We have concluded that mutual understanding in abnormal discourse should be approached through hermeneutic understanding or language learning. However, the real possibility of restoring mutual understanding in abnormal discourse seems to make our incommensurability thesis lose its revolutionary edge and philosophical significance. Let me explain. Since communication breakdown between two incompatible P-languages is caused by lack of mutual understanding, the restoration of mutual understanding appears to render the doctrine of communication breakdown between two P-languages unsubstantiated. To paraphrase Davidson’s mock on conceptual relativism (1984, p. 183), ‘incommensurability as communication breakdown is a heady and exotic doctrine, or would be if we could make good sense of it. The trouble is, as so often in philosophy, it is hard to improve intelligibility and tenability—by removing the extreme relativistic flavor, i.e., the failure of mutual understanding, off the thesis of incommensurability—while retaining the excitement’. Thus, we seem to face a dilemma: We can either defend the doctrine of cross-language understanding failure in order to keep the sharp edge of the thesis of incommensurability as communication breakdown, but at the same time we have to bite the bullet of extreme relativism; or we have to give up the doctrine of communication breakdown altogether. But if there were no communication breakdown per se, then there would be no incommensurability as communication breakdown either.
Fortunately, we do not have to accept the above ‘either/or offer’. We can accept the universal understandability of any languages and traditions, and, at the same time, maintain the doctrine of incommensurability as communication breakdown. This is certainly achievable so long as we realize that cross-language understanding (in the propositional sense, as comprehension) should be distinguished from cross-language communication. I will argue that cross-language understanding (in the propositional sense) is a necessary but not sufficient condition for cross-language communication.1 Although the failure of mutual understanding surely puts communication at risk as I have argued in chapter 12, the success of cross-language understanding does not guarantee successful communication. Successful cross-language communication requires much more than understanding. Even if complete communication breakdown can be overcome in terms of hermeneutic understanding or language learning, there still exist some much more interesting cases of communication breakdowns that are inevitable in abnormal discourse, namely, partial communication breakdown. It is this kind of communication breakdown per se in abnormal discourse that substantiates our doctrine of incommensurability as communication breakdown.
Hermeneutic understanding is supposed to be an alternative to avoid antithetical poles of the projective and adoptive approaches to understanding in order to achieve authentic understanding in abnormal discourse. As attractive as it is, what concerns us here is not whether Gadamer’s hermeneutic understanding could provide us with a sort of understanding in abnormal discourse, which it does, but rather whether it could restore full communication by achieving effective understanding in abnormal discourse, which it does not.
Based on our truth-value conditional account of understanding, effective understanding involved in cross-language communication is propositional understanding by nature. According to it, interpreters understand effectively the sentences of an alien P-language if and only if they can grasp the speaker’s language-meaning or its propositional contents. In normal discourse, propositional understanding can usually take place smoothly since both sides have shared or compatible M-presuppositions as a common ground to establish necessary logical relations between the propositions under discussion. In contrast, in abnormal discourse the would-be communicators do not even agree on the truth-value status of the sentences under discussion. One sentence that clearly expresses a proposition when considered within the context of one P-language may lack any propositional content when considered within the context of the other. It makes propositional understanding (effective understanding in particular) in abnormal discourse problematic and difficult (although it is still possible).
Gadamer drives home the point that propositional understanding is doomed to failure in abnormal discourse in the following passage:
There are no propositions which can be understood exclusively with respect to the content that they present, if one wants to understand them in their truth. ... Every proposition has presuppositions that it does not express. Only those who think with these presuppositions can really assess the truth of a proposition. I maintain, then, that the ultimate logical form of the presuppositions that motivate every proposition is the question. (Gadamer, 1986, p. 226; first italics are my own)
Put into our terminology: Gadamer is actually arguing that proper propositional understanding of an alien language depends on the grasp of the M-presuppositions of the language. In abnormal discourse, if interpreters cannot comprehend the M-presuppositions of an alien language (‘do not think with these presuppositions’), they cannot grasp the propositions expressed by the sentences of the language. Consequently, the propositional understanding of the alien fails. The best alternative in this situation is to work out the M-presuppositions of the alien language by engaging in a dialogue to ask questions, to hypothesize the alien way of thinking, and to make comparison between their own language and the alien’s. To do this, interpreters inevitably become involved in the hermeneutic circle between their own language and the alien’s. It is when propositional understanding in abnormal discourse fails that hermeneutic understanding takes over. This is where Gadamer’s notion of hermeneutic understanding comes into play.
However, to say that propositional understanding begins where hermeneutic understanding ends seems to give the reader an impression that hermeneutic understanding is a continuation of propositional understanding. This is not the case. Here we are actually dealing with two very different kinds of conceptions of understanding. Heidegger’s and Gadamer’s notion of hermeneutic understanding is not intended to be a solution to the traditional epistemological problem of understanding, i.e., propositional understanding. Instead, it is supposed to offer an alternative revolutionary notion of understanding. In fact, Heidegger’s and Gadamer’s notions of hermeneutic understanding arise as a rebellion against the traditional notion of propositional understanding. The notion of propositional understanding presupposes the monologue model of meaning. According to the model, a sentence possesses a self-sufficient unit of linguistic meaning or a self-contained propositional content that is given by the author and is determined the conventions of his or her language, and thus has nothing to do with the interpreter’s involvement (the interpreter’s language, tradition, or horizon). It is, therefore, monological.
