Contrary to what Truth and Method might imply, Gadamer’s masterpiece turns out not to be about the relationship between truth and method at all. Gadamer explicitly denies that his hermeneutics, which is ontological and universal by nature, is not simply a method of understanding or interpretation. To Gadamer, the practice of understanding and interpretation is a mundane, common, ordinary feature of language that cannot be captured by methodical rules or strictures as with a scientific method. It would be a mistake to suppose that Gadamer is in the business of providing concrete rules for the carrying out hermeneutic inquiry. Nevertheless, we can see that Gadamer does offer us a very effective methodology of understanding others, especially those initially totally alien to us: to ‘reach a substantive agreement’ through a dialectical interplay between the horizon (tradition, prejudices) of the interpreter and that of the interpreted (the hermeneutic circle) that leads to a partial fusion of horizons. Such a ‘method’ of interpretation is in essence ‘the art of conversation—argument, question and answer, objection and refutation’.
However, to my dismay, we do not see that truth plays any significant role at all in his hermeneutic understanding and communication. Gadamer has explicitly denied that it was his intention to play off truth against method. But this does not mean that truth should not play any significant role in his hermeneutic method. After all, ‘a primary intention of Truth and Method is to elucidate and defend the legitimacy of speaking of the “truth” of works of arts, texts, and tradition’ (Bernstein, 1983, p. 151). For Gadamer, to understand something is to grasp its ‘true meaning’, to let its true meaning speak to us, or more precisely, to allow the ‘claim to truth’, which ‘makes upon us’, to reveal itself. Clearly, truth is essential for his hermeneutics. Besides, according to Gadamer, one can understand and communicate effectively on some subject matter with someone only by engaging in an authentic conversation or dialogue with that person, and an authentic conversation is the process of question and answer, argument and counterargument, objection and rebuttal. Some notions of truth and truth-value are indispensable for both sides to engage in argumentative dialogue. Without a proper understanding of the role of truth in argumentative dialogue, Gadamer’s conversation model of communication is substantially incomplete.
It is notorious that Gadamer’s notion of truth is elusive except that we are sure that he has no taste for the traditional correspondence notion of truth and for propositional truth. For Gadamer, truth cannot be exhausted by scientific method alone. It is available to us through hermeneutic method. R. Bernstein suggests that Gadamer is in fact appealing to a concept of discursive notion of truth—a truth is whatever is warranted by appropriate forms of argumentation. But what distinguishes appropriate forms of argumentation from inappropriate ones? It cannot be discursive truth; otherwise we would fall into a vicious circle. What is required, in order to safeguard and promote authentic dialogue or successful communication, is a valid form of argumentation. Gadamer is properly blamed to be ‘at his weakest in clarifying the role of argumentation in all the claims to truth’ (Bernstein, 1983, p. 174).
We can see why the link between truth and argumentation becomes the weakest link in Gadamer’s hands from a different angle: Rather than clarify the role of appropriate forms of argumentation in the notion of truth, Gadamer needs to clarify the role of truth in appropriate forms of argumentation. If the notion of truth is provoked to (as it usually does) establish valid forms of argumentation used in dialogical communication, it has to be primitive, a notion we could appeal to intuitively without justification, such as P. Horwich’s minimal theory of truth—a version of the deflation notion of truth (the truth predicates, ‘is true’ and ‘is false’, exist primarily for the sake of a logical need: to ensure that we can stick to argumentative dialogue). This is in fact the notion of truth we have employed so far.
Habermas, in his theory of communicative action, brings the role of truth in cross-language communication back to the central stage. Habermas’s main concern has been with such questions as: ‘How is social action possible?’ ‘How is a social order or societal integration based on coordination of social action possible?’ To address those questions, Habermas needs a comprehensive critical social theory, which in turn is based on his notion of communicative rationality. The concept of communicative rationality makes sense only against the background of Habermas’s theory of communicative action. To see the central role of truth played in Habermas’s view on linguistic communication, we will only focus on those parts of his theory of communicative action that focus on the crucial role of validity claims, especially truth claims, in communication.
Before we present Habermas’s theory, it would be instructive to see how much Habermas, in developing his theory of communicative action and rationality, is deeply in debt to Gadamer’s philosophical hermeneutics. Here are only some similarities and contrasts between them.
First, for both Gadamer and Habermas, understanding is the process of coming to understanding, which is oriented toward agreement or consensus, not simply comprehension. ‘The goal of coming to an understanding [Verständigung] is to bring about an agreement [Einverständnis] that terminates in intersubjective mutuality of reciprocal understanding, shared knowledge, mutual trust, and accord with one another’ (Habermas, 1979, p. 3). In fact, ‘one would fail to grasp what it means to understand an utterance if one did not recognize that it is supposed to serve the purpose of bringing about an agreement (Einverständnisy (Habermas, 1992, p. 78). For Gadamer, understanding is linked to coming to understanding and agreement through a fusion of horizons in conversation; for Habermas, understanding is connected with reaching understanding and consensus in terms of the evaluation of validity claims in argumentation.
Second, Habermas enthusiastically endorses the situatedness of understanding. The theme of our historicity is no less fundamental for Habermas than it is for Gadamer. With Gadamer, Habermas insists that we cannot completely escape from our own tradition and horizon (the lifeworld for Habermas) in seeking to understand what is apparently alien to us. Understanding always starts with one’s own tradition or lifeworld and moves to fusion of traditions.
Third, in Gadamer’s view, tradition is linguistically constituted and transmitted. Language is worldview. Understanding is no longer a sharing of consciousness and mental exchange; instead, all understanding is linguistic. Habermas’s emphasis on language is, if anything, even greater. For Habermas, language is a kind of metainstitution on which all social institutions are dependent; social institutions depend on social actions and social actions are constituted in ordinary language communication. As well as serving mutual understanding and the reaching of consensus, language is also a medium of domination and social power that serves to legitimize relations of organizing power.
Fourth, both Gadamer and Habermas are concerned with the vitality of dialogue, conversation, questioning, solidarity, and community to human society. Both agree that the breakdown of communication in the form of the non-agreement of reciprocal expectations is a threat to social life.
Fifth, Gadamer’s philosophical hermeneutics aims to unearth the fundamental conditions or presuppositions of human understanding (which is in essence communication). Similarly, Habermas’s theory of communicative action is directed toward uncovering the universal conditions or presuppositions of all communication (understanding). We will see the connection between Habermas’s notion of communicative action and Gadamer’s notion of understanding when we discuss communicative action in detail in the following.
Habermas identifies communicative action as the primary mode of action coordination within the lifeworld. Like Gadamer’s notion of lifeworld (the world in which we live out our everyday life, which is a coherent web of meanings, beliefs, values, and norms structured and transmitted linguistically), Habermas thinks of lifeworld in linguistic terms. The lifeworld is the world structured by language and cultural tradition, more precisely, a ‘culturally transmitted and linguistically organized stock of interpretive patterns’ (Habermas, 1987, p. 124). Since culture is the patterns of interpretation transmitted in language, it would better to say that the lifeworld consists of a linguistically transmitted cultural stock of knowledge that is always familiar. This stock of interpretive patterns, intuitive know-how, and socially established practices serves, in the form of implicit assumptions or presuppositions, as a background to all understanding and communication. Similar to Gadamer’s tradition, lifeworld is ‘the horizon-forming context of communication’ (Habermas, 1985), which can both restrict (as a limit) and promote (as a resource) communication. On the one hand, we can no more step outside the lifeworld than we can step outside language. On the other hand, the structure of the lifeworld lays down the forms of possible intersubjective understanding and meanings for communication. In modern societies, communication is achieved through a highly reflective mode of communicative action (in the form of critical, open-ended argumentative dialogue) against the background of plural lifeworlds.
Within the lifeworld, we coordinate our actions primarily in the form of communicative action, a form of social interaction in which the social actions of various agents are coordinated through an exchange of communicative acts that are oriented toward reaching understanding [Verständigung], i.e., toward reaching both comprehension and consensus/agreement [Einverständnis]. In its simplest terms, communicative action is the action oriented toward understanding (agreement). Clearly, not all actions are oriented toward understanding. Thus, it is important, in Habermas’s view, to distinguish communicative action from nonsocial instrumental action and social strategic action (purposive-rational action). Communicative action is oriented toward understanding, coordinated through consensus, and rationalized in terms of discursive argumentation, while instrumental action and strategic action are oriented toward success, coordinated through influence, and rationalized in terms of empirical efficiency of technical means and the consistency of choices between suitable means. In Habermas’s view, communicative action is the primary mechanism for social integration and purposive-rational action is merely a secondary one.
Although communicative action could be carried out by non-linguistic means, such as by extra-verbal body or facial expressions, Habermas focuses his attention on those communicative actions exhibited in the use of language, the primary mode of communicative action. Two points are worth mentioning here regarding the use of language in communicative action. First, according to Habermas, the use of language oriented toward understanding (the communicative mode) is the original or primary mode of language use. Other uses of language—such as the indirect mode of language use (the figurative, the symbolic) and the instrumental or the strategic mode (such as the perlocutionary effects of language use, ‘Give me the money or I will shoot you’, or ‘May your children die before you’)—are parasitic on the communicative mode of language use. Second, be aware of the fact that the language used in communicative action, according to Habermas, is not language conceived as a syntactic and semantic system, but rather language as it functions in social interaction, namely, language in use under certain circumstances, or simply, language as speech act (the utterance of a well-formed sentence in a particular situation). According to this pragmatic explanation of language, the basic unit of language is not the sentence in isolation but the sentence-as-uttered (the utterance), i.e., the speech act. Thus, communicative action is primarily a communicative speech act, i.e., the speech acts oriented toward understanding (not all speech acts are oriented toward understanding though).
Habermas calls the ability to perform speech acts communicative competence. To distinguish his own theory from other semantic theories and empirical pragmatic theories of linguistic understanding and communication, Habermas calls his theory of communicative action universal or formal pragmatics. It aims to identify the ‘unavoidable’ universal conditions or presuppositions underlying all communication and to reconstruct systematically the structures of communicative competence or implicit rules involved in the know-how of competent speakers in modern societies.
Habermas believes that we can reconstruct the universal conditions of communicative action through a close analysis of the formal features of our everyday communication activity. Indeed, it is not so hard to identify one precondition of successful communicative action. Imagine a professor instructs her lab assistant to fetch a frozen tube from the refrigerator by saying, ‘go to get the tube for me’, during a lab experiment on April 1st. The speaker, the professor, through performing the speech act by uttering the sentence under this certain situation, establishes an interpersonal relationship with her hearer, the assistant. The speech act makes an offer that the hearer can either accept or reject. Before the assistant can accept or reject the request, he has to understand it first. For this speech act to be successful (i.e., the assistant understands it and carries it out as requested) two conditions have to be met. On the speaker’s side, the professor has to have some good reasons to back up the validity of her request if the assistant questions it. The professor’s request could be valid in different senses, (a) It could be valid in the sense that what the professor said and what is implied are true (validity as truth): that there is a frozen tube in the refrigerator, that the refrigerator is nearby, and that the task does not require abnormal physical ability and is not hazardous to perform, and so on. (b) It may be valid in the sense that it is appropriate or normally right for the professor to make such a request (validity as Tightness): The professor feels that she has a right to request her assistant to do this for her. (c) It may be valid in the sense that the professor is sincere (validity as rightfulness): she really needs the tube for the experiment and the request is sincere, and so on. On the hearer’s side, the assistant, in order to understand the utterance and to want to carry out the request, has to be able to recognize what kind(s) of validity is (are) claimed (validity claims) by the request and what are the implicit reasons behind each validity claim, such that the professor would be able to defend them if questioned. This example actually reveals a defining characteristic of all speech acts: Everyday communication is connected with some validity claims that demand ‘yes’ or ‘no’ responses; and to understand a speech act is to understand the validity claims it raises.
Based on the intuition gained from the above example, we can formally introduce Habermas’s theory of validity claims, which is at the heart of Habermas’s theory of communicative action. According to Habermas, participants in communication can take up three basic attitudes—objectivating, norm-confirmative, and expressive—toward the world. Corresponding to those three different attitudes toward the world, we can relate ourselves, through communicative action, to three different (conceived) worlds: the objective world of what there are (objects, facts, and states of affairs), the social world consisting of shared norms and values used to regulate interpersonal relationships, and the subjective world of the totality of inner experiences (intentions, desires, feelings, and so on). Although a competent participant in communicative action can only take up a certain primary attitude to a certain world (the objectivating attitude toward the objective world, the confirmative attitude toward the social world, and the expressive attitude toward the subjective world), this does not prevent him or her from adopting different attitudes toward one and the same world. For instance, one could take up both an objectivating attitude (as the primary attitude) and an expressive attitude (as a secondary attitude) toward the objective world. For this reason, Habermas calls those different attitudes ‘the performative attitude’.
Accordingly, when performing a speech act, one can and must raise and recognize three different types of validity claims about something in each of the three worlds, as well as suppose that they can be reciprocally vindicated and redeemed if questioned, and can be accepted or rejected by the participants, (a) There are (propositional) truth claims—a claim to propositional truth—implicitly raised with and presupposed by the speech act about the objective world. For example, in our above example, in issuing the instruction to her assistant, the professor claims that there is a frozen tube in the refrigerator, and further supposes that the refrigerator with the tube stored is nearby and is easy to obtain. If the assistant replies, ‘but I have only five minutes before the next class’, then the professor has to either revise her implicit assumption or to make sure that he knows the truth of her claims, (b) There are (normative) Tightness claims—a claim to normative rightness—about the social world. In our example, the professor assumes, as a social norm, that she has a right to request the assistant carry out her instruction and he has an obligation to obey. If the assistant complained, ‘Why should I do that for you?’ The astonished professor would suddenly realize that her assistant lacks basic common sense of what his duty is as a lab assistant, (c) There are (subjective) truthfulness claims—a claim to truthfulness—about the subjective world. For example, the assistant could question the truthfulness of the professor’s intention by asking, ‘Do you really need the tube now?’ (Implication: ‘You are not pulling my leg, right?’).
Habermas’s theory of validity claims is theoretically based on his speech-act theory of meaning. In a speech act, the speaker comes to an understanding with the hearer about something. Following K. Bühler, as we have introduced in chapter 12, Habermas claims that we can identify three dimensions of an utterance’s meaning: its propositional content—which is supposed to represent states of affairs and can be defined in terms of its truth conditions; its illocutionary content—what is used in a speech act to enter a relationship with the hearer, and its expressive content— what is intended by the utterer. Every communicative speech act makes reference to each of the three dimensions of meaning simultaneously. Needless to say, those three independent and mutually irreducible dimensions of the meaning of the speech act correspond to the three types of validity claims: the propositional content to the truth claim, the illocutionary content to the rightness claim, and the expressive content to the truthfulness claim. Habermas thereby concludes that performing any speech act simultaneously raises all three types of validity claims corresponding to the three dimensions of meaning. Even so, Habermas insists that every speech act can be shown to raise only one type of claim directly, which is usually the primary intention of the speaker, and two other types of claims only indirectly. Thus, we can classify speech acts based on the type of the direct validity claim raised with a speech act: the constative speech acts loaded with direct truth claims, the regulative speech acts with direct rightness claims, and the expressive speech acts with direct truthfulness claims.
Closely related to the performative attitude that competent participants in communicative action take up toward each of the three worlds, the participants can (and should) adopt a reflective relation to each of the three worlds, namely, the participants relativize their utterances against the possibility that the validity claims behind them will be contested by others. In performing a speech act, the speaker does not blindly assume the validity of his or her validity claims and take a dogmatic stand on them (taking them as literally true or as they are); and the hearer does not blindly accept or deny those claims. Instead, those validity claims should be reciprocally contested, disputed, and defended by the giving of reasons and arguments. Therefore, with every speech act, in terms of the validity claims it makes, the speaker enters into an interpersonal relationship of mutual obligation with the hearer: The speaker is obligated to support his or her claim with reasons and the hearer is obligated to accept the claim unless he or she has good reasons not to do so. From this it follows that, to achieve the goal of understanding and agreement (mutual recognitions of the valid claims from others), participants in communicative action have to cooperate. Success of communication is not at the disposal of any individual participant.
To sum up, mutual recognition and reciprocal redemption of related validity claims made in communicative actions are a precondition (as a universal condition or presupposition of the possibility of communication) for all communicative actions. More precisely:
The success of a communicative action depends on (i) mutual recognition of the validity claims raised with the action as well as on (ii) the fulfillment of reciprocal obligation of redeