Translation as a filter


Translator : WALKER Gavin

  • fr
  • en
  • Multilingual mode
  • Simple mode

Naoki Sakai

Translated from Japanese by Gavin Walker and Naoki Sakai

When we seek an explanation of translation, we too often and too facilely speak of translation as if its central aim consisted in the transfer of a text written in one language to a text written in another language. Here, I choose to utilize this compound “central aim” (“primary lens,” shugan) precisely because it is difficult to draw narrow conceptual limits around the word “translation,” which is almost always used metaphorically. We might say that translation possesses a strongly amplificatory character because it is utilized flexibly from a single word or combination of a few words, to the proposition, the paragraph, the single work, and then the complete works of an author.

If we consider the translation of a single work such as Notes from the Underground, we can also speak of the translation of the collection of works by the author known as Dostoevsky. That is, we become able to speak of a sequence as a single bundle or assemblage by means of translation, moving from the commensurable to a unity that is combinatorially formed from the multiple. 

Precisely in this sense, translation operates by exceeding the narrow meaning of language. A novel is translated into a film, just as a political idea can be translated in action. A human being’s creative capacity can be translated into capital, their desires translated into dreams, their aspirations translated into seats in parliament. Translation passes through and circulates in the intervals of different instances of meaning, threading together discontinuous contexts. As a consequence, translation is conceived of as something that is particularly metaphorical within the metaphor and thus often referred to as “the metaphor of metaphor.”

We cannot forget that the term translation contains a doubled sense of meaning. Translation is the work or process of re-writing and re-stating, but at the same time is the text, which emerges as a result of passing through this work or process. To speak of the translation of Proust’s À la recherché du temps perdu on the one hand signifies the task of repeating the original text in another medium, and at the same time, it also connotes the new text that is born as a result of this task. From the outset, the work of translation differs from that of literary creation in that translation has the character of imitation performed on citation. It is possible to cite words, sentences, books, images, human actions and so forth, but the cited referent is presented as an assemblage, a bundle (hitomatome).  Citation is “shifting something elsewhere” and at the same time, “assembling or bundling together,” in other words, it is an act of unifying. We cannot speak of translation in the case of something which cannot be cited. If you cannot cite, you cannot translate. That which is cited is separated from the actual context of citation – the cited text possesses a different temporality from the context of citation. It is precisely for this reason that translation can be something like the “afterlife” of the work. Yet translation is never wholly unrelated to the original work. Translation is a “repetition” of the cited original text. The fact that the original work and the translation mutually reference each other neither indicates that they share the same message, nor that they convey the same signification. Nonetheless, some creative moment is never negligible in translation, and this is creativity that accompanies repetition.

In this essay, I aim to liberate the possibility of translation from the curse bestowed on it by the view of translation organized around the image of  communication: the communication of a written text from one language to another. Translation is not a task limited to the written word, but a concept which grants us the possibility of examining social action in general anew, something which offers us an invaluable gateway by which to enter an inquiry into sociality itself. Nevertheless, the traditional view of translation has elided this potent sociality that suffuses it, through its collaboration with the substantialization of “national” and “ethnic” languages. It goes without saying that the argument regarding translation that I offer here tries carefully to avoid lapsing into another systematic differentiability - known as phonocentrism - of the written and the spoken. But this is not all. By “text” I do not mean the traditional view of the text which limits it to documents or books, nor do I adopt here the widely-disseminated dichotomy between the practical task of oral interpretation (tsūyaku) and the translation (honyaku) of scriptures, philosophy, and literature in written form. I simply do not accept the distinction between interpretation and translation precisely because I want to examine the operation of metaphor, which suffuses the situation of translation, while simultaneously historicizing the traditional view of translation.

In studying translation, we must pay close attention not only to how trope operates, but also to how it malfunctions. In other words, in order to devise shifts in the discussion of translation, we not only need a transformation of the basic concepts, but also a recomposition of the tropes and figurations that we employ. Today the very presumption that a language has its inside and outside must be scrutinized, and we must call into question the regime of translation according to which one language is represented as external to and exclusive from another language spatially.  I have referred to this regime of translation, in which translation is represented through the strict distinction between the interior and exterior of a language, as the “homolingual address.” In my view, we must historicize the stigma of this regime of translation while at the same time, turning ourselves towards thinking of translation as a “heterolingual address.”1 The “homolingual address” derives its legitimacy from the vision of the modern international world as a juxtaposition of state sovereignties as well as the  reciprocal recognition among nation-states. Of course, the international world and the nation-state offer mutual reinforcement and form a system of complicity. In order to unravel this traditional view of translation, and to recombine the tropes of translation towards a forum for the elucidation of  sociality, the trope of “translation as a filter” provides us with an appropriate thematic. Let us attempt to begin thinking about translation from the title of this paper.

On the Title

This title, “Translation as a Filter” was something given to me rather than something that I proposed. Nevertheless, since it was this title that prompted me to write the present essay, I ought to briefly discuss the relationship between the title and the argument I make here. I accepted the title, neither because it accurately named a guiding thread for the discussion of translation that I intended to develop here, nor because I particularly want to take this title as my thematic subject and formulate a justification of it. I do not dare to say that it summarizes my argument well, either. Rather, what sparked my interest was precisely that this title encompassed a certain complexity that cannot be resolved in a typical manner. It contains numerous pitfalls that discussions of translation have often fallen into. Thus what I intend to do here is to utilize this title as a springboard for discussion, disentangling its intricacies in order to attempt to find an escape from the traditional view of translation.