Heidegger considers propositional understanding to be secondary and derivative from universal hermeneutic understanding. For Heidegger, to understand is to ‘be at home with something’. In this sense to understand something is to master it. This is actually what we usually call practical understanding or ‘to understand how’ in contrast with propositional understanding, i.e., ‘to understand what’. Understanding in this sense means less a ‘kind of knowledge’ than ‘knowing one’s way around’ (Heidegger, 1975, p. 286).
Like Heidegger, Gadamer considers the ‘construction of logic on the basis of the proposition’ to be one of the ‘most fateful decisions of Western culture’ (1986, p. 195). For Gadamer, ‘Language is most itself not in propositions but in dialogue’ (1985, p. 98). Against the primacy of the notion of propositional understanding, which assumes a self-contained propositional content for each sentence, the hermeneutic approach adopts the dialogue model of meaning. It reminds us that the literal meaning or the propositional content of a sentence can never be prescinded from the context in which it is embedded when it is understood, and therefore can never be detached from the process of understanding itself. There is no fixed, self-closed propositional content for each sentence that is waiting to be discovered by the interpreter. Understanding of what a sentence is saying thus becomes an activity of participation, engagement, assimilation, and dialogue, that is, the interpreter’s participation in the reformation and enrichment of its meaning, the engagement between the speaker’s and the interpreter’s traditions or the fusion of horizons, assimilating what is said to the point that it becomes the interpreter’s own, and ultimately a dialogue between the interpreter and the text. ‘For language is by nature the language of conversation; it fully realizes itself only in the process of coming to an understanding’ (Gadamer, 1989, p. 446). In a dialogue, there are no fixed propositions, only questions and answers, which in turn call forth new questions.
In general, the process of hermeneutic understanding well involves a complex process of reinterpretative reconstruction on the part of the interpreter, which is not just purely reproductive, but also a productive activity. The advocate of propositional understanding may suspect that the so-called hermeneutic understanding is actually what we usually call the process of interpretation. Just as to know an object, to understand a text is a monological or unilateral process in which the subjectivity of the interpreter (the fore-meanings and prejudices coming with his or her own tradition and language) is not supposed to be part of understanding. This is why we always strive for a value-free, objective understanding (at least in scientific understanding). The goal is to attain some finally adequate understanding so that we can obtain a full intellectual control over what is to be understood. But interpretation is supposed to be a different process from understanding. The former is less stringent compared with the latter. To interpret is to see things through the eyes of the interpreter. It is a bilateral and dialogical process in which the interpreter comes to an interpretation through the interplay between his or her own fore-knowledge and what is to be interpreted. ‘[A]n interpreter’s task is not simply to repeat what one of the partners says in the discussion he is translating, but to express what is said in the way that seems most appropriate to him, considering the real situation of the dialogue’ (Gadamer, 1989, p. 308; my italics). Therefore, ‘to interpret means precisely to bring one’s own preconceptions into play so that the text’s meaning can really be made to speak for us (Gadamer, 1989, p. 397). Since both the fore-knowledge of the interpreter and the context in which an interpretation takes place can change, we can never have a final interpretation. Clearly, the advocate of propositional understanding continues, Gadamer confuses understanding with interpretation.
To this charge, Gadamer accepts its conclusion but rejects its premise: The critic is right on the point that hermeneutic understanding is interpretation, more precisely, interpretative understanding; but the critic is wrong to separate understanding from interpretation as two independent processes. In the early tradition of hermeneutics, which seeks to draw a rigorous distinction between understanding (subtilitas intelligendi) and interpretation (subtilitas explicandi), there is first a pure, objective understanding of the meaning existing in the text, which is free from all prejudices and not ‘contaminated’ by ‘subjective’ interpretation. Propositional understanding is supposed to be such an objective understanding. Interpretation functions as a means to the end of or a supplement to understanding. In bold new contrast to this tradition, Heidegger and Gadamer argue that there are no essential differences between understanding and interpretation.
