On Discourse

With graduation approaching, I feel an urge to settle accounts with my own past. A while ago I wrote “Academic Language and Everyday Language,” which I think of as my “first critique.” This piece is still about the conventions of academic conferences themselves. The first critique actually began as a “PPT design tutorial.” While writing that tutorial, I realized that designing slides is basically an act of telling, not of visual presentation. So the root of everything is expression.

This expression takes two forms, spoken and written. The first is the talk given at a conference or lecture; the second is the paper. I will start with the strange things about papers, and then say something about lectures.

Talking with a friend recently, I said that the overwhelming majority of humanities papers these days (I mean specifically in art studies, philosophy, literature, and journalism and communication) seem to me to follow a remarkably uniform pattern. First there is a thesis, or an argumentative aim, A. Then the author picks out countless theoretical resources and bits of jargon (sometimes mixed with a peculiar rhetorical style) to wrap and cover it. Call this B. What the reader finally receives is A wrapped in B , and we can call this whole C. As readers, the sensory whole we take in is supplied by C (the “academic flavor”). But in order to understand what the author’s A actually is, and whether the argument for A has been properly carried out in the paper, we first have to struggle through B to get to A. Yet our cognitive energy is limited. Often, by the time we have made our way through B, it has been used up, and we no longer have the strength to tell whether that A holds up at all. And in many cases, once we do manage to extract the paper’s “A,” it is not hard to see that it is very likely a self-evident conclusion, or one not worth stating at all, and certainly not one that needed so many theoretical heavyweights to stand up for it. The thicker the layer of B , the more “scholarly” the paper looks, and of course the harder it is to “understand.”

Thickening B benefits almost everyone, except those readers who actually want to learn something about the world from it. For the author, investing in B rather than A does three things. First, it lets them feel, while writing, that they really are grappling with some tremendously complex problem. Second, it keeps their real views from facing the reader’s scrutiny on their own terms. After all, anyone who wants to question the paper’s “A” has to make their way through the dense fog laid down by B, and this threshold of cognitive effort alone screens out most people. Vague expression is always safe; transparency means taking responsibility. The third is perhaps the most important: it gives people who have not really mastered the ability to argue a shortcut for surviving inside the academic system. That is, rather than seriously wrestling with the problem in front of them, they need only imitate others and cite more theorists, and this will get them through most of the time, because hardly anyone is willing to pour much energy into a subject they are not very familiar with to begin with. The training for this skill is simple and certain. It only requires you to learn the crafts of the tailor and the parrot, not to stay for a long time inside the uncertainty that complexity brings, grinding away at that complexity. “B,” as an object of knowledge, is always stable, and therefore always within reach; a question or idea that does not know where it is going is not.

Furthermore, caring about B leads authors to think that they really are using the resources of B to advance their arguments, but very often these theoretical resources are applied extremely crudely. They are interchangeable, and they are never tested or calibrated against empirical material. Theory works like a mold: any empirical material placed in it will smoothly produce something academic, but the theory itself never takes on a new shape under pressure from experience. In this relationship, both experience and theory are inert. One might also look at this: when an author claims to be arguing for something, do they first sketch what the problem itself looks like, or do they first “borrow so-and-so’s perspective” and use that to push the paper forward?

As for editors, since they inevitably bear part of the responsibility for a paper’s quality, the best way to avoid “mistakes” is to publish papers in which “B” has a very strong presence. On the one hand, this proves the author’s “academic competence” (guaranteed by the atmosphere created by a stock of terminology and a set of syntactic relations), so the editor can say that they have lived up to what an academic journal is supposed to be. On the other hand, it keeps readers from actually testing the paper’s arguments and conclusions. After all, if you published a paper with few theoretical resources and transparent language, some readers might, because the language is so plain, “try to understand” it. That risk is often high, even though simplicity and transparency of linguistic form do not mean, and cannot guarantee, simplicity in what the language says. Thickening B in turn thickens the protective walls around the academic community made up of authors and reviewers, because in this situation the main function of B is to block and defend: to keep people outside the field from coming in and disrupting its rules of the game, and to keep those inside the community who have not yet caught on from discovering the trick. Here B is a kind of identity card for the community; it proves that we are all playing the same game. Unfortunately, for the past few years I too have been in the business of constantly reinforcing “B” in order to make the lies hang together, and predictably, in order to scrape by in this line of work, I will have to keep doing it.

