to understand the relationship between history, ideology, and action in ways that we could not have foreseen, and is therefore crucial to understanding both why things remain the same and in thinking about how dominant powers and authority can be challenged, that is crucial to revolutionary social change.
Others may borrow from us, may try to emulate us, are welcome to join us, but it is we-in the discipline of anthropology-that even when the forces are stacked up against us, who create the space for and fight for participant observation.

Why is participant observation a potentially revolutionary praxis?
Anthropology is often critiqued by those who don't know it or can't understand its potential to produce at worst detailed descriptions or at best interesting case studies-Ingold's charge of ethnography. While there is nothing wrong with a case study, the implication is that we can't produce much else through our localized fieldwork. This supposition misses the point that unlike other disciplines-which most commonly rely on finding data to confirm or negate preexisting theoretical frameworks or hypotheses-the in-depth and holistic experiences and understandings of participant observation provide to us the possibilities of reaching general propositions we could never have arrived at before we embarked on fieldwork and to thus question dominant theoretical and political positions.
Since I do not wish to argue that participant observation as revolutionary praxis is the preserve of anthropology, let me illustrate with examples from the two ethnographies I began this essay with, to show how and why the in-depth and holistic research of participant observation can lead to new general propositions. In Learning to labour, Paul Willis (1978) challenges the proposition that working class kids are left to working class jobs because they are less capable of acquiring the skills to move on. Through participant observation with a group of working class boys in their last year-and-a-half at school and their first few months at work, Willis shows how in fact the lads recognized that there was no such thing as an equal opportunity for them-that no matter how hard they tried, they would always be far less successful than middle-class students. They thus developed an antagonism to the ethic of modern education as enforced in the school system and developed a counterculture against it that came across in their linguistic and visual expression and style. Over the formal knowledge prioritized in school, they glorified hard manual labor and privileged practical knowledge, life experience, and "street wisdom. " They challenged obedience and shunned meritocracy for they knew all too well that their fate was not going to be determined by their individual ability to acquire a skill but the ways in which the labor market works to marginalize them and was beyond their control. Not only does Willis show the complex dynamics through which working class kids get working class jobs but also illustrates how their youth culture of resistance in fact buttressed the status quo.
Similarly, in Weapons of the weak, James Scott (1985) challenges dominant theories-in this case, theories of class struggle that look for or focus on large-scale resistance and epochal events-by living for eighteen months with Malaysian peasants. Instead, he shows how with the polarizing effects of the Green Revolution that spread capitalism in agriculture in Malaysia, the far more significant form of class struggle was the everyday resistance that stopped short of outright defiance-the foot dragging, the lying, the arson, the stealing from the rich, the gossip about and the jokes that are told of the rich-that most often were unseen by the powerful groups. The Malaysian peasants were dominated without hegemony but their everyday forms of resistance would-without Scott's participant observation living with them-have gone unnoticed by those coming in looking for symbols, outcomes, and rhetoric that match the book theories of class struggle.
Of course, these books-like all good books-are not without problems or critiques, but what they illustrate is that participant observation is a potentially revolutionary praxis and understanding why entails a consideration of the relationship between method and theory. Willis once said about participant observation's relation to theory, "It has directed its followers towards a profoundly important methodological possibility-that of being 'surprised, ' of reaching knowledge not prefigured in one's starting paradigm" (Willis 1980: 90, cited in Malkki 2007. Indeed, I would argue that participant observation is not merely a method of anthropology but is a form of production of knowledge through being and action. It is thus praxis, the process by which theory is dialectically produced and realized in action. It has more in common with and is better thought of in relation to Marx and Gramsci's "Philosophy of Praxis" (Thomas 2015), Mao Zedong's On practice (2008), in which knowledge arises from practice, or Paulo Freire's reflection and action in the Pedagogy of the oppressed (1970) than any methods manual.