Interpretation is not an occasional, post facto supplement to understanding; rather, understanding is always interpretation, and hence interpretation is the explicit form of understanding. In accordance with this insight, interpretative language and concepts were recognized as belonging to the inner structure of understanding. (Gadamer, 1989, p. 307)
Understanding and interpretation are internally related or fused, both of which compose one unified process. Every act of understanding involves interpretation, and every act of interpretation involves understanding. Understanding is not just reproductive but, because it involves interpretation, is also a productive process. Understanding always involves something like the application of the text to be understood to the interpreter’s present situation. Since we are always understanding in light of our anticipatory prejudgments determined by the effective historical situation in which we live, we always take ourselves along whenever we understand. Meaning is always coming into being through the ‘happening’ of understanding. ‘The interpretative process of understanding ... is simply the concretion of the meaning itself (Gadamer, 1989, p. 397; italics as original). C. Geertz describes the indeterminacy of hermeneutic understanding vividly as follows:
Understanding the form and pressure of, to use the dangerous word one more time, natives’ inner lives is more like grasping a proverb, catching an allusion, seeing a joke—or, as I have suggested, reading a poem—than it is like achieving communion. (Geertz, 1979, p. 241)
Therefore, there is always room for different understandings or interpretations. ‘There cannot, therefore, be any single interpretation that is correct “in itself. ... Every interpretation has to adapt itself to the hermeneutical situation to which it belongs’ (Gadamer, 1989, p. 397). Hermeneutic understanding is interpretative understanding.
To sum up, although the would-be communicators in abnormal discourse can still understand one another to a certain extent by utilizing the process of hermeneutic understanding, effective understanding (in the sense of propositional understanding) is inevitably partial. When effective understanding in abnormal discourse is incomplete, cross-language communication between two P-language communities is inevitably also partial.
‘Hold on’, a critic may object, ‘your argument starts with a very suspicious premise that understanding used in cross-language communication has to be propositional, which you call effective understanding. Why does it have to be so? Why could we not have a different notion of communication that is itself hermeneutic by nature and thereby does not presuppose propositional understanding at all?’ I have to admit that our imaginary critic has raised a very good point. Indeed, whether propositional understanding should be an essential part, a necessary condition, of cross-language communication depends upon the notion of cross-language communication to which we subscribe. In fact, as we will clarify shortly, there are at least two competing models of communication: the standard transmission model, and the hermeneutic dialogue model. The transmission model of communication, which is propositional by nature, does presuppose propositional understanding while the dialogue model does not. Considering my specific purpose, there is no need for us to engage in a touchy debate as to which model of communication is better (although I personally favor the dialogue model). My strategy is to show that both models lead to the same conclusion: Partial communication breakdown is inevitable in abnormal discourse even if understanding (especially propositional understanding as comprehension) is no longer an obstacle. For the sake of argument, I will assume that either full effective understanding or (partial) hermeneutic understanding is in place, and see whether full communication can be established between two incompatible P-languages in abnormal discourse.
Before we tackle the two communication models directly, let us first consider what a bilingual would face in the context of abnormal discourse. By definition, a bilingual of two P-languages should master each language equally as well as any competent native speaker does. A bilingual should be able to understand, speak, use each language, as well as to think, reason, and act in terms of each language. Obviously, understanding is no longer a problem for a bilingual in cross-language communication. We do not have to debate whether such an understanding is propositional or hermeneutic either. A bilingual of two P-languages can achieve sufficient understanding of both languages and thus overcome complete communication breakdown between the language communities in abnormal discourse. Nevertheless, the question remains: Is successful full communication possible?
One may argue that although it is difficult to put oneself into the other’s skin, it does not mean that it is impossible. A member from one P-language community can become a member of another P-language community by immersing oneself in the community and living their way of life. With tremendous effort and sufficient time, one could eventually become a bilingual of both P-languages. For example, our bilingual Ms Jones can become a bi-medical ending up talking both medical languages.
Unfortunately, the life of our bilingual who walks on the thin line between two incompatible P-language communities is not a ‘happy’ one. A bi-lingual—who has mastered two P-languages separately—is not a meta-lingual—who can speak a meta-language or common language with the two P-languages as its sublanguages. Lack of a common measure, especially shared M-presuppositions, between two substantially different P-languages (such as the Chinese medical language and the Western medical language) often puts our bilingual (Ms Jones) who dwells on the boundary between the two language communities in an awkward situation. It is impossible for Ms Jones to commit herself fully to the two incompatible P-languages since one can neither be incorporated into the other, nor can they coexist peacefully. She cannot think in terms of both P-languages at the same time. She might not be able to live the pre-modern Chinese way of thinking. If she could, then her present Western way of thinking would not survive. In any case, at best Ms Jones can start to think and talk in terms of the pre-modern Chinese way of thinking only when she becomes alienated or dissociated from the Western medical language. By doing so, she will eventually drop out of the language community of Western medical theory. At some point, our bilingual eventually has to decide to which language she wants to commit herself unless she can ‘happily’ live such a ‘double life’. Nevertheless, the decision is not going to be an easy one since she cannot hold both P-languages side-by-side in mind to make a point-to-point rational comparison.
