Constantine Nakassis on his book, Onscreen/Offscreen

Interview by Chris Ball

Christopher Ball: Onscreen/Offscreen tracks Tamil film of the first decade and a half of the 2000s through analysis of different feature films, including interviews with actors, directors, and audience members, and ethnography of filmmaking and reception. This approach shows the intertextuality that ties together Tamil cinema, in spite of its own internal divisions between “class” and “mass,” “film” and “cinema.” At the same time, Chapters 1 to 4 deal, in turn, with the narrative figures “the hero,” “the heroine,” “the fool/jester,” and lastly “the auteur/director.” Can you say something about how as an author you discerned the presentation of your analysis and your interaction with, and presentation of, these characters?

Constantine V. Nakassis: Thanks for this question, Chris and for engaging with the book! This book is based on my work on the Tamil cinema of South India dating back to my very first fieldwork as a graduate student almost twenty years ago. But I didn’t want to write an ethnography of the industry per se. Rather, I had certain theoretical issues – for example, about the ontology of the image, its performativity, about spectatorship, the production format of images, realism – that I wanted to explore and elaborate on in each chapter of the book via the ethnographic analysis of a particular film or set of films. I also think having worked on Tamil cinema for an extended time gave me a longitudinal perspective that was invaluable in shaping how the book developed and how I came to think about its internal organization. From my first work in the early and mid 2000s to my dissertation work in the late 2000s to post-PhD work in the 2010s, I repeatedly encountered claims and desires for a “new” realist (“class”) cinema – and the announcement that it had (almost!) arrived – coupled with a lament about the untimely performativity of fantastical images of the industry’s larger-than-life “mass” heroes; as I came to realize, this ambivalence was a constitutive feature of the political field of images, despite whatever real changes there were across this time period (and before). Why and to what effects?

This question helped frame the book, specifically splitting it into two parts: the first (“Presence/Representation”) primarily concerning the presence of images and the second (“Representation/Presence”) their representationality. Each of those two parts then split into two chapters: Chapter 1 focusing on the performative presence of images of so-called mass heroes and Chapter 2 of heroines and dancers (whose presence manifests as a kind of negative, sexualizing stigma), and Chapters 3 and 4 focusing on attempts in the industry to create representationalist images, either in the mode of parody (of heroes) or realist, director-anchored films. But if Part I focuses on the insistent presence of the image, and how this traverses onscreen and offscreen (the image spilling beyond the screen, but also the social getting introjected into the film text), each of the cases also feature a countermovement, where such images get framed representationally by their directors, causing an interesting kind of friction internal to the image. Part II, by contrast, explores how filmmakers and other stakeholders to the image attempt to bracket and denude the image’s performativity, often by putting the hero under (narrative) erasure. But even in doing so, a certain kind of performativity bubbles through the citational frame. How is that managed? And to what ends?

While I didn’t know that the book would take this form when I was doing my fieldwork, I did know that there were certain incidents, or films, or industry dynamics that came up in my research that I wanted to understand and unpack: like in what became Chapter 1, where a slap in a film scene caused an uproar among fans because it involved a junior actor playing a character who slaps another character played by a huge hero-star, Ajith Kumar; or, in Chapter 3, the first full-on spoof in the industry, Thamizh Padam (‘Tamil Film’) or, in Chapter 4, the outcrop of small-budget “realist” films in the 2000s; or, in Chapter 2, the fact that most heroine-actresses in postmillennial Tamil cinema are not considered ethnolinguistically Tamil. As I worked through these various incidents and films and dynamics, I saw how they spoke to certain theoretical issues that I kept circling around; in particular, this issue of representationality and performativity. 

This had another effect, which is that certain characters – the hero, heroine, the comedian, the director – came to the fore in each chapter. This wasn’t by design, but it had to do with the fact that each of these characters is a critical node on which semiotic dynamics/achievements like representationality/performativity are themselves reflexively centered in the Tamil cinema: the hero and heroine as sites of (positive and negative) plenitude and performative presence, the comedian as piercing that plenitude with parodic self-reflexive humor, and the director as an alternate site of authority over filmic and cinematic semiosis. Here, too, emerged something else that became a central preoccupation of the book, the question of who or what is responsible for the image (a Goffmanian question of the politics of the production format). In certain moments we see the actors – those animating heroes and heroines and comedians – held responsible for the images in which they appear (as presencing them in some way); but at other moments, it is the opposite, where the director holds the burden for the image (as representing their vision or agenda or whatever). So how and when does one or the other (or both or neither) happen? And what follows from this? 

Chris Ball: As you were just saying, the play of representation and performativity is a central theme of Onscreen/Offscreen. In attempting to bring together linguistic anthropology and film studies, you wade into territory that is often characterized by a rejection of “linguistic” or “representationalist” accounts of filmic or other image texts. How did the framing of realist representation and performative presence as distinct but connected, indeed complexing mutually implicating, allow you to reframe such divisions or disciplinary impasses?