Participant observation is potentially revolutionary because it forces one to question one's theoretical presuppositions about the world by an intimate long-term engagement with, and participation in, the lives of strangers. It makes us recognize that our theoretical conceptions of the world come from a particular historical, social, and spatial location. There are of course other disciplines that can help us for similar ends-for instance the study of history. I am so often struck by the explicit similarities in participant observation as potentially revolutionary praxis and the attitude to the relation between theory and practice by those that have focused on the "small voices of history, " the "history from below, " as e. P. Thompson (1971) or Ranajit Guha ([1983Guha ([ ] 1999 have. but it is participant observation that consistently forces us as anthropologists to always question our ideas of the world by engaging with those of others; revisiting and revising the questions that we enter the field with, often making our initial ideas redundant.
In my own case, for example, I applied for a PhD in anthropology to study a UNICeF program for the Development of Women and Children in Rural Areas, but preliminary fieldwork revealed it only existed on paper. In the first year of my PhD, I changed my proposal to comparing the poverty alleviation work of a government and an NGO in Jharkhand, eastern India, but quickly realized as I began fieldwork that the Adivasi people of the region were far more interesting than the institutions that tried to work among them. I could never have imagined when I began that I would end up following Adivasi migrant labor to brick factories in Calcutta, chasing wild elephants' night after night with the Adivasis, trying to understand why they sought to keep the state away from their lives while the local elites moralized about their siphoning off of public funds for their private pockets, or that I would help my village friends escape from the pressure of having to comply with Maoist revolutionaries. I could not, in my wildest dreams, have guessed that I would write a thesis on understandings of the state that would turn into a book on indigenous politics and that many years later I would return to live with the Maoist guerrillas themselves in a different part of the country. Participant observation makes us consider people's lives and social relations in their totality, makes us always be open to considering new possibilities in an open-ended study, and forces us to constantly revisit our theories and assumptions. It is this requestioning with which we often seduce potential students to anthropology when we say that the beauty of subject is that it makes peculiar what we once thought was normal and shows us that what is strange might in fact be the basis of new normative analysis of the world, that is new theory.
What is this theory? Another issue on which there seems to be much confusion. Theory is a set of general propositions about the world that cannot possibly belong to any department or discipline. Participant observation enables us to contribute to theory no matter under what rubric or in which discipline. So the debates about whether something is called anthropological theory, ethnographic theory, or sociological theory seem to me to be superfluous. The point is that the knowledge we produce-while inevitably emerging from a history of ideas in particular fields and institutional frameworks-is and should be beyond the confines of any disciplinary boundary. Anthropologists practicing participant observation may contribute with theoretical propositions to theories of cognition just as they may to theories of the economy, and in all these fields our propositions may lead to alternative ideas and action to that which prevail. Our theoretical propositions need not be new. We are always building on the work of others and we may be reviving old ideas long buried. Whatever we propose though should always at its heart be interdisciplinarycross-disciplinary is in fact a more ambitious way to put it-for theory (or action) can never be the preserve of any discipline.
Moreover, in some parts of the world where anthropology departments are taken over by other traditions or there are no anthropology departments, the praxis of participant observation may only take place in other disciplines. In the Indian case, for instance, anthropology departments tend to be populated by those describing and measuring the world (sometimes even through cranial indices) and it is sociology that is more likely to engage in participant observation. (Though it is in fact rare in the Indian academy to find the doctoral scholar who truly embarks on participant observation and perhaps it is more alive there outside of the institutional structures of academia). In the United Kingdom or in the United States, anthropology then is the name given to the institutional structures-whether it is departments, journals, or seminars-that keep alive the praxis of participant observation and its traditions. It is important that we teach what has theoretically developed from participant observation and its history as the subject "anthropology, " that we critically analyze each other's fieldwork and ideas in anthropology seminars to refine them, interrogate our propositions from within. but these-our activities in the discipline of anthropology as it exists in the universities-are just the starting point, the home from which we reach out beyond our institutional confines (a point I will return to).