One might ask, why does our bilingual have to commit herself to one of the two competing languages? Could she suspend her judgment and hold onto both languages without committing to either? This is fine if Ms Jones does not need to communicate with the other side. The trouble emerges when she attempts to communicate the ideas of the Western medical language to the Chinese, or vice versa. These ideas are intelligible for the Chinese only if the underlying M-presuppositions are comprehensible to them. This requires that Ms Jones emerge from the way of thinking embedded in the Western medical language and interpret the ideas in the way that the Chinese can understand. Therefore, a bilingual like Ms Jones who inhabits a certain type of frontier situation between two rival P-languages always faces a choice between immerging and emerging: to immerge into an adopted alien language to effectively understand it; to emerge from one’s own native language to successfully communicate with the speaker of the alien language. Such frequent switches between immerging into and emerging from a P-language very often put our bilingual in a predicament if she is confused about in which language community the discussion is occurring. The use of one mode of reasoning or lexical taxonomy to make assertions to the speaker who uses the other incompatible one makes understanding problematic, and thus places communication at risk. For example, as Kuhn points out, one cannot speak an alien language (such as the language of Aristotelian dynamics) while using the projectible kind-terms of the other competing language (such as the Newtonian notion of ‘motion’). Bilinguals are forced to remind themselves at all times which language is in play and within which language community the discourse is occurring to avoid improper use of a kind-term of one language in the other language community. The inability to use one way of thinking embodied in one language to understand the other language makes full communication between two substantially different languages problematic.
Of course, it is possible for our bilingual to understand the Chinese medical language but not commit herself to it. She might identify and comprehend very well the pre-modern Chinese way of thinking and corresponding categorical framework while rejecting them as either illegitimate or unsuitable. In this case, she actually becomes a spectator, not a participant, therefore not an engaged conversation partner and communicator. We have more to say about this when we discuss Gadamer’s conversation model of communication later.
More significantly, let us put those worries about bilinguals surviving on the boundary between two incompatible P-language communities aside and turn our attention to cross-language communication. The achievement of a bilingual is understanding; but understanding is only part of communication. The real issue is this: Even if our bilingual can understand perfectly an alien P-language, it does not mean that she can communicate successfully with its speakers. This is because cross-language understanding between two P-languages itself does not guarantee successful communication between the language communities. Cross-language understanding is only necessary, but not sufficient for cross-language communication.
For our bilingual (Ms Jones) to be able to communicate with another P-language community (such as the Chinese medical community), she needs first to master both languages (the Chinese and Western medical languages), that is, to be able to understand and use effectively each language involved. But the fact that our bilingual can understand an alien language does not automatically make her an engaged communicator with the speaker of the alien language. Communication is not assimilation. A constructive dialogue between two distinct P-languages requires that each side commit to their own language and manage to communicate their thoughts to the other side. It is not an easy task as Gadamer has realized:
It is well known that nothing is more difficult than a dialogue in two different languages in which one person speaks one and the other person the other, each understanding the other’s language but not speaking it. As if impelled by a higher force, one of the languages always tries to establish itself over the other as the medium of understanding. (Gadamer, 1989, p. 384)
To be an engaged communicator, in addition to understanding the other, our bilingual also has to make herself understood well by the other or to convey effectively the information from one language (the Western medical language) to the other,2 to respond properly to the other’s requests, and to engage in constructive dialogue with the other. It is exactly this aspect of cross-language communication that the bilingual of two incompatible P-languages cannot fulfill.
For an illustration, consider the following situation faced by our bilingual, Ms Jones, and our old friend, the Chinese physician Mr Wong. Suppose that both of them work together in a clinic, and both speak Chinese (or English if you wish) as the working language. Suppose that Mr Wong, like our Ms Jones, knows both medical theories inside out. For some unknown reason, Mr Wong practices the Chinese medicine only and Ms Jones only practices the Western medicine. One day, a patient by the name of Jennifer, who is a friend of both physicians, comes to visit the two physicians separately for her painful spleen. Mr Wong diagnoses her illness as an excess of the yin within her spleen (an asthenic spleen). According to Ms Jones, however, Jennifer’s illness is caused by a bacterial infection. Jennifer, perplexed with two very different diagnoses, asks the two physicians to meet and explain the difference between them to her. As a friend, she requests that each physician convince her of the correctness of their own diagnosis by presenting arguments to the other. You can imagine what would happen under such a situation. Although both physicians understand the other’s diagnosis well, it is hard for them to become engaged communicators. They cannot engage in a productive conversation, not because of misunderstanding, but because there is no common ground for them to engage in a constructive conversation and argumentation. There is no way to match what Mr Wong wants to say against what Ms Jones wants to say. Each side cannot make a point-to-point comparison with the other side. Neither can even deny directly the correctness of the other’s diagnosis by claiming that it is false. In fact, the question of whether the counterpart’s diagnosis is true or false simply does not arise. Although both can understand one another very well, the communication between them is inevitably partial.
The point of the above thought experiment can be driven home when we compare the previous situation with the following one. Suppose that Jennifer visits two Chinese physicians. One physician diagnoses her illness as an asthenic spleen while the other diagnoses it as a sthenic spleen (the excess of the yang within her spleen). When asked to present their own case, each physician can effectively present his or her argument based on the evidence shared by both (such as symptoms of abdominal distention, loose stool, inappetence, phlegm-retention, oedema, diarrhea, blood in stool, and so on). In this case, both sides can not only understand one another well, but also engage in successful communication by exchanging and comparing their thoughts as well as engaging in constructive rational argumentation.