Constantine V. Nakassis: Yeah, it’s true that American cinema and media studies, but also visual and media anthropology, often pit “language” (or “representation”) and “image” against each other; and along this axis of difference (as Susan Gal and Judy Irvine put it) a whole host of other contrasts cluster: logic/affect, mediated/immediate, aural/visual, sensuous/abstract, and so on and so forth. In film studies, I think that this had to do with the limits and failures of structuralism, which invested so deeply in the notion of a “film language.” There is something so ironic in this, of course, since what such structuralisms-beyond-language did was to take what is unique about language and then ask how it could be used to illuminate other semiotic systems. But the whole thing was bound to fail, definitionally! In film studies, scholars like Christian Metz realized this contradiction at the very outset, but they gave it a try anyway and working through the failures pushed into new territory (for example, by coming to think with Benveniste on enunciation) that interestingly resonated with what was going on in linguistic anthropology (not that anyone realized this at the time, since they weren’t reading each other’s work).

In any case, and by contrast, in the US, this failure became a reason to throw the semiotic baby out with the semiological bathwater. But for me, the very division of image versus language is misguided from the outset; it’s a semiotic ideology, as Webb Keane calls it. The more interesting move, it seems to me, is to treat semiotic ideologies (like this one) as ethnographic components of how people experience semiosis, and thus how semiosis itself unfolds. If seen in this way, then we can start to ask, well, how and under what conditions does a film image represent and seem to stand apart from what it represents? And when does it performatively presence and act? When does the image seem to reach through the screen into the events of its happening and presence itself? And when is that possibility blocked, and the image and screen are kept “within” themselves via some proverbial fourth wall? Tamil cinema is rife with these performative perforations; but it’s also filled with realist representations too! And my point in the book is that an image isn’t necessarily one or the other but may be one or the other depending on the point of view, or maybe both at one and the same time, or at different times, and so on and so forth. And here is where things get interesting for me. Because what I found in the cases I looked at is that the image harbors both potentialities and tendencies, each brimming out of the screen, each in dialogue with the other, sometimes amplifying, sometimes dampening each other. So, there is a tussle and a multiplicity – a dialogism, indeed, a politics (and in the Tamil case, where actors often form populist political parties, a real electoral politics) – internal to the image. Yet when we try to study this, we realize that whatever is “inside” the image is the effect of semiotic processes (of entextualization and contextualization, in our technical jargon) that take place in a sort of constitutive “outside” to the image.

And this raised a further question for me: what kind of a thing is an image such that this is the case? And if this “is” – a question of “ontology” – is political, which it is (claims about “what is”are THE site of the political, aren’t they?!), then these semiotic questions take us directly into what I discuss in the book as an “ontological politics of the image.” Hence, to return to your question (sorry for the long tangent!), the book tries to provide a way thinking about cinema that allows us to get beyond the stale oppositions of image versus language, or of “film language.” Hopefully, this will resonate with folks in film and visual studies, and put them in touch with linguistic anthropology, and vice versa. It always struck me as a shame that there hasn’t been any meaningful traffic between these fields (since, at least, Sol Worth and Christian Metz), especially since both deal with the most complex of problems of meaning and each has developed incredibly sophisticated semiotic approaches to them. The book is trying to create that dialogue.

Chris Ball: I’d like to ask more about your concept of the “ontological politics of the image” and its potential links to debates in anthropology at large. You introduce the concept in the Introduction by questioning the givenness of the image as object, and by suggesting a pragmatic approach to the multiplicity of the image’s construction in varying sociohistorical and cultural contexts. The cross-cultural stability of the image, if we can hold on to something concrete, lies in its existence as a “political fact,” such that for filmmakers, critics, and viewers alike, in Chicago or Madurai, “Contestations about what an image is, is the very nature of images.” 

Constantine V. Nakassis: Right, exactly, and this connects up with what we were just talking about, this question of whether and when an image is a performative act or a realist representation; these, of course, are just two very simplified ways of talking about the heterogenous being of images; as we were saying, they are dimensions or potentialities or tendencies that get actualized in particular contexts in particular ways. And if such actualization is up for grabs, as I found it to be, then we need to be attending to the “ontological politics” (this being Annemarie Mol’s phrase) that condition the being and becoming of images. This, it seems to me, is the nature of images as we find them: stakeholders to images (makers, audiences, various institutions, including various organs the state; in India, e.g., the Censor Board) contest and shape what images come to be, what they should (or should not) be, and so on. So, there is a kind of semiotic relativity at the heart of the question of being. And this itself requires that we take a particular ontological position, a semiotic realism (as you yourself have argued in other contexts) that insists on the reality of generals – in this case, semiotic-cum-social facts – but sees them as (the outcomes of) dynamic processes. 