The title of this essay might appear provocative, but it may ring hollow with certain readers since it contains few unexpected insights. The reason that I presume to call it “provocative” is that, like the solving of a puzzle, this title invites a variety of interpretations and is open to multiple definitions. At first glance, proposing a metaphorical relation between translation and a filter seems understandable, but in fact, one quickly develops a nagging feeling of incomprehension. In conjoining “translation” and “filter,” there are too many indefinite elements that intervene between these two terms, and thus even the provisional judgment “translation is something like a filter” immediately renders this title unacceptable. In what ways and as a result of which aspects, can the term “filter” serve as a metaphor for translation? Is it not the case that precisely because we utilize this term “filter” we become incapable of moving past the restrictions it places on translation? It is my contention that, in order to gain an understanding of this type of metaphorical judgment, we cannot avoid the fact that we lack something urgent, that we need a more persuasive explanation.

Nevertheless, this title anticipates a certain view of translation – the mode of being that the term “filter” describes, in fact expresses this view perfectly. In the traditional view, translation is often grasped as if some already determined “meaning” passes through a barrier, and thus the figure of the filter effectively corroborates this representation of translation. From such a viewpoint, the filter is a curtain or barrier permeated by a fluid mediator. Of course, the term “filter” describes something which allows certain things to pass through while blocking others; thus it is only at the point where permeability and impermeability coexist that a certain blocking entity comes to acquire the characteristics of a filter. A filter is precisely a semi-permeable membrane. Permeability presumes the existence of a mediator which passes through it, and therefore within it are flows and movements; a filter, by blocking a flow which has a certain directionality, is put under pressure by this mediator. Thus, we might unfortunately imagine, inspired by this figure of “translation as a filter,” that translation is a situation arising only when there are two sides: something that passes through and something that does not pass through. In this view of translation, the coexistence of permeability and impermeability is presumed, and thus there must be a flow with directionality. Furthermore, the filter indicates a site where there is a curtain or barrier as obstacle. This is often imaged as a line that bisects a surface, or as a surface which bisects a space. The basic material property of a filter is to be something that obstructs, something that hinders movement, even if it is full of holes or permeable, and thus those things that cannot pass through it are gathered in the filter and held in stasis. As a result, the impermeable objects that previously circulated freely are held in place at the site of the filter, and prevented from slipping through to the opposite side. This is the metaphor that first emerges when we intertwine the terms “translation” and “filter.”

A crucial function of translation is frequently alluded to at the starting-point of this trope. A filter selects and classifies what is permeable and what is impermeable from something mixed. Differentiating what can pass through and what cannot pass through is precisely the act of filtration, that is, the term “filter” always indicates this act of filtration. Practically speaking, the function of filtration, as a metaphorical connotation, has often insinuated its way into the discussion of translation. In other words, it is precisely here that we are encountering the pitfalls inherent in the metaphorical statement, “translation is something like a filter.”

Communication and Translation

The discriminatory function of the filter is not limited solely to the classification of the permeable and the impermeable. It also differentiates into two distinct areas a space which is presumably connected on this side and that side. It splits one contiguous space in to two. This function of filtration is possible only when this filtration is unidirectional, when the filter operates as a threshold, and only on condition that the upstream and the downstream flows are not blended together. Through the exclusive partitioning of space, the filter acquires another trope of discrimination: border. The filter thus takes on the sense of a national boundary or enclosure, that is, not only the partitioning of space but also the partitioning of the surface. On the one hand, the national border discriminates between those who can pass and those who cannot. Further, the national border is the site of the customs boundary; it distinguishes between certain things that can pass through it and others that cannot. Nevertheless, on the other hand, the national border constitutes the outer edge of the territoriality of the sovereignty of the nation-state. If you cross the border, the sovereignty of the nation-state operating on one side becomes invalid on the other. In other words, the enclosure is an apparatus that discriminates between those allowed to enter and those who are not, but at the same time marks the outer edge of the land as property. Thus, the figure of the filter can be expanded to encompass the distinguishing of heterogeneous areas of a surface, the establishment of demarcations between interior and exterior on the land, and the mapping of sovereignty and ownership; thereby it governs the communication between areas. Thus, in our examination of translation, the filter acquires yet another metaphorical function. Translation serves as a boundary that distinguishes the space. Its role is in introducing the threshold into a space.

It is not particularly difficult to discern how various characteristics of language are being articulated within this economy of trope. A bundle of articulatory paradigms and generative rules such as the regularities of phonetics, morphology, and syntax are seen as the special characteristics of a given language, and are often thought of as the archetypal examples of what does not typically pass through the filter. Can we not say the following: the paradigms according to which an enunciated voice is articulated into phonemes, the generative rules of comprehension and composition, the criteria which combine words together, and indeed the systems of classification that distinguish words as morphologically significant units express the particularities of a given language, and constitute precisely the typical example of what is erased by translation? Or maybe we should put it another way: to transmit a text into another language is to erase the particular characteristics of the original language; the filter as translation manifests itself through the erasure of the grammatical traits of a particular language.