What we learn from the above case is not only that cross-language understanding is only necessary, not sufficient, for cross-language communication in abnormal discourse; but also that cross-language understanding is actually necessary for our realization of partial cross-language communication breakdown. Communication breakdown between two language communities with incompatible M-presuppositions arises not just from one side’s inability to understand the other side (a complete communication breakdown), but also precisely from those discourses in which one side is able to understand and see how different the other’s beliefs, categorical framework, and mode of reasoning are from their own (a partial communication breakdown). Ms Jones and Mr Wong cannot engage in a constructive dialogue with one another, not because the other’s beliefs appear bizarre, but because each can understand how the other is tied to a form of life and intellectual tradition manifested as the M-presuppositions of the language to which they are so committed themselves. Ironically, it is precisely our ability to understand the other language that reveals the inability to communicate successfully with the other. In this sense, a bilingual can only make one aware of and appreciate the occurrence of partial communication breakdowns in abnormal discourse, but cannot provide a solution.
We have contented ourselves so far with a vague and common sense notion of cross-language communication. To distinguish cross-language communication from cross-language understanding and to argue for the thesis of inevitability of communication breakdown in abnormal discourse, the time has come for us to be more specific about the notion of cross-language communication.
Although humans were anciently dubbed both the ‘rational animal’ and the ‘speaking animal’ by Aristotle, only since the late nineteenth century, especially after the linguistic turn in the early twentieth century, have we realized that it is our ability of using language to communicate with ourselves and others, more than our ability to reason (we cannot reason without language), that defines who we are. We might simply define humans as the linguistic communicator. Indeed, there is nothing people do more often than communicate with one another by language. However, the act of linguistic communication is easily accomplished, but not so easily explained conceptually.
As other notions hailed as unmixed goods, the notion of communication suffers from the misfortune of conceptual confusion. Through tracing the historical development of the idea of communication, J. Peters (1999, pp. 6-10) finds that there are at least four clusters of meanings associated with the notion of communication, (a) The word ‘communication’ originated from Latin word ‘communicare’, meaning to impart, share, or to make common. Accordingly, one dominant cluster of meaning in ‘communication’ has to do with imparting, a unilateral process quite apart from the bilateral process of exchange, dialogue, interaction. It could mean ‘partaking’ (partaking of Holy Communion), an act of receiving, not of sending; it could mean a message or a notice to a certain audience, an act of sending, not receiving or exchanging; or it could mean physical connection or linkage. We can call this the communion model of communication. (b) Another dominant cluster of meanings involves transmission of physical objects, such as heat, light, electronic signals, or of information, which could be psychological entities, such as ideas, thoughts, or speaker’s meanings, or linguistic entities, such as propositional contents, literal meanings. As we will see later, like the communion model, this transmission model is not necessarily a bilateral process of exchange and interaction between the sender and the receiver. The transmission process could be one-way, from the sender to the receiver without any feedback from the receiver, (c) If we fold up the linear transmission model to connect its one end with the other so as to make it a two-way, bilateral process of exchange between the sender and the receiver, we have the exchange model of communication. It is supposed to involve interchange, mutuality, reciprocity, and engagement, such as exchange of ideas in dialogue, psychosemantic sharing, even fusion of consciousness, (d) Last, ‘communication’ can mean something very general, as a blanket term for various modes of communication identified above. Very roughly, any attempt at manifestly having a specific impact on somebody else’s cognitive attitudes may be called an act of communication.
The above four models of communication cover linguistic and non-linguistic communication. Linguistic communication stands out from any other forms of non-linguistic communication in that the vehicles or mediums employed in communication belong to a system of semantically interdependent meaning items, which are conventionally established, such as a natural language. Although there is no consensus on what linguistic communication is, especially, what counts as an engaged successful linguistic communication, we can still identify the two most popular philosophical models of linguistic communication directly pertaining to cross-language communication. One is the transmission model, which is the standard model of linguistic communication; and the other is what I shall call the dialogue model, roughly corresponding to the exchange model identified above. We will focus on the transmission model in this chapter and discuss the dialogical model in the next chapter.
The transmission view of linguistic communication is the most prevalent conception of communication in our culture, and perhaps in all industrialized cultures (J. Carey, 1992). The philosophical foundation of the transmission model was established by John Locke’s empirical epistemology. It is John Locke, more than anyone else, who provides articulate defense of the two doctrines foundational to the classical transmission model: the private mind filled with ideas and linguistic signs as empty vessel to be filled with ideational content. Locke treats the meaning of words as a sort of private property in the individual’s interior, that is, the internal stream of ideas derived either from reflection or from sensations which are neither social nor linguistic. For Locke, language does not play any of the significant roles that many philosophers attribute to language, neither as a source of knowledge, nor as a shaper of thinking (B. Whorf), nor as the way we articulate our being in the world (Heidegger), not even as a mode of performing actions (T. Austin); it is rather ‘the great instrument’ that makes the inner life of ideas publicly accessible and a means of transporting ideas from one speaker to another. Words are at best conventions; they refer to meanings inside people’s minds and to objects in the world. When we communicate with others, we trust our private ideas to public symbol proxies by virtue of encoding them as linguistic signs. An act of communication is successful if the hearer can replicate the speaker’s ideas without distortion in terms of decoding the linguistic signs. ‘When a man speaks to another, it is that he may be understood; and the end of speech is, that those sounds, as marks, may make known his ideas to the hearer’ (Locke, 1996, 3.2.2).