Chris Ball: So how does your approach to the contested construction of images and the possible multiplicity of their ontological status connect to wider trends in anthropological theory? I’m thinking about the Ontological Turn, obviously, but also about the general (post-)post-structuralist, post-colonial, post-modern, post-human, and so on, engagements with multiplicity, fractures, and frictions between worlds? What is your stake with this book in such conversations?

CVN: In this book, much more so than those trends in anthropological theory (e.g., the Ontological Turn), my discussion is in a direct conversation with the foundational film critic and theorist André Bazin and his “ontology of the photographic image.” This has and continues to be a central issue in film studies. So, my aim, in addressing the question of what an image is, is to reformulate the question of being through a pragmatist approach to semiosis, but also with attention to the open horizons of possibility, and relativity, as I was saying before, that ethnographic theory at its best can provide. To do so, then, I am thinking with, but also against, Bazin to ask, how can we reconceptualize the so-called “ontology of the photographic image” processually and ethnographically, from a disciplinary perch (linguistic anthropology) and empirical archive (commercial Tamil cinema) that has been very peripheral to these discussions? 

Of course, the so-called Ontological Turn, among other movements in anthropology, has also tried to decenter certain philosophical positions that also animate the (Bazinian) questions in film studies (i.e., there are common touchstones). And insofar as this is the case, there is a resonance there. (And, certainly, my uptake of Mol’s phrase also connects this work to actor network theory and Latour, if indirectly.) But there are also major differences between what I am doing and the Ontological Turn. For example, there is a kind of relativism in the Ontological Turn that I find problematic. And, as you’ve pointed out in a recent paper at the 2022 AAAs, (semiotic) relativity is not the same as (cultural/ontological) relativism. In fact, a comparative account of the relativity of languages or images (or physical processes, for that matter) demonstrates that there are general properties that undercut the strong relativism in certain versions of the Ontological Turn. Indeed, the fact of the multiple ontic tendencies within images that we were discussing a moment ago implies a set of processes that traverse “worlds” or what not; as I argue in the book, it implies a politics of being that produces that very ontic multiplicity.

It’s worth underlining that an ontology is an account or theory (-logy) of being (ōn, ont-); it is a claim on being. And to say an account – or theory or claim – is to imply a plurality (hence the political aspect!): ontologies and modes of being (in a dialectical relation, of course). But it’s a mistake, in my opinion, to think that for every “world” (or every “culture” or “epoch”) there an ontology (or “worldview”), or that for every medium there is a different mode of being, like analogue versus digital images, as some people have suggested in visual studies. And again, this is because the multiplicity of being is intrinsic or internal, and yet also distributed, extimate. It’s not a pluralism, and that’s because being is a dynamic political fact. 

Another implication of this is that while this book is about the Tamil cinema of South India, with a real serious consideration of its particularities, ultimately through theorizing those particularities the story is about much broader features, relations, and processes of images (and semiosis more generally), whether in film or otherwise, whether in Tamil Nadu or elsewhere. 

Chris Ball: You talk about realism throughout the book but approach it most directly in Chapter 4. You end up mostly agreeing with Bazin on cinematic realism, though you qualify that it is not because of an indexical trace of light between reality and (celluloid) film, but rather because the connection to the real is political. Realism is an achievement, an effect rather than a cause in this formulation. I’m struck by the similarity of this move to that which linguistic anthropology has been articulating with respect to language for a long time. Rather than starting with the assumption of language as an ontological object that is given, analysts have turned increasingly to articulate how multimodal communicative practices and their politics, together with their (metapragmatic) frames of interpretation, effectively yield familiar domains we label as grammar, or discourse, or language writ large. Does that analogy make sense, and did you think about that sort of cross modal iconicity, either taking a set of analytic tools from one domain such as linguistic anthropology to another like film studies, or claiming that language and image are after all similar, in writing the book?

Constantine V. Nakassis: Yes, it definitely makes sense! I am a linguistic anthropologist and Onscreen/Offscreen is a book of linguistic anthropology. In the book I call this a “linguistic anthropology of cinema,” and some amount of discussion is devoted to explaining what that phrase means. So, in one sense, I am treating cinema in ways similar to how linguistic anthropologists have approached “language,” and precisely how you characterize it. Part of what that means is transferring advances and tools made in one domain (“language”/discourse) to another (film/cinema). 