The  conception of translation according to the model of communication finds its raison d’être precisely in the tropic economy that I have sketched out here roughly. This is to say that the model of communication cannot be maintained unless the transmitted content and the rules of communication can be clearly separated. Transmitted content is generally seen as “information.” Throughout the latter half of the twentieth century, the term information swept through the fields of economics, cognitive science, manufacture, technology and so forth. “Information” means a knowledge transferred by means of the act of informing. In other words, it is what one is informed of. To inform is to advise or teach, by giving form and shape to the spirit of the other, and the information thus communicated has the characteristics of a message delivered by a messenger. Whether or not the institutions and technologies develop, from a paper letter carried by the courier to the postal service managed by the national state, or from the cabled telegram to the international wireless internet, the theory of communication is incapable of shedding its reliance on this schema of the message handed over by a messenger.

In the model of communication, the transfer of information is understood as parallel to the metaphor of the messenger communicating a message, from addresser to addressee. Generally speaking, however, this point is not understood even in arguments that attempt a scholarly classification of translation, such as Roman Jakobson’s2. The apprehension of translation according to the model of communication conceives of translation as a specific example of this type of general communication. The textual reading strategy known as “deconstruction” has already demonstrated in detail the strict impossibility of maintaining a clear discrimination between the communicated content (message) and the rules of communication (code).

Translation studies, a scholarly discipline established in an increasing number of universities in the world today, for the most part conceives of translation premised on the model of communication, thus trusting in the possibility of a principled distinction between message (content) and code (rule). When content is translated from one language (to adopt the vocabulary of translation studies, the “source language”) to another (the “target language”), this content contains elements that, like a proper noun, do not necessarily follow the rules of a particular language. It is generally accepted that a proper noun is not translated, nor is there any need to do so strictly following the code of the target language. Aside from such an exception, translation is expected to be an all-encompassing transformation of rules (code). When content (message) in the source language is translated into content in the target language, the rules of the source language should presumably be completely erased from the content expressed in the target language. Translation conveys content to us, but does not teach us the grammar of a different language. Thus, by seeking the distinction between the translatable and the untranslatable solely within the communicated content (message), the communication of rules (code) is foreclosed from the outset, separated by the mutual exclusion between content and rule in this economy of metaphor. It is anticipated either as a translatable message or as an untranslatable message. Because the grammatical rules or the particular qualities related to the organization of language are excluded from things that can pass through the filter, the materiality of the text is not examined as something translatable, and is thus neglected. Consequently, the distinction between the translatable and the untranslatable is anticipated only on the level of the communicated content (message). That is, according to the model of communication, the untranslatable is determined from the start as something in the content of the communication: the untranslatable is only anticipated as “the part of the message that does not arrive.”

The Symbiosis of Culturalism and Subjectivity

Further, we can also associate this tropic economy with the typical arguments on subjectivity. A person cognizes things in the world through a certain system of categories. It is not easy to objectify this system of cognitive categories as a whole, but it may appear comparatively simple to identify it in terms of differences between one language and another. The confinement to one’s (native) language may well explain the confinement of one’s subjectivity to one’s (native) culture. We can see most starkly the conspiratorial linkages between the model of communication and culturalism precisely in the discussions of subjectivity that are bound up with translation in the representation of language. Here too, the trope of the filter exhibits a new force.

We are born within a given language, and acquire the ability to cognize the world under the grammatical rules of this language. Many might accept this as a valid claim. It should follow then that, well before we produce words, before we gain a knowledge of other languages, our cognitive capacity should be determined through already given cognitive categories; we should be able to perceive the world only through a given filter. In this way, discussions of subjectivity jump too quickly to conclusions by way of the spatialized trope of a language, a spatialized figure with a clear contour. I have no intention here of reducing the discussions of transcendental subjectivity that emerged in the 18th century to the problem of culturalism, but the trope of “translation as a filter” clearly exposes the symbiotic relation between discussions of subjectivity and anthropological culturalism, such as that inspired by American Structural Linguistics presented by Edward Sapir and Benjamin Whorf.

The filter, which distinguishes the permeable from the impermeable, enables the representation of two different spatial areas, but that is not all. It also forms these areas into spaces saturated by differing systems of grammatical rules (rules that are organized by means of phonetics, syntax, and so forth). Here, “different space” carries the connotation of “different language.” Language, which is assumed to be a potential system of rules (of phonetics, morphology, syntax, and so on) is given a spatialized figure as if it were a closed area. If individuals enter the world burdened by their language of birth (their “native language”), they would have been born already located in one of these specific spatial areas to the extent that they depend on the trope of “translation as a filter”. In other words, the area distinguished by translation becomes a space that expresses a primordial belonging, which symbolizes the destiny of the individual. The location in this space is imagined as the destiny of one’s cognitive capacity that cannot be changed by the individual’s own initiative; it is an innate trait, like colorblindness, an ability given with one’s birth. I experience the things of the world through a certain system of categories, so in principle I have no access to a position from which to judge the relevance or irrelevance of my experiences as a whole . There is no way for me to judge in advance whether or not the world I am given is biased or distorted. I might see the world through colored lenses, but these colored lenses are the only lenses I was born with, which goes by the name of “native language.” Regardless of whether my eyeballs themselves are in color or not, I cannot be given a perspective that could correct my own prejudice towards the world that I embrace in perception. One reason that this phrase “translation as a filter” has a certain persuasive power is that it prepares us for the deployment of a trope which makes it possible to mobilize translation within this type of argument on subjectivity.

At the same time, let us not overlook the following point. By inciting us to construe the representation of translation in terms of the tropic figures of the upstream flow and downstream flow of the filter, or as two spaces separated by a barrier, “translation as a filter” enables us to figure language as a space distinguished by boundaries.  This trope serves as a schema for the spatial figuration of language.