A “paper” (lunwen, 论文) must of course “argue” (lun, 论), and this “lun” means argumentation. It is a craft that has to be learned. The aim of argumentation is to show that some idea of mine is right. That is to say, argumentation is necessarily accompanied by an awareness of whether the argument is valid. In this sense, what does cultivating the craft of argument mean? On the one hand, it is undoubtedly a kind of meta-awareness of one’s own thought process, without which there is no argument. For argument demands that I not only think about the problem, but also check, at the same time, whether my way of thinking about it is reliable. Once this meta-awareness is in place, it keeps running. On the other hand, this self-awareness about validity means that we have accepted another kind of constraint. This constraint does not come from the B I described above, but from the internal demands of the structure of argument itself. An argument is either valid or invalid; the judgments I put forward can in principle be checked by anyone, and do not depend on standpoint, titles, clout, feelings, or rhetoric. That is to say, accepting this constraint means admitting that there is a standard beyond my personal will, and that my judgments must answer to it. The objective effect of investing in “B,” as described above, is to let authors evade this constraint. As long as one first overwhelms the reader with a feeling of sublimity, an avalanche of sensation, this kind of checking can be kept from happening, and the question of “valid” or “invalid” never has to be faced.

Putting these two aspects together, we might say that training in argument is fundamentally a decentering of the self. It requires authors to be able to step back from their own standpoint and examine their reasoning in an impersonal way. That is to say: refusing to treat one’s own perspective as the only perspective, and refusing to treat one’s own feelings as an argument. For feelings remain thoroughly one’s own, while argument is directed toward the other.

What is rather interesting is that a person can have an extremely strong capacity for argument and at the same time choose to spend it on questions that do not matter at all. For self-awareness about validity does not automatically bring the judgment to see which questions are worth arguing about. That is a value judgment. Value judgments are the source from which all forms of knowledge-discourse arise, and value judgments can only come from the lifeworld. A person may know how to make an argument rigorous without knowing why arguing this question matters more than arguing that one. That is, the skills and craftsmanship become ever more refined, while the sense of a problem that drives the argument grows ever weaker, until it finally becomes a technical operation inside an already established framework of problems. It is like the metaphor Mu Xin once used: the academy is inhabited by a crowd of boors who cultivate their plots with meticulous care. So if the cultivation of argumentative skill is not accompanied by a continual asking of “why argue about this?”, it becomes a capacity that seals one off. In the end we learn to do more and more precise work in a smaller and smaller space, and dignify it with the names of specialization and scholarship. But is this space itself worth the energy we put into it?

My mentor, Professor Qiu Xiaolin, once said to me: “Argument already contains an awareness of facing the other, so it is naturally decentering.” I could not agree more. Argument is not talking to oneself; it is talking to someone who may disagree with oneself. Every step I take must be one that another person could in principle follow. Without this presupposition, it is not an “argument” at all; it can only be called a “monologue.” A monologue, at bottom, is solipsism, because what it implies is: whatever holds for me holds for everyone.

“Argument” can be understood as a kind of teaching. In Inquiry (I), Kojin Karatani draws a distinction between “teaching” and “speaking” that made a deep impression on me. “Speaking” is passing along information on the premise that the rules are shared, whereas “teaching” is passing on the rules themselves to someone who does not at all understand the rules I hold. That is, when we “teach” a foreigner to speak Chinese, or “teach” a child to talk, we clearly know the “rules” of Chinese, but we cannot state them explicitly. It is easy to point out that here one should use “le” rather than “guo,” but why? The crucial point is that the teacher is in a weaker position relative to the listener. This, I think, is the difference between argument and propaganda: propaganda is also directed toward the other, so why is argument different? The one who argues is like the teacher Karatani describes. Whether the teacher’s words mean anything is not up to the teacher. Whether a teacher has taught a student to speak Chinese depends on whether the student has learned it, and the teacher is often powerless here. They cannot force the other person to acknowledge the effectiveness of their teaching by appealing to a premise both share, because that premise (the grammatical rules of Chinese) does not yet exist for the student at all. All the teacher can do is put forward what they have, then wait for a result they cannot control, and take responsibility for it. This is the substance of argument: when I argue, I am handing over a judgment, handing it to someone whose reaction I cannot control. Whether the argument is valid is for the other person to decide. This is also the difference between “argument” and “dogmatic assertion”: the dogmatist presupposes that their judgment holds automatically, while the one who argues knows that this depends on the other’s checking. In the structure of speech, and even in an ontological sense (for Levinas), the “I” is absolutely in the weaker position relative to the other.

So in what sense is this an “other”? The point is simple enough: it is a living individual whose lifeworld is entirely different from mine (the arguer’s). Before this person is an “academic expert,” they are first of all a being with a lifeworld I cannot reach. And we turn a blind eye, almost systematically, to this glaringly obvious fact. The setting in which this blindness is at its worst is the ubiquitous academic lecture and conference talk.

In “Academic Language and Everyday Language” I already touched on the topic of “academic talks nobody can follow.” But I want to come at it from another side: why is a 15-minute talk such an exhausting thing to sit through? The reason is clear enough: we cannot follow it. And the reason we cannot follow it can hardly be that the speaker “does not know enough” about their object of study. Quite the opposite: they usually know it extremely well. Nor can it be put down to a lack of learning or comprehension in the audience, since the people sitting below can often write academic papers every bit as good as the speaker’s. So where is the problem? It lies in whether language, as sound, can be parsed by the ear. I have already analyzed this split in language in the “Ordinary Language” piece, and will not repeat it here.

What I want to say is this: as speakers giving academic talks, we all seem to hold a strange presupposition. Since everyone is speaking in an academic setting and has the relevant background, I can proceed directly at the level of concepts, in something close to written style; that is, I can communicate in “the language of my peers.” But let us think about it: what is the basic task of a lecture or talk?

The reason the lecture or talk exists as a form is that a group of people are sitting in front of us, having spent real time out of their lives to bring their ears and take in our words. This is the basic fact of a lecture. And if this basic fact is not the starting point from which we organize everything we mean to express, then we have chosen the wrong form: you should be writing a paper, not speaking on stage. If, after a lecture, no one in the audience knows what the speaker was saying, then however precise the thought and however coherent the handling of concepts, the lecture has failed as a “lecture.” It may be a good paper, a good piece of research, but it has not done what a “lecture” or “talk” is supposed to do. If we go back to the plainest question, “What is a lecture actually for?”, this whole logic looks utterly absurd. Before those listeners are members of an academic community and experts, they are first of all living people. They have lifeworlds unknown to us, which appear mysterious precisely because they are foreign; these worlds are in principle inaccessible to me, and they are what make us different from one another. As living people, carrying the identity of “the academic community,” they have spent real time coming to hear us speak. We have a responsibility to make ourselves understood. This responsibility is not automatically waived because your ideas are complex, nor because the people in the audience are Changjiang Scholars, experts, and professors. Living ears, living attention, and living working memory all have limits that hold for everyone.

By contrast, the default mode of most academic conferences goes like this: first hand everyone a paper abstract that is hard to make head or tail of, then, on site, pour out content at extremely high density in “the language of peers,” presupposing that the audience shares the same body of background knowledge. The reason this practice is so widespread is that it was never meant for the other in the first place. Its sole function is self-certification; it is a business card. By “business card” I mean: I am selling off my whole identity as an academic researcher, rather than handing over to you my real judgment about some question. So this self-certification is not at all what Karatani calls “teaching,” nor is it argument; it contains no awareness of the other whatsoever. It is merely a display.

We should organize our language according to other people’s standards, not according to our own views. The complexity of thought requires us to put more effort into expression, not less. This is the speaker’s responsibility. Again it was Professor Qiu Xiaolin who, years ago, kept telling us that a good speaker should be like a “tour guide.” It is only a pity that it took me many years to really grasp what he meant. The guide’s task is to remind and to point the way, to take people through territory that was entirely foreign to them before. Language is the guide’s pointing finger; it reminds us to notice things we could not notice without someone else’s guidance. So expression itself should be as transparent as possible. If, during the tour, the visitors keep asking themselves, “What does he mean by saying this? Why is he saying it?”, that shows their attention has been drawn away from the things and onto the guide’s act of “speaking” itself. Then the guide has failed. As Wittgenstein says in §§126–129 of the Philosophical Investigations, “The work of the philosopher consists in assembling reminders for a particular purpose,” and “The aspects of things that are most important for us go unnoticed because of their simplicity and familiarity. (One is unable to notice something—because it is always before one’s eyes.)” The task of the guide is the task of philosophy. These truths can almost only be lived through, not grasped by the understanding. They are effective and whole only when they happen as actual lived experience. So although I knew this long ago, I did not truly understand it until quite recently.

Thinking of oneself as a “tour guide” really means placing others at a higher rank than oneself. Karatani has already spelled out this hierarchy very thoroughly. Placing others above oneself also means that the “responsibility” for a lecture belongs to the speaker, not the audience. The guide has to chew through the tough bones in their own research. For the guide is the one handing something over, and whether it can be received depends entirely on the other side. If a product does not sell, you cannot blame the buyers. The actual working logic of academia is the reverse: this responsibility is usually assigned to others, which frees the speaker from the weaker position. They no longer have to bear the uncertainty of “Have I actually gotten my meaning across to them clearly?”, because they have already shifted the responsibility for failure onto the other side in advance. This is refusing to face the other, only dressed in academic sheep’s clothing. If we truly accepted our weaker position as speakers relative to the audience, then these people could not be bystanders who happen to be present while we talk; they would be constitutive conditions of our act of “speaking” itself.

Most academic talks come with a basic limit: 15 minutes. This is often felt as a constraint: my research is so complex, how can I explain it clearly in 15 minutes? But this complaint rests on two premises that I think are unreasonable. The first: the speaker has not achieved a free command of their object of study, and so cannot draw sharp distinctions, and can only use length of time to hand the responsibility for drawing distinctions over to the audience. In this case, even if the time were longer, the problem would not fundamentally change. The reason more time seems to avoid the problem is that it gives listeners time to digest and organize the content themselves; this does not mean that the structure of responsibility has changed. So this looks as though it were simply a question of depth of research: if I understood my research more deeply, would the problem be solved? Of course not. That brings me to my second point, which is more fundamental: the speaker has no awareness of being a “guide.” That is to say, the people sitting below simply do not exist as constraints in the speaker’s mind. Whether preparing the script, organizing the language, or opening their mouth to speak, all they ever have in mind is “what I want to say,” and the people down there merely happen to be watching their monologue. So their behavior is still, in fact, dogmatic.

A guide’s most important skill is reading the room. What reading the room yields is “talking like a human being.” By “talking like a human being” I mean that language must stay in constant contact with experience. Which raises the question: when someone giving a talk begins to speak, how do they understand themselves and others in this setting?

If they think that the most basic situation of the people present is that each is a vivid, independent individual, then they will naturally start from the basic experience of the lifeworld that we all share. For the only thing living people are sure to share is the recognizability of experience. The people in the audience may not be able to reproduce or follow my chain of argument, but every one of them can, on the spot, compare my account with their own life experience, and discover where the two fit together or come apart. So language has to do an extremely difficult job: by making itself transparent, translate concepts into experience. Or, to put it another way, let concepts grow out of experience on their own, rather than deriving them from other concepts. This is in fact a rigorous test of the speaker: if they can anchor a concept in experience, that shows they really understand what the concept means. If they cannot, and can only explain one term with another, then their own understanding of the concept is often left hanging in the air, merely going around in circles among concepts without intuitions. In this sense, being aware that the other and I coexist in the lifeworld is not simply a matter of “being responsible to the audience”; it is fundamentally a matter of being responsible for one’s own thinking.

On the other hand, if the speaker assumes from the start that the most basic situation of those present is that they are experts and scholars fluent in academic jargon, then they will inevitably keep reinforcing the “B” I described at the beginning. Experience then becomes an optional aid; their language will necessarily start from concepts, and will grow more and more opaque. This setting is highly efficient and offers little operational resistance, since it lets one skip all the work of bringing concepts down to earth. But the price is that it costs one the ability to check whether one really understands what one is saying. For when language always runs along the high ground of concepts, its user cannot tell whether they are really thinking or merely arranging terms skillfully. The arrangement of terms can produce the appearance of thought (Schein, in the double sense of the German word, “illusion” and “radiance”), and we ourselves can be persuaded by this appearance. More importantly, this brings us back once again to the question of “responsibility”: choosing the first setting means exposing oneself to being checked, and taking responsibility for one’s judgments and one’s way of speaking. If we choose “living people” as the default setting, the experts will still understand, because they are living people too, and they will in fact see our thesis, the “A” from the beginning, all the more clearly for it. Whereas if we choose “experts” as the default setting, the obstacle of “B” appears, and at the same time it becomes a kind of torment for living ears.

What I have been talking about all along is “responsibility.” “Responsibility” can be parsed as “response” plus “ability.” To be responsible means to have the ability to respond to others. To respond to others through language means that I try to hand over the rules I hold to the other person, and invite them to test those rules against their own experience. In this process, the responsible person lowers themselves into a weaker position, not out of submission to some superior will, but out of a more honest attitude: I am willing to cede my own presence to the truth that temporarily appears through my mouth. In this ceding, a more complete existence is able to show itself, and “I” am merely its medium. Eliot, Nietzsche, Goethe, and many other great masters have said as much. Without the recognition of the other, this structure simply could not hold.