Locke’s notion of private ideas and his commitment to the individual as sovereign in meaning-making make linguistic communication both necessary and impossible. It is fundamentally impossible to communicate accurately the ideas in the mind of one person to stimulate the same ideas in the mind of another, unless we can read each other’s minds. Therefore, for Locke, all communication, both intra- and inter-language communication, is inherently imperfect. Communication breakdowns loom large in Locke’s scenario.
Could we remove the mentalistic elements of the Lockean classical transmission model (meanings as private ideas or mental images given by individuals), but preserve its basic spirit—briefly, linguistic communication as the process of transferring messages from one speaker to another by means of language? The Lockean classical transmission model hinges on the premise that communication is a process whereby an idea, as a private mental entity, transfers from the mind of the speaker to the mind of the receiver. If we substitute a neutral term ‘message’ for Lockean ‘ideas’ as what is to be transferred and accordingly de-emphasize the role of the private mind of the speaker/receiver in the transmission process, we can identify four basic components involved in the process of linguistic communication: message, speaker, receiver, and linguistic expression.3 Like our treatment of cross-language (propositional) understanding, ‘the speaker’ still refers to the representative speaker of a language community; the same is true for ‘the receiver’. By ‘the linguistic expressions’, I mean declarative sentences of a P-language, a system of semantically interdependent meaning items that can be conventionally established by the language. Especially, I will identify ‘message’ with thoughts, propositional contents, or literal meanings expressed by declarative sentences. To repeat what we have emphasized in chapter 12, I treat the notion of thought here as extensional: A thought is not the intensional content of the mind, such as Lockean ideas or mental images, or the content of beliefs and other propositional attitudes, which characteristically differ from one individual to another, but is rather taken to be what is expressed by a declarative sentence and corresponds to the propositional content asserted by the sentence. What distinguishes (objective) thoughts from (purely subjective) mental contents is that thoughts can be, or are at least capable of being, true or false while mental contents, such as Lockean ideas, cannot. As such, thoughts are neither inside the head nor outside language. They are publicly accessible to all the competent speakers of a language and are communicable among them.
In contrast, it is common among many to treat linguistic communication as the branch of the use of language that focuses on the speaker-meaning, the intention of an individual speaker, conversational maxims, propositional attitudes, and the illocutionary force the speaker expresses by saying something (a speech act). For our limited purpose, I will not burden us with those pragmatic aspects, but only focus on the semantic aspect of linguistic communication, which I will call informative communication:
Cross-language (informative) communication is essentially a process of the transmission of thoughts (the literal meanings or the propositional contents) from the speaker of one language to the interpreter of the other language.
This is actually the standard model of linguistic communication adopted by most analytical philosophers since the linguistic turn.
Let us see what counts as successful communication between two disparate P-language communities in abnormal discourse based on the standard model. There are two sides to any act of linguistic communication, the speaker’s side and the interpreter’s side. Ideally, in a successful cross-language communication, the two sides in discourse understand a sentence from the speaker’s language in the same way: The interpreter understands precisely what the speaker means. As before, I call the literal meaning of a sentence in the speaker’s language ‘the speaker’s language-meaning’ and the meaning that the interpreter ascribes to the same sentence ‘the interpreter’s language-meaning’. In an ideal situation, the interpreter’s language-meaning and the speaker’s language-meaning should be the same; in other words, the propositional content expressed by the speaker is the same as that taken in by the interpreter. If such an agreement is achieved, then we say that the act of linguistic communication between two P-language communities is successful; otherwise, we say that the communication is defective or breaks down.
Corresponding to the two sides of the act of linguistic communication, i.e., the speaker’s and the interpreter’s sides, we can identify conceptually two directions or two parts of the act of cross-language (informative) communication. On the one hand, the speaker of one P-language expresses a particular thought (or a propositional content) to the interpreter of another P-language. Such a speech act is performed whenever a thought is put forward to the interpreter in terms of a declarative sentence. On the other hand, the interpreter apprehends or takes in the thought put forward by the speaker. We can call the act of putting a thought to the interpreter ‘the act of communicating’ (in the narrow sense) and the act of apprehending a thought put forward ‘the act of understanding’.