Translation, insofar as it is a filter of permeability, separates space into two areas, but whether or not these separated spaces are necessarily formed as enclosures has not yet been problematized. For now, however, the filter does not merely divide the continuous space into two; it implies an overarching condition that restricts my capacity to cognize the external world, like the camera accessory that selectively transmits light coming into a camera obscura. Here, it is as if this “translation as a filter” has come to possess the character of an optical filter, covering the main lens of a camera, rather than the form of a filter as a semi-permeable membrane.

Yet, curiously, although the figure of the filter is called for in order to postulate the innateness of cognitive capacity in terms of spatial belonging, it becomes a principle by which to explain the typical situation of cognition prior to translation. Knowledge that is acquired by means of translation is postulated in order to explain the apparent existence of innate capacity prior to translation. To gain an awareness of the constraints immanent in the cognitive capacity in one’s ‘native’ language, one must have the experience of othering /being othered (ika) in relation to a foreign language, and consequently, those who can neither speak nor read a foreign language are not even capable of being aware of the constraints imposed upon them by the ‘native’ language. Without “the incomprehensible,” or without an encounter with someone one does not comprehend, one cannot become aware of the limitations of one’s own cognitive capacity. Translation folds over upon itself and gives rise to the moment of what may well be called ‘reflection.’ Let us note by passing that the reflection into native language has a fundamentally different temporality from that of transcendental reflection, and we should not confuse these two. Nevertheless, it is still worthwhile to keep in mind that translation provides a negative moment in relation to the native language, and that, therefore, without the presumption of a foreign language, the awareness of something as native language should be impossible from the very outset. Thus, it should be argued that the nativism that posits native language as an innate condition for one’s cognitive and practical capacity becomes possible for the first time only in passing through a moment of negativity in relation to native language itself. Is it not the case, as far as the tropic economy of “translation as a filter” is concerned, that to establish the identity of a native language is to posit translation as a “mediation (Vermittlung)”? Of course, in this argument so far, I have not overtly doubted that a native language could ever exist in and of itself.

The filter of permeability, divides two spaces, but the properties of each space are determined only in relation to the other space. Each space is determined relatively, dependent upon the other space it is paired with. However, as soon as the trope of filter acquires the sense of an optical filter, the determination of the space’s property is altered: it is no longer relative. Each of the spaces separated off from each other by the filter come to be represented as if they already possessed these properties as determined within themselves. Each space is represented as if it were pre-determined in its properties irrespective of the other space from which it is differentiated. Then, people tend to forget that translation is first and foremost expressed in the verb “to translate,” that translation is an event, an action, a movement. This is analogous to the way in which people forget that the national border is not a natural condition, that it is an institution created through acts of sovereignty by the state, the ruler, the national people, and so forth. In precisely the same sense, we must not ignore the fact that the border itself cannot exist independently of the act of discriminating.

What Enables Translation to be Represented as the Filter

The representation of translation as a filter attributes a fixed and definite image to translation. While it may assist in terms of thinking of translation, it also creates pitfalls for the discussions of translation. The trope of ‘translation as filter’ is not only sporadically utilized in certain exceptional ways, but tends to function as a means of organizing multiple examples in a broad sense; it serves as a schema in the comprehension of translation.

In order to attempt to think what is difficult to think – the representative example of a difficulty for thought is time, and because it is difficult to represent time directly to ourselves, people depend on the schema of time. For this reason, the schema of time is well known in modern philosophy – we rely on tropes and images to provide us with some equivalents in the sensible. The  image or figure drawn on a piece of paper or on a computer screen, a flow chart, and various other visual figures guide our thought in instead of the direct representation of complex topics to ourselves. Thus, when we encounter the representation of a complex topic, we rely on an image, a shape, a trope, and a figure. We attempt to think of the complex topic that is difficult to represent to ourselves by relying on these images. The schema is precisely this sort of equivalent in the sensible that is applied to an object difficult to directly conceive of; through the utilization of the figure or the trope, we render a subject-matter for direct thought representable, and  we call this operation of schema “schematism.”

Of course, it is exactly the role of the schema to construct a relation of equivalence between a theme difficult to represent and a certain image. Equivalence is always a relationality between two terms, that is, without assuming a contrasting term, it is impossible to speak of equivalence. Yet,  in the case of schematism, a schema is called for precisely because one of the two terms is posited as a representation. The relationality in schematism therefore appears to be unidirectionally-determined by only one of the two terms. It may appear that the theme is passive while the figure or image is active. Let us not forget this point: how something is made equivalent depends on the schema. As a schema, an image has the power to postulate something that cannot be represented.

The figure deriving from the trope of “translation as a filter” serves as a schema for the representation of translation. Yet, it is important to note that, in the representation of translation, not only one but also two figures are called for; translation is represented as filtration between two schemata.

The way in which translation is represented is sustained by a certain institution. It can give us a false sense of familiarity and self-evidence. But in what way is this convention of representing translation through the trope of a filter itself formed? How can we historicize this type of institution? First, let us point out two opposite directions in which the image can unfurl. The first is oriented towards the presence of the filter itself, and in doing so, leaves indeterminate the two areas divided by the filter. The difference between the two areas is dependent upon how the filter classifies and what it filters, explicitly showing us the difference between the upstream space, a mixture of the permeable and the impermeable in the process of permeation, and the downstream space, purified by the sole presence of the permeable. Yet, precisely the opposite conception is also possible: an orientation towards each of the divided areas. Here, the filter becomes a void or absence. It is possible to see translation as an act that links the gaps or ruptures between the two areas, rather than as a substantial barrier of filtration dividing a continuous space into different areas. Instead of considering it as a positive blockage that divides space, we can understand it as a negative interruption that produces impossibility of passing through by rupturing the ground. Then, translation is seen as an act which links two areas that are detached by an abyssal gap, as an operation of crossing over, as a leap to the opposite bank. The filter is transformed into something negative, a symbol of absence. The two orientations suggest two contrary views of the filter, one a rupture isolated by the abyss and the other a barrier preventing permeation, as a void that separates humans in contrast to a porous obstructing entity.

The orientations are in opposite directions, but are complementary alternatives in the model of communication; in either orientation we are led to the common presumption. Should we say that “the cup is half full,” or that “the cup is half empty?” Do we see two faces joined in profile or simply the body of a vase? A shape is dependent on the viewpoint; it can unfold itself as a trope of either substantiality or absence. When we approach it as substance, translation is close to the figure of a filter; if we approach it as a void, we associate it with the image of crossing a bridge.

This is why the metaphor of filter invites the other metaphor of abyss, for these two metaphors are alternatives complementary to one another. However, when the filter’s existence is transformed from something substantial like a barrier or semi-permeable membrane to something non-substantial like an abyss or gap, something happens that cannot be overlooked: the determination of the space divided by the filter is surreptitiously transformed. When the space is divided into two spaces by the substantial filter, neither space possesses a principle of unity. In other words, these spaces can only manage to maintain their own positions precisely by being divided by an obstacle. However, if the filter is taken to be something non-substantial, then conversely, this space begins to take on substantial characteristics. We begin to be able to conceive of these divided spaces as if they possess internal unities. If we go one step further, these spaces each acquire the capability of becoming self-sustaining unities with organic organization.

Although we are now coming from a contrary direction, this contrastive grasp of translation is still predicated on common assumptions; that translation inscribes and confirms two areas as different; translation is treated as the establishment of a connection between this side here and that side over there, two sides that pre-exist. This is precisely the reason why the two figures of the filter and of the void are jointly called forth.

In either case, the representation of translation mobilizes two schemata, and translation is represented as if it were an interaction or bridging of two separate and distinct spaces. In other words, the representation of translation is a schematism of duel schemata, a process of co-figuration.

Obviously, in the model of communication, which sees translation as a kind of switching from one code to another, the de-substantialization of the filter and the substantialization of space are two corollary processes. Thus the two spaces divided by the filter become spaces saturated by languages. As a result, translation is represented as an operation of crossing the crevice between one language and another. When language is represented by an enclosure, the filter is associated with a lens covering the optical entry into a camera and linguistic nativism begins to exercise its power.

A new moment arises when we move towards this trope of an optical filter: space becomes specifically enclosed space. In the trope of an optical filter as constraint to subjectivity, it is not only that space is demarcated; but it also implies that “I,” or “we,” are enclosed in a certain interior. When this metaphor is adapted to the question of language, we move towards a perspective which sees native language as interiority.

What then enters the picture is the image of the “I” or “we” as confined to the space of the native language. Thus, culturalist standpoints such as the national character study or the discourse of Japanese uniqueness are accompanied by the presupposition of a “we” or a “they” as nation or ethnicity confined to this space of native language. In this regard, national subjectivity is a matter of epistemic nationalism in the first place.

When the space demarcated as interiority is determined as enclosure of the national language, national culture, and national subjectivity in the tropics of translation, I call it “area.” It goes without saying that the area in parenthesis refers to the area of “area studies.”

The Schema of the Filter and the World

This schematism presents translation in a representation of the world. The space split into upstream and downstream by the semi-permeable membrane soon slips towards the space of one language and another, and is finally amplified into the split between the spaces of one national language and another. Thongchai Winichakul depicts the historical transition in which the Kingdom of Siam was transformed, over the course of the 19th to 20th centuries, from a kingdom that did not yet possess clear territorial boundaries to a sovereign state with a territory enclosed by national borders3. He traces the process by which inter-state relations linking different states through a tribute system were transformed into inter-national relations between sovereign states identified by their respective territories. In Siam Mapped, Winichakul employs the concept of the “geo-body,” and, by depicting the discourse of cartography in Thai politics, analyzes the gradual transformation from a sovereign power without national boundaries to a modern state sovereignty with a clearly-delineated territoriality. The geo-body does not simply refer to the cartographic image of state territory, but also describes the nation as a community represented as an interiority, newly-determined as an enclosed area. In other words, the geo-body is an apparatus of imagination demarcating an interior “us” from an exterior “them,” facilitating the formation of state sovereignty through its symbiosis with the figure of the homogeneous national community as a result of which the subjects of the Kingdom of Siam began to live as Thai for the first time. Furthermore, according to Winichakul, the inhabitants were gradually unified into a single sovereign state, whereas previously it was normal to belong simultaneously to multiple states. The unification of the people or the nation of the Thai state was established through negotiations with English and French who were gradually colonizing areas around the periphery of Thailand. It was the process through which the Kingdom of Siam acquired legitimacy as a sovereign state within inter-national relations. The English and French colonization of Indochina and the emergence of the modern Thai state were not in contradiction but were instead mutually-facilitating processes.