Our project is ambitious in a world where theory is controlled by people sitting in the offices of other disciplines who are usually more powerful than we are. It is easy to give up and talk only to ourselves, produce simply description, or conjure up some "ethnographic theory" that may mean something (if anything) only to a small number of people within the discipline, or engage in interdisciplinary work to merely produce case studies for other disciplines to theorize. but we must keep up our spirit, challenge ourselves, not get too comfortable, and at least try to realize the full potential of participant observation.

What is participant observation?
The foundations on which we build participant observation as revolutionary praxis critically entail attention to history and the possibility of its transformation through human action. It must have at its heart the idea that we and the people we live and work with come together in a particular conjuncture of time and space, are inheritors to specific ideological and material histories, are working with and against their grain. Attention to and awareness of this history, the structures of ideology and material forces within which our fieldwork is placed, is crucial to our ability to engage in participant observation that produces new knowledge and has the potential to transform ideology and action, challenge dominant theory, propose alternative theories, and take action that may potentially challenge the forces around us, change the course of history.
To dispel a few misconceived notions, I would like to go over some well-rehearsed ground about what participant observation ought to mean. Some readers may find some of the ground covered one over which much ink has been shed. Nevertheless, I feel it is important to reiterate not only in relation to our friends in other disciplines who valorize ethnography but usually do something far less radical than what we hope to embark on, but also because even in our own discipline, in this era of quick results and impact, we can usefully reignite its spark.
Participant observation centers a long-term intimate engagement with a group of people that were once strangers to us in order to know and experience the world through their perspectives and actions in as holistic a way as possible. For short, I will refer to these four core aspects that are the basis of participation observation as long duration (long-term engagement), revealing social relations of a group of people (understanding a group of people and their social processes), holism (studying all aspects of social life, marking its fundamental democracy), and the dialectical relationship between intimacy and estrangement (befriending strangers).
Let us turn first to the relationship between intimacy and estrangement. Why is it important to work with people who you feel at first sufficiently alienated from? It is nothing to do with exoticism or cultural relativism (the stereotypes that are sometimes lazily attributed to anthropologists) but it marks the very basis of our ability to contribute new insights. Working with people who are similar to us or have the same histories as we do risks us perpetuating what most other disciplines do-that is, to work from theoretical premises that ultimately only demonstrate their own assumptions.
Working with strangers does not mean we all have to take the next boat to the last tribe that Survival International has found. Far from it. It could mean working on our own doorstep with people who are at first new to-and different to-us. It could, in some rare cases, also mean working with the communities that we come from if we have been sufficiently estranged from them over the course of our lives. For what is important, as Kirsten Hastrup (2004: 468) once said, "is a deliberate alienation from the world under study in order to understand it as it cannot understand itself. " Of course, the point is that over time we will be profoundly intimate with the people we study, sometimes become kin, certainly no longer strangers, and will be able to maintain that productive-but-difficult tension between involvement and detachment as friends and scholars.
Duration, like estrangement, is important because it takes a very long time to become a part of other people's lives, to learn to speak, think, see, feel, and act like them (Malinowski's recommendations in the Argonauts). Participant observation, it is typically suggested, should in new communities be conducted over at least a year (that is if there is already some grasp of the language), ideally eighteen months or more (though some anthropologists have life-long engagements with the people they work with), preferably living with the people one is studying or in very close vicinity. This commitment of time to those one is studying is important to get to know people intimately, to see and understand the conflicts and contradictions between them, and most importantly to challenge our own ideas and assumptions. The insights of participant observation are based not only on what is said but also that which is left unsaid and demonstrated only through action. Foundational to participant observation is the recognition that knowledge itself is practical and that theoretical or abstract knowledge-that which is communicated in languageis a very particular kind of knowledge that must be situated in relation to practice. Participant observation thus enables us to explore the disjuncture between what people say and what they do.