Clearly, full cross-language (informative) communication requires successful cross-language (propositional) understanding since understanding is a precondition of transferring thoughts. The failure of cross-language understanding would lead to complete cross-language communication breakdown as I have argued in chapter 12. However, the success of cross-language understanding does not guarantee the success of full cross-language communication. One (such as our Ms Jones) can understand a P-language (the Chinese medical language), but may not be able to communicate effectively to its speakers (such as our Mr Wong). This is what we shall turn to next.
It would be an arduous task (if it is possible at all) to specify a complete list of sufficient conditions under which a successful cross-language communication is fulfilled. Fortunately, since we are only concerned with cross-language communication breakdown in the case of incommensurability, we only need to identify one significant language-dependent necessary condition for successful cross-language communication, namely, a common measure shared by two different P-languages in order for the communication between them to be successful.
According to the standard model, cross-language communication is the transmission of thoughts from the speaker of one P-language to the interpreter of another P-language in terms of declarative sentences. More precisely, we have identified the two closely related acts in the process of thought-transmission: the act of communicating and the act of understanding. For the communication to be successful, the thoughts apprehended by the interpreter have to be the same thoughts put forward by the speaker. Although it may be controversial as to what counts as the same thought (propositional contents) transferred and received, the truth-value of the thought transferred should, at least, be preserved during the transmission. That means that the interpreter should at least get the truth-value of the thought put forward by the speaker right in order to receive the same thought transferred. In particular, in formal communication such as scientific communication, we obviously have an interest in not acquiring false thoughts or propositions. As such, the primary goal of thought-transmission is, at least, to preserve the truth-value—or more stringently, to preserve the truth—of what is transferred. Therefore, truth-value-preserving thought transmission should be a necessary condition of successful cross-language communication.
Let me be more specific about the thought transmission process in cross-language communication with an eye on what is required to preserve truth-values. There is a speaker S of a P-language PLS and an interpreter E from another P-language PLE. Suppose that S wants to communicate to E a particular message— e.g., a thought T or a proposition p. To convey p, the speaker searches in PLS for a sentence that is true if and only if p. Eventually, she finds a sentence A of PLS that fits this truth condition. She utters A for E to hear or writes down A as a text for E to read, and hopes for the best. Of course, she cannot just hope that the interpreter can catch what she said. For her speech act to qualify as a successful act of communicating, the speaker is responsible to prepare the way of smooth transmission of her thoughts. More precisely, she has to express p in such a way that the interpreter is able to recognize that, first, she has asserted something to E when uttering A (the condition i); and second, what she asserts is p (the condition ii). To ensure the condition i (E could recognize that S has asserted something to E), the speaker needs to make sure that she has said something that is true or false to the interpreter. However, she cannot simply assume that the interpreter could somehow conceptually grasp its truth-value (i.e., to assume that the interpreter is a bilingual). Thus, she needs to make sure that sentence A is not only actually true or false (has an actual truth-value) from the perspective of her own language PLS, but also true or false when A is considered within the context of the interpreter’s language PLE. In addition, to ensure the condition ii (the interpreter could understand what is asserted by A), the speaker has to make sure that the interpreter is able to establish the logical connection that A is true if and only if p, which in turn requires that the interpreter can recognize that A is actually true when considered within his own language PLE. The fulfillment of both conditions i and ii requires sentence A to have an actual truth-value for both the speaker and the interpreter. Furthermore, based on our analysis of the concept of truth-value status in chapter 9, the interpreter could recognize that sentence A of PLS —which is actually true or false when considered within PLS —as actually true or false also when considered within PLE only if he accepts or could accept—not just recognize in the case of understanding—the M-presuppositions of PLS. This is possible only if the M-presuppositions of PLS and PLE are shared and compatible.
I thus conclude that shared or compatible M-presuppositions between two P-languages are necessary for carrying out a successful communication between the two distinct P-languages. In other words, shared or compatible M-presuppositions are the significant necessary common measure (the pre-established semantic harmony between the two languages) for successful cross-language (informative) communication.
This explains the awkward situation faced by our bilingual Ms Jones and Mr Wong since mutual understanding is necessary but not sufficient for successful cross-language communication. To understand a thought T put forward by the speaker S of an alien P-language PLS, the interpreter E from another P-language PLE has to put himself in a position to decipher and acquire T as S intended to be understood. According to the truth-value conditional theory of understanding, to comprehend T, the interpreter does not have to accept the M-presuppositions of PLS. Instead, E only needs to be able to identify and comprehend them such that sentence A has a conceptually possible truth-value for him. This can be done even if PLS and PLE are incompatible, either by virtue of language learning or the hermeneutic interpretative understanding. But in order to communicate successfully with one another, we have concluded that the two languages involved have to be compatible. Therefore, the cross-language (informative) communication in abnormal discourses is inevitably partial.
One may be considering that although informative communication cannot be reduced to one-way propositional understanding, either on the interpreter’s part or on the speaker’s part, it could very well amount to mutual understanding. There is, in fact, a strong tendency in the related literature of confusing communication with mutual understanding. For many analytical philosophers (such as Davidson, Dummett, and others), cross-language communication is nothing but the understanding back-and-forth between the speakers from two different languages. Insofar as we are interested in communication, they believe, we are interested in mutual understanding. Such a reading of communication as mutual understanding is logically implied and widely promoted by the transmission model of communication. The goal of communication, according to the model, is simply transmission of thoughts from one side to the other. The purpose of transmitting thoughts from the speaker to the interpreter is to have the interpreter understand them. As long as the interpreter comprehends the thoughts transmitted, the goal of communication is obtained. Hence, the act of communicating could be reduced to the act of understanding; the former is only the means to the latter. Simply reversing the above process from the interpreter to the speaker, we have the act of mutual understanding. Therefore, communication equals mutual understanding.
However, as I have argued, what is involved in the act of communication is more than just mutual understanding. Mutual understanding could be achieved by two bilinguals, one on each side, such as the exchange between Ms Jones and Mr Wong. However, the fact that our bilinguals can understand one another’s languages does not necessarily make them engaged communicators. As will become clear when we discuss the dialogue model of communication, the goal of communication is not merely mutual understanding in the propositional sense—simply passing on information to one side and taking in the information passed on from the other side.
To think of the act of cross-language (informative) communication as the act of transmission of information and further as mutual understanding actually reduces cross-language communication into cross-language translation. To illustrate, we can reformulate the process of thought transmission on the speaker’s part by substituting ‘translation’ for ‘transmission’: To transfer a proposition p from the speaker’s language to the interpreter’s, the speaker first finds a sentence A from his or her own language which is true if and only if p. Second, the speaker translates A into a counterpart sentence B of the interpreter’s language (if there is any) such that B is true if and only if p. However, it could be argued that the last step of translation cannot be carried out between two radically distinct languages since the interpreter’s language does not have the necessary semantic resources (due to meaning change, reference difference, unmatchable lexical structures, or whatever other reasons) to express A. Cross-language communication breakdown turns out to be translation failure. I think the reader should have recognized by now that this reading of the transmission model brings us back to our old foe, i.e., the thesis of incommensurability as untranslatability. This is, I think, a major reason why the received interpretation of incommensurability as untranslatability sounds so natural for many, is so widespread, and is so hard to dispose of, since the interpretation is sanctioned and conditioned by the transmission model of communication.
These and other limitations of the transmission model can be attributed to its semantic foundation, namely, the monologue model of meaning. The standard transmission model of cross-language communication is propositional in essence. The model reduces the act of communication, which is supposed to be an ‘alive’, interactive, and dialectic process, to a ‘dead’, static, and monological propositional understanding: there is a fixed, self-sufficient thought in the mind of the speaker; the speaker conveys the thought to the interpreter in terms of a sentence. The sentence is self-closed and its meaning is self-contained, independent of the interpreter, simply there to be discovered. For the communication to be successful, the interpreter has to recapture the author’s intention and to understand the sentence as the speaker intends and expects it to be understood.
To think beyond understanding and translation, we should ask ourselves a question: What do we want to get from communication, especially from cross-language communication? To understand each other, of course, one may answer. But why do we care whether or not we understand one another? What is the purpose of understanding anyway? Or, more precisely: What does understanding one another enable us to do? The answer cannot be that we want to understand simply for the sake of understanding. We have at least as much interest in learning from one another and coordinating our actions in a social setting through understanding. For those and other purposes, we need to hammer out disagreement and reach consensus. ‘In a certain sense the question of commensurability and incommensurability with respect to competing paradigms or research traditions is the question of the possibility of objectively resolving the differences between them’ (Boyd, 2001, p. 56). To do so, only passively understanding one another by passing on information is not good enough; it requires critical engagement between two sides: to respond effectively to the other side’s requests, to exchange ideas effectively, and to engage in constructive dialogue and argumentation with one another. Unfortunately, this crucial aspect of genuine communication is missing from the standard model of communication.
We need to go beyond the transmission model.
1 This rather simplified statement, although it is sufficient for the time being, will be qualified later when we discuss further Gadamer’s hermeneutic understanding. As we will see, for Gadamer, understanding is an open-ended process of a speaker coming-to-an-understanding and reaching-an-agreement with someone or a text about something. Such an understanding can only be achieved through conversation, which is in essence a process of communication. Therefore, no sharp distinction could be made between understanding and communication. Habermas holds a similar view. In a sense, for both Gadamer and Habermas, understanding is communication.
2 Suppose that the other side is not a bilingual, which is true in most cases of cross-language communication. But it will not affect our analysis even if both sides are bilinguals as will be illustrated in the following thought experiment.
3 By distinguishing medium from message, I do not intend to take the message, such as thoughts or propositions, to be communicated by words or a sentence (the medium) as naked and fully determined prior to any linguistic expression. The full meaning of a message is determined in the process of being expressed by some linguistic expressions. Thoughts cannot exist antecedently and independently of the vehicle of expressing them, namely, linguistic expressions. The process for a sentence to express a thought is the process of showing what the thought to be expressed is.