Let me present two points as to why Winichakul’s investigation is crucial for the thinking about translation. First, he clearly demonstrates that the area in which the nation or people are located was formed through the demarcation of the national border. Prior to this, no technological means were available – neither modern cartography nor the method of triangulation – for the construction of a constant national border, and there was simply no need for unification of the territory because there were many ways to determine interior and exterior in relation to the sovereignty of the state. With the spread of customs that rigorously differentiated human beings of the interior from those of the exterior or compatriots from foreigners, the systematic legitimation of the sovereignty of the modern state was completed. There is one more point to note; this is not one directly addressed by Winichakul but is one that we can logically deduce from his argument. The establishment of the national border does not merely imply the recognition between one sovereign state that monopolizes the territory and another neighboring sovereign state, but it rather indicates a sanction by the inter-national world in general. The establishment of the national border goes hand in hand with the recognition of the international world consisting of the sovereign states, each one determined as an enclosed areas. To return to the trope of the filter, the recognition of the national border is not the division of space into two, but rather the creation of these spaces divided by the demarcation of space as enclosed areas, in correspondence with the creation of independent interiorities formed by their respective languages. In other words, the national community for the first time became possible in the inter-national world, and the destruction of the ancien régime of diplomatic relations based on the tribute system was necessary for the formation of the national community.

Of course, the international world does not refer to the system of natural relations among peoples of the world. It is a global order for mutual recognition among the modern state sovereignties that developed from the 17th century and continued well into the 19th and 20th centuries, in which the institution of the nation-state and modern colonialism came into existence.

What Thongchai Winichakul clearly shows us in the concept of the geo-body is that three processes are interrelated; 1) the representation of the world as the topos of mutual relations among  “national bodies” (kokutai); 2) the determination of every site on the earth’s surface as a coordinate by means of cartographic measurement, and 3) the clear division of the population into the interior and exterior of each of the nations/ ethnicities. The relation of “here” to “there,” or to the “neighbors,” represents the life we live in a variety of ways, and brings a certain order to life by the representation of these relations. In this order we encounter things, events, and people, and comprehend them in our life experience. Let me call the framework of our life experience “the world” and in the world a multiplicity of registers exist; from the quite familiar register of space-time in which “here” and “now” are assigned, the contextual relations of before and after in actions and events, the registers that express the arrangement of rooms in a dwelling or the placement of daily objects, the registers of the passage of everyday time, the registers of the knowledge of places in which to buy and consume the daily necessities of life, the dates of the calendar, to the register in which the territory of the nation-state is represented. These registers are intricately linked to each other and are continuously shifting.  The total order of the representations and the comprehension consisting in these multiple instances is what I can call the schema “world.” We encounter various phenomena and comprehend them within the order by representing them: the world is this order of representation and comprehension.

The filter as the semi-permeable membrane and the filter as the optical lens are tropes that operate in differing registers. In the phrase “translation as a filter,” neither the national border nor state sovereignty is explicitly mentioned. Nevertheless, the instances of these tropes are integrated into the schematism of the world and operate in that world. Today, the reason that translation is seen as a transfer of information from one national language to another is precisely that, at present, we only attempt to conceive of translation within this modern schema of the world. As I briefly mentioned earlier, in Roman Jakobson’s theory of translation, translation is determined from the very outset as communication from one language to another. He classifies translation into three kinds: intralingual translation, interlingual translation, and intersemiotic translation. Intralinguistic translation - re-wording between the different fields of specialty or genres and intersemiotic translation – transcoding between different media – are treated as secondary to, or derivative of, the most authentic genre of “interlingual” translation, and the unity of a language is taken as a natural given4. Language here is conflated with national or ethnic language. Thus, for Jakobson, “proper translation” is from the very outset determined as interlingual translation. Consequently, the trope of the filter wields an awesome power in Jakobson’s discussion of translation. As long as translation is understood in terms of the model of communication, we cannot put down the analytical weapon of schematism.

The Representation of Translation and the Act of Translation

When translation is represented as the transfer of information from one language to another, it may seem that the trope of “translation as a filter” functions smoothly. This trope posits that, in translation, information is communicated  while its signs are transformed from one system of code into another. In the trope of “translation as a filter,” both the utterance prior to translation and the utterance after translation are given as ones with determinate meanings. Certainly, the instant of translation indicates a rupture between one language and another, but this gap is created by means of the figure of the filter. Here, the trope of the abyss that divides this shore from that shore seems to capture the essential moment of translation – the abyss is an image that seems to beautifully give form to this gap or rupture - but the fact remains that this shore and the other shore are within a continuous world. Something fundamental is omitted from the working of this trope. Of course, precisely because it is a trope, it neither expresses adequately nor as a whole, but we must consider that it elides the indispensable moment of translation.

What is this indispensable moment? For translation to occur, one must encounter some form of incomprehensibility or unintelligibility. One translates because one does not comprehend. However, incomprehensibility cannot be reduced to the absence of correct interpretation or the lack of proper meaning5.