To access and understand as full a reach of knowledge production as possible, and the contradictions that lie within, one needs to be very intimately embedded in the communities one is working with, be involved and observe them through different seasons and stages of a lifecycle. This is why we guide our PhD students to spend at least eighteen months in the field with the communities they are studying, that indeed six months into field research they may realize that they need to discard much of what they thought was the case in the previous six months. It takes time to question and undo what one thinks one knows and to acquire new knowledge not only by speaking with others but through becoming-in practice and in part-others.
Then there is holism-that is the significance of understanding the total social context of the people we are trying to understand. Participant observation is an inherently democratic form of knowledge production because it takes seriously the lives of ordinary people in a holistic way. Holism is important because it recognizes that we cannot understand one aspect of social life in isolation from another. We do not need to define ourselves as a political anthropologist or an economic anthropologist or an anthropologist of ethics, as though these labels are a badge to differentiate ourselves from other anthropologists, for the wonders of our holism means that the most insightful work we do is that which does not limit our scholarly boundaries. To quote Jonathan Parry reflecting on Malinowski, "it was not possible to understand chieftainship in the Trobriand Islands without understanding its links with international trade in the form of the kula" (2012: 44). We cannot separate a domain of study that is politics from one that is economics, any more than we can separate religion from kinship, because it is the interrelations between them that are the most significant.
Finally, we have the revelation of social relations of a group of people. The potential theoretical contributions of participant observation are dependent on knowing intimately a group of people-defined in various kinds of ways-and those that affect them and are affected by them. This inevitably leads one to understanding people, places, and forces beyond the group of people one is living with, at different scales of analysis. There is often some misunderstanding around this focus on revealing the social relations of a group of people in relation to "sites" and "scales. " In the quest for depth of understanding, focusing on the situation of a group of people does not mean confining oneself to one site or one village or one scale. In fact it demands that we follow processes in and out of our field sites, follow the impact and bearing they have on the people we are concerned about; and this may mean following processes (e.g., spread of neoliberal financialization) that are often beyond person-to-person relations, are often hidden.
I have often heard students remark in somewhat confused terms, "My study is not just a village ethnography; it is multisited. " What they seem to misrecognize is that there is really no such thing as the kind of bounded village ethnography they imagine, if there ever was one, and that studying a group of people and all the different social relations that have a bearing on them inevitably leads one, through the people and processes, to different physical locations, different sites, exploring different scales. To cite Parry again in relation to his first fieldwork based in a Himalayan village, "In order to investigate the hypergamous marriage system of the Kangra Rajputs I found that, avant la lettre, I needed to do some 'multi-sited' fieldwork at different points on the east-west chain along which brides are given, and in order to understand how it had changed over time I needed to examine the policies that the british army applied to recruitment and promotions in its Dogra regiments" (2012: 44). Inevitably, to understand the social relations that the people we study are embedded in, we must work across time and space. Perhaps even more significantly, the danger is that research that seeks to be multisited without recognizing the importance of a focus on the social relations of a group of people, ends up studying groups of different people in different places who don't have much of a bearing on each other, thus producing thin bits of evidence that are not only incommensurable but also do not challenge the theoretically presuppositions that we began with.

The risks of participant observation
Participant observation of the kind I describe-that centers a dialectical relationship between intimacy and estrangement, long duration, holism, and revealing the social relations of a group of people-makes the fieldwork we do perhaps more difficult than that of any social science.
It requires us to dive deep into the sea of other people's lives and find a way to swim with them. It requires commitment, endurance, constant improvisation, humility, sociality, and the ability to give oneself up to and for others. It also entails the ability to retrieve oneself and be prepared to rethink, from this position, everything one thinks one knows. And then it needs one to swim back to the shore and be prepared that this shore is almost always going to be different from the shore where one began.
This swimming in the sea is risky. It is not just that participant observation is practically risky but more profoundly the risks are at least four-fold: The first is that we may be unable to leave our shoreline because we fear the water we need to dive into. So though we may spend a long time in our field site, we remain on the peripheries of what is going on there. essentially, we never leave our own lives to delve into those of others and are then not able to challenge ourselves and our ideas about the world in any significant way, running the risk that we contribute only in terms that have been preconceived.
In today's era, we do not have the forced emotional and physical separation that existed in previous times to help us. In most of our field sites, there is the easy possibility of a constant emotional intrusion via technology-mobile phones, emails-from our shoreline that can be limiting to our learning, to the production of democratic pedagogy. This intrusion need not of course be prohibitive but it means that we must work ever harder at alienating ourselves from our worlds to dive into those of others.
The second risk is that we may sink into the sea of other people's lives, unable to cope with the challenges of trying to swim back to the shore. While physically it may result in "going and staying native" altogether, there is also another risk-a theoretical one akin to that which David Graeber (2015) has highlighted in his critique of the ontological turn in anthropology. We may remain swimming with the people we are studying, refusing to or unable to come back to the shore, thinking that it is enough to simply present their constructive imagination as "radical alterity, " without challenging ourselves to think about what implications our experiences with them have for the general questions of what it means to be human and to have social relations anywhere in the world. There is of course nothing wrong with presenting cultural critique as radical alterity. but it is politically and theoretically limited and may result in us simply producing a collection of pretty butterflies, for other people to collate, theorize, and act on; rendering what we do as not much more than laboratory specimens for other disciplines and their theoretical suppositions. Then the risk is that we are indeed drowned out.
The third risk is that there is potentially no end to our enquiry. While our openended quest through participant observation and search for holism is the beauty of what we do, it is also important to acknowledge there is a delicate tension between reaching our conclusions and getting an adequate purchase on the whole. In the end, we do have to make compromises and judgments about what to include and where to stop (see Parry 2012).
The fourth risk is that there is a tension between participation and observation. Most of us most of the time are doing just the observing and not the participating. It is the participation, in particular, that has the potential to reveal unique new insights as Parry shows of his bhilai work when he got involved in the arrest of his research assistant, Ajay T. G. who was jailed for allegedly being a Maoist revolutionary. but once we become truly participants it makes us lay our stakes on the ground and potentially alienates related others and prevents us from observing them (in Parry's [2015] case, once he was seen as a part and parcel of Ajay's arrest-the local press reported that Ajay had brought a professor from the London School of economics to help in the recruitment of the Maoists and to fund the movement-the aristocracy of labor with whom they had worked shunned them).
While participant observation is foundational, so as not to be misunderstood, it is worth noting a few further points in relation to its limits.
The first is in relation to fieldwork. That is, while participant observation is crucial, our fieldwork must involve a range of methods that will depend on the fieldwork context and the issues we become compelled to contribute to theoretically and politically. So, for instance, household surveys are crucial not only to have basic information about the people we are working with but also to show general patterns and configurations emerging from our analysis. We are only shooting ourselves in the foot by not being rigorous enough to capture statistically some of the more general patterns we propose. More generally, collecting data and empirical evidence is important if we wish our work to be convincing to others and have an impact in debates with those working in different disciplines.
Similarly, archival research into the places and people we are working with is important to deepen our understanding of the relationship between the present and the past, to better understand the present. Learning the political or ritual language of the groups we are interested in, apart from their everyday language and practice, is also significant. In certain cases simulated environments can help us get deeper into particular issues whether it is the kind of experiments described by Astuti or getting a group of people together to have a focused discussion or debate on a particular issue (what other disciplines often call "focus groups") or even formal interviews (in contexts where this may be expected of a researcher). Following flows of processes-for instance, the movement of capital or finance-as it impacts on the people we are living with but are beyond their immediate environment may be important. These are all different methodological tools we may also use in our fieldwork.
A second point is the place of participant observation in relation to the anthropologist's lifecycle. It is rare to find the anthropologist who immerses herself totally in a new setting twice in their lifetime-three times is almost unheard of today but was perhaps not so unusual in an earlier era. There are many reasons for this. Participant observation is so emotionally and practically challenging that it is not only hard to keep putting yourself through it again and again but also difficult to inflict it repeatedly on one's family and friends. In today's era, it is also extremely difficult to get away from our institutional contexts-the responsibilities we have as Heads of Departments, Conveners, Chairs of Committees-that tie us to our bureaucracies. It is more common for the anthropologist to return to the same site. Astuti is correct to suggest that the point, in these return encounters, is to continue to challenge our theoretical suppositions and ourselves.
It is perhaps even more common to find anthropologists who do not return to their field site, engage in much thinner fieldwork elsewhere, or make speculations about the world based on the work of others. Howell says we have no future as such "armchair anthropologists. " This may perhaps be the case if that is all we were to do as a community, but there is clearly also a value in anthropologists who have once experienced the revolutionary praxis of participant observation, casting their net out to try to read against the grain of the knowledge produced by others from a different era or from different spaces, consolidating the work of others. For participant observation crucially teaches us how to read the work of others-to be able to assess the validity of the claims that are being made, the evidence on which they are forged, and to read against their grain. Perhaps it is no surprise that we so often say, when reflecting back on an anthropologists lifeworks, that it was the first book that was the best! A final point is in relation to what we do with participant observation, for there is so much to do. In the early part of our lives as anthropologists there is a certain degree of policing-a compulsion self-imposed and increasingly institutionally imposed-to prove that we have passed the test, the rite de passage. We write ethnographic texts and we submit articles to anthropology journals mainly for other anthropologists to read. There is nothing wrong with this and indeed in this neoliberal university environment where everything needs to be utilitarian, measured according to what impact it can have on business or policy, it seems to be more important than ever to have the production of knowledge just for the sake of it. but we can and should not hesitate to also reach beyond the university. We may do this in different ways and through different mediums: whether it is through the engagement we have with people in other disciplines, publishing in their journals; whether it is through reports we may write for those who we want to influence with our ideas; whether it is reaching a broader public through radio, film, or other forms of reportage; or even in the ways in which we write our monographs so that they aren't just the preserve of other anthropologists. And that is that we are inheritors, harbors, and protectors of a unique form of knowledge production that is based on the dialectical relationship between method and theory, which is praxis.

Participant observation is only just the beginning
There is no other form of knowledge production-as found in the praxis of participant observation-that is so deeply democratic through its very premises, that requires, even forces, one to throw away one's assumptions about the world and seeks to understand social life anew through our engagement with distant others and their social relations. In most cases, this means taking seriously all those people and their histories that would be easily and willingly ignored by others. There is also no other form of knowledge production that is so democratic that it demands us to dwell into all aspects of people's lives, making us focus on the whole rather than just a part.
Participant observation enables us to literally turn things on their head. It enables us to challenge received wisdoms and produce knowledge that previously had no space in the world, was confined to its margins, was silenced. engaging in participant observation is thus a profoundly political act, one that can enable us to challenge hegemonic conceptions of the world, challenge authority, and better act in the world. but a small, parting warning on the tensions between our scholarship and activism, on acting too quickly, too soon: With the crisis of the left and the spread of the right across the world-whether it is Trump in the United States, May in the United Kingdom, or Modi in India-there is a compelling need for participant observation as revolutionary praxis, to offer better theory and action in the world. Indeed, we have seen the surge of a call from within for a militant anthropology (Scheper-Hughes 1995) and it is tempting to join hands with various forms of activists. but there is a real tension between the democratic commitment to the truth in a holistic sense demanded by participant observation and the commitments of partisanship expected of the activist (see Shah 2010). In the short run, we may need to suspend our moral desire to become a part of those activists whose political engagements we wish to serve-to pursue any naïve kind of militant anthropology-and to recognize that participant observation may force us to reconsider the theoretical premises of even those we morally feel we should explicitly form alliances with. Participant observation may in fact inhibit our revolutionary zeal. That is the point and democratic potential of participant observation in politically engaging in the world. In the long run this makes us better politically engaged anthropologiststhat is, better activists.
I would like to end by reflecting on how my own thoughts on participant observation as a potentially revolutionary praxis have emerged and acknowledge the debt I owe to Maoist-inspired Naxalite revolutionary guerrillas in the tribal dominated forests and hills of eastern India, the people who became part of my own fieldwork in the last decade. I was amazed to find these guerrillas-fighting what is now a fifty-year-old revolutionary struggle to transform the world into a classless egalitarian global society-living amid the people who they were working with, almost as anthropologists. In fact, they appeared to follow deeply anthropological methods. They endeavored to adhere to Mao Zedong's (1961) advice, "the guerrilla must move amongst the people as a fish swims in the sea"-that is, they immersed themselves so closely amid people (the water) that they were able to move through the water (people) like fish. Although they didn't call what they did participant observation (they talked about it as the mass line, "from the masses to the masses"), some of them in fact took participant observation so seriously that they seemed to me better anthropologists than many that I had come across.
Yet the moral commitments of their activism stopped them from acknowledging some of the contradictions and realities they were faced with, even as it arose from their own revolutionary praxis of "from the masses to the masses. " It is my experiences with these Maoist guerrillas and my emerging critique of them that urged me to recognize that, in fact, what our forefathers and mothers since Malinowski have given us in participant observation is a profoundly important possibility of political engagement, whether we recognize it or not, for it is a potentially revolutionary praxis.
The very idea of participant observation as a potentially revolutionary praxis is for me then itself influenced at least in part by my fieldwork among these Indian Maoist revolutionaries. The theory that participant observation can be a revolutionary praxis is itself a product of participant observation as revolutionary praxis. théoriques à propos du monde qui nous entoure, et à produire un savoir nouveau sur ce qui était mis en marge, sur ce qui était tu. Il est expliqué ici que l' observation participante n' est pas seulement une méthode de l'anthropologie, mais une forme de production de savoir à travers l' expérience de la vie et de l'action. Cette expérience est praxis, le processus par lequel la théorie est dialectiquement produite et réalisée dans l'action. Quatre aspects de l' observation participante sont ici évoqués: sa longue durée (un engagement de long terme), la mise en évidence des relations au sein d'un groupe (comprendre un groupe d'individus et leurs processus sociaux), le holisme (étudier tous les aspects de la vie sociale, marquant la qualité démocratique de cette méthode) et la relation dialectique entre intimité et mise à distance (se lier d'amitié avec des étrangers). Même si les risques et les limites de l' observation participante sont soulignés, de même que les tensions entre activisme et anthropologie, nous suggérons que l' observation participante est un acte profondément politique, un acte qui nous donne les moyens de contredire les conceptions hégémoniques du monde, de contester l'autorité et de mieux agir dans le monde. Alpa Shah is Associate Professor (Reader) in the Department of Anthropology, London School of economics and Political Science. She is the author of In the shadows of the state: Indigenous politics, environmentalism and insurgency in Jharkhand, India (Duke University Press, 2010) and has led the research and writing of Ground down by growth: Tribe, caste, class and inequality across India (Pluto Press, 2018). She has edited seven volumes and written numerous articles on issues ranging from affirmative action, agrarian change, revolution in India and Nepal, emancipatory politics, the underbelly of the Indian boom, and Adivasi and Dalit political pathways, and is finalizing a book manuscript on India's Maoist inspired Naxalite revolutionary struggle.

Alpa Shah Department of Anthropology London School of Economics and Political Science
Houghton Street London WC2A 2AE UK a.m.shah@lse.ac.uk