But I’m not sure if we should see this as an analogy, as if cinema was like language (or discourse), and thus one can extend linguistic anthropology outwards, as if its center was always and only language. Part of my discomfort with this way of stating it is that I don’t see my work as an application of linguistic anthropological tools to something else that is not proper to the field, as much as I see this as linguistic anthropology, period. And part of the reason I insist on this is because to not so insist (and to take the book as an application) would reinscribe a certain center of linguistic anthropology as the study of “language” “in” …, and then you can fill in the blank with all those substantives that we are familiar with (“culture,” “society,” “interaction,” “use/practice,” or whatever). Doing this recapitulates a picture of language—a formalist, referentialist, structuralist vision of language as autonomous, sui generis “system”—that the very move of adding the “in culture/society/interaction” is meant to deconstruct! So, part of what I try to show in Onscreen/Offscreen, and in other work, is why this is ultimately an unproductive gesture. Of course, it’s not untrue historically, since much of the field has been concerned with language in culture, society, and so on. But, as I’ve tried to argue, even when this seems to be the case, what we are studying in such cases is not language as such, but rather the limits of the very construct of language to capture semiosis in (and across) its contexts of happening; and further, that by pushing through language we open up all sorts of interesting horizons beyond where we started. And those horizons are what ground us (even if we always start from where we came). Hence my claim that Onscreen/Offscreen is a book of linguistic anthropology even though – and maybe especially because – what I am focused on is cinema (and not language, or speech, in films). 

But interestingly, what this also means is that this is a book that is not only about cinema. Rather, as you point out, and as I was saying earlier, it is trying to get at certain semiotic dynamics that are not medium-specific. I have come to realize that thinking in terms of medium specificity is a cul-de-sac. It is, at best, self-limiting and, at worst, reductive and essentialist. It may be productive for disciplinary political projects (to zone off some domain of study, say, language for linguistics, film for film studies, and so on and so forth), but intellectually it narrows thought. So I am absolutely not interested in purifying the (film) image and trying to define its uniqueness, or definitional specificity. That would just reproduce the very problematic, blinkered mode of inquiry that linguistic anthropology, in its critique of that very form of thinking as concerns language (focusing on what makes it peculiar and unique: its essence as a system for denotation), has rendered unstable and untenable. So, in moving outwards from the yoke of language to images, my aim is not to suggest that we rest there and call it a day. Rather, I would endorse Peirce’s metaphor for scientific inquiry as walking through a bog. We must plant our feet somewhere, but only to always have to pick up and move, for if we stand content too long in one place, we sink. We have to move forward and explore, in this case, looking for what traverses what has been categorized as film and language. Now, I don’t mean to collapse each to the other, or to deny that there may be specificities; rather, my aim is to open up a clearing for thought in ways that propel us forward and make connections with other fields. 

For me, this general attitude toward inquiry implies an open vision for the field, for a linguistic anthropology of …, where that ellipsis is to be filled with as many possible possessors of the genitive that we can find, from cinema to spam filters to painting to ethics to cryptocurrency to … . It can all be linguistic anthropology. And it is: our wonderful colleagues are already pursuing all these things and more! So, that ellipsis is an invitation for us to open up and for others to find their way in. From my point of view, we should be pitching a big tent, not policing the borders or redoubling on the centers. This book is an effort to that end. 

Onscreen/Offscreen is open access and available here.

Barney Bate’s book, Protestant Textuality and the Tamil Modern

Cover of Protestant Textuality and the Tamil Modern by Bernard Bate, Edited by E. Annamalai, Francis Cody, Malarvizhi Jayanth, and Constantine V. Nakassis

Hannah Carlan interviews E. Annamalai, Francis Cody, and Constantine V. Nakassis

Hannah Carlan: Bernard Bate’s posthumous book Protestant Textuality and the Tamil Modern is a major contribution to the study of political agency and nation-building, which has, through Habermas, Anderson, Taylor, Warner, and others, long privileged “the press” over “the platform” in the formation of the democratic public sphere (pp. 3–4). Bate demonstrates that this emphasis on print capitalism is both due to the difficulty of studying the unrecorded, as well as Western semiotic ideologies that privilege the denotational functions of language over its poetic ones, leading to an elision of the sensuous, affective, and embodied aspects of speech as social activity. Bate invites us to consider how communicative genres like homiletic oratory, devotional songs (bhajans), poems, and mass meetings together constituted a novel “interpellative infrastructure” of politics (p. 66). Through these practices, previously disparate groups were called into being as a singular political public at the turn of the 20th century in Tamil Nadu. Rather than reproduce the dichotomization of speech and materiality, Bate emphasizes their mutual imbrication. In the Indic context, vernacular political oratory gained semiotic value not merely from its denotational accessibility—an influence of Protestant ideologies of textuality—but from what Bate calls “poesy”: the corporeal, sonorous, musical qualities of speech that are prized in Tamil textual traditions. How might this intervention contribute to broader efforts in anthropology to theorize language and/as materiality, and how might such an approach allow us to decenter Western theories of politics and democracy? 

Continue reading