When we comprehend or articulate something in the world, the following presupposition seems to ensue. Let us take up a cut between “here” and “there” or between “now” and “soon” in order to clarify what I imply by this presupposition . What is at issue here is a situation such as this: “here” signifies my house grounds and “there” is the premises of my neighbor’s house, or “now” is today, while “soon” is tomorrow. The calendar’s daily demarcation is drawn between “today” and “tomorrow.” Or for example, a boundary line between my house and my neighbor’s property is drawn in the land registry. What allows us to insert a cut is that my house and my neighbor’s house are on a continuous land, or that tomorrow follows from today. Only when there is continuity can we insert a cut. When the surface of the earth consists solely of property with coordinates, when time is arranged in the chronological order of past, present, and future, then we are within the world as a schema. Consequently a cut occurs only within a given, continuous world, and only as long as it cannot destroy or alter the continuity of the world. In other words, the continuity of the world guarantees that we can smoothly move and create cuts within this given order of comprehension. From this world of continuity, incomprehensibility is excluded. It is supposed not to exist in the world. In as much as a cut is possible in this continuous world, we are not supposed to encounter any situation that cannot be comprehended. If translation is a response to the situation in which “I do not comprehend,” then how should we rethink the relation of translation and the world that we have held onto up to this point?

Obviously, the filter is a trope for this sort of cut. It is a marker of both “comprehensibility” (wakaru koto) and “divisibility” (wakerareru koto). Yet, it is expected to indicate the locale of incomprehension! That is, it is a device whose basic feature is to represent “not comprehending ” as if it had been “comprehended.” That is to say, when we think of translation we must guard against the trope of the cut.

Thus far, I have insisted upon a rigorous distinction between the representation of translation and the act of translation because we have to be cautious about the tropics of the cut. The cut does not express discontinuity; on the contrary, the cut serves as an affirmation of continuity. Correspondingly I have tried to be exact about the idiom “translation as a filter.” It is precisely because the workings of the cut are preserved within this trope: it seals translation within “comprehensibility,” thereby eliminating its most crucial moment, which is expressed by the idiom “I do not comprehend.” In order not to confuse the act of translation with its representation, we must confront this situation of “incomprehensibility.”

The Heterolingual Address

First, we cannot overlook the fact that “incomprehensibility” is essentially a matter of sociality. “Incomprehensibility” can only occur when I co-exist with “you,” and in this the basics of sociality appear. Of course, I might immediately be rebutted by someone asking whether or not any act can occur at all outside of the scene of sociality, but for the time being, let us simply state that translation occurs in the scene of sociality. As I have discussed earlier, translation is something about citation; it cannot occur in the modality of immediacy such as “I speak” or “I write.” Translation is an enunciation, but insofar as it is a citation, it is imitative and retrospective, and it is because of this retrospective referentiality to an other past text that, in translation, I necessarily betray my own spontaneity. Furthermore, translation anticipates a differential between one who comprehends and one who does not. And it is the strange subject called  “translator” who articulates this differential between the one who comprehends and one who does not. Therefore, it is rather misleading to claim that  “comprehensibility” demonstrates social connectedness between people while “incomprehensibility” expresses the lack of social connection. For a situation of incomprehension to take place there must be a relation between people. If “comprehensibility” simply expresses a situation in which communication is accomplished, “incomprehensibility” must also be a situation in which communication occurs. “I do not comprehend” clearly expresses the situation in which the limitations of the model of communication are most explicitly revealed. Translation occurs in aleatory sociality, a social relation of the wager.

Translation does not occur in between one language and another. Rather, the image of a language as an enclosed and unitary totality is posited precisely through the representation of translation. Put another way, the figure of “translation as a filter” is what regulates the representation of translation. This is because the schematism of translation renders “incomprehensibility” into “comprehensibility” by projecting (tōsha) – or project-ing (kitō) – translation into the world. Therefore we can say this much: by representing translation as communication between one language and another, these two languages come to be represented as enclosed ‘areas.’ This is why earlier I referred to the schema that operates when we represent translation as “the schema of co-figuration.6” Thereupon, the schema of co-figuration gives rise to the institutionalized expectation that the difference of languages ought be the cause of “incomprehensibility.” However, at the locale of  “incomprehensibility, translation is attempted by various people in various ways, and we cannot always mold this locale of “incomprehensibility” to the image of the semi-permeable membrane or abyss located between spatially-represented languages. “Not comprehending” takes place everywhere – or in an “exteriority” that is not simply the obverse of the interior – yet, thanks to the configuration of the international world, it is assumed that we should be able to comprehend one another  within an identical language. Thus, we come to imagine a world in which incomprehensibility and comprehensibility are allocated to the cartographically demarcated territories of state sovereignties and the locations of national languages and cultures. In other words, the act of translation is located  in the schema of the world as the continuous totality of those who say “I comprehend.”

Here, translation holds an ambiguity. The representation of translation is a work of presenting “incomprehensibility” as “being already comprehensible,” and the act of translation also aims at turning what I do not comprehend into “something that I comprehend.” But, the dimension of turning “what I do not comprehend” into “something already comprehended” in the representation of translation is entirely different from the dimension of turning “not comprehensible” into “comprehensible” in the act of translation.

By chance, I happen to meet a stranger, but I do not understand anything she says. Perhaps I explain to myself by claiming that the reason for my incomprehensibility stems from the gap between my Japanese and an unknown that I suspect she speaks; by alluding to this gap between the two languages, I can turn my incomprehensibility into a comprehension of my incomprehensibility. However, by representing a different language as an enclosed area  from which I am no doubt excluded, I comprehend my inability to understand what she is saying as a gap between two languages. This is an idealistic resolution of “incomprehension,” I must hasten to add. However, there is another manner of interaction. I do not understand what she is saying, so I look for common terms, fragments of some colonial heritages she and I might share, and pursue the possibility of communal work through non-linguistic texts like gestures or maps. Evidently this is what children do when they meet someone they do not understand. What one attempts to gain by this method is neither the original meaning nor the correct interpretation. It is simply a way of turning incomprehensibility into a kind of comprehensibility. This attempt to say “let me try to understand you” is from the very beginning something collective, something co-eval (kyōsonzaiteki). I would not hesitate to call this approach a materialist resolution to incomprehensibility. Incomprehensibility is a matter of sociality. We learn that incomprehensibility is multilayered and involves many potential relations. Our attempt to translate reveals that we are engaged in many potential relations of incomprehensibility in which we discover ourselves.

An act of translation invites another because it takes place in sociality. Translation is repetition, and one can capture a glimpse of this essence of translation. When I talk about  translation, I also want to approach the act of translation from this dimension.

At the same time, it is not the case that the act of translation can completely exceed the dimension of its representation. To comprehend through translation is to return to the world, and to discover that the world to which we return has been changed. As I discussed at the beginning of this essay, the “heterolingual address” is a refusal of the idealistic resolution to the situation of incomprehensibility.

In Place of a Conclusion

    Since the advent of German Romanticism in the 18th century, the arguments on interpretation put forward by the Sorai school or other projects whose monuments are beyond my comprehensibility, translation has been the central institution of the Humanities, and one would not be able to understand either the formation of modern European languages or of the modern Japanese language without taking the institutionalization of translation squarely into account. Moreover, the regime of translation have always accompanied the project of nation building. For instance, it was common to argue that the ideal of democracy could be realized only in the medium of a homogeneous national language. However, this logic of imagining a society based on the presupposition of national or ethnic language, and then developing democracy within that society, no longer holds the relevance that it once enjoyed. Now the democratic subject resides not in the nation or ethnicity, but in the immigrant and the refugee, those who are heterogeneous to the assumed homogeneity of the nation. It is necessary to think of democracy not after the figure of the nation, but rather after the figure of the foreigner in us – that is, to envision a democratic society founded not on national language but on translation7.

In pursuing the tropics of “translation as a filter,” I have tried to show how the representation of translation is subject to its historical limitations. There are other issues I could have treated here by analyzing the multiple forms that translation can take, and by extension, the impoverishment of the form as a result of the formation of the nation-state. For instance, I could have devoted more space to illustrating that the unity of a national language is not given in experience but can only exist as a regulative idea. But these issues have been dealt with elsewhere. By pursuing the trope of translation as a filter with regard to the figure of the international world, I have emphasized that the representation of translation operates by reproducing the international world.

The modern international world developed around two fundamental axes. One is the continually-expanding movement of commodification, while the other is the movement of containment that entraps the multiple “incomprehensible” differences in the attempt to resolve them in the direction of the “comprehended” differences of the system of co-existing nation-states. The former is well-known as the movement of the accumulation of capital, but as to the latter, we do not yet have an adequate understanding. The nation-state nurtures and cares for the “life” of the population that resides within the territory of that state in order to manage it, but this is only one aspect of the biopolitics of nationality. It also constitutes its sovereignty by selectively expelling bodies passing through its territory beyond its national borders, or by enclosing them within. The manufacture of the national subject is always accompanied by the apparatuses of expulsion as well as integration of those who are not part of the national people. Moreover, the self-legitimation of state sovereignty is moving further and further towards the domain of security. This is precisely why people such as Zygmunt Bauman and Giorgio Agamben compare the nation-state to the “concentration camp.” The movement of capital accumulation and the movement of the classification of global humanity into nation-states operate together and are complicitous. Therefore, in order to confront globalization, we must first confront the structure of the nation-state’s manufacture of subjectivity. The question of translation cannot be evaded, precisely because a critical analysis of capitalism cannot be carried out on the presumption of the national subject8. For, the representation of translation posits the unity of a national language, and the unity of a national language has formed the inner kernel of the techniques for the manufacture of national subjectivity. It is by inventing a different way of representing translation that we can continue to seek a mode of collective being that is neither national nor ethnic.


1 Naoki Sakai, “Introduction. Writing for Multiple Audiences and the Heterolingual Address” in Translation and Subjectivity: On “Japan” and Cultural Nationalism (Minneapolis: University of Minnesota Press, 1997).

2 Roman Jakobson, “On Linguistic Aspects of Translation,” in Selected Writings, vol. 2 (The Hague & Paris, Mouton, 1971), 261.

3 Thongchai Winichakul, Siam Mapped: A History of the Geo-body of a Nation (Honolulu: University of Hawaii Press, 1994).

4 Jakobson, op. cit., 261. 

5 On Watsuji’s ethics, see Watsuji Tetsurō zenshū, vols. 10 & 11 (Tokyo: Iwanami Shoten, 1962); Watasuji Tetsuro’s Rinrigaku, Yamamoto Seisaku & Robert Carter trans. Albany, SUNY Press, 1996; A Climate, a philosophical study, Geoffrey Bownas trans. Hokueseido Press, 1971.

6 See the “Introduction” to Translation and Subjectivity, op. cit. 

7 Étienne Balibar’s “Europe: Vanishing Mediator?” in We, the People of Europe?: Reflections on Transnational Citizenship, trans. James Swenson (Princeton: Princeton University Press, 2003).

8 See Naoki Sakai and Jon Solomon, “Introduction: Addressing the Multitude of Foreigners, Echoing Foucault” in Traces (4): Translation, Biopolitics, Colonial Difference, eds. Naoki Sakai & Jon Solomon (Hong Kong: University of Hong Kong Press, 2006), 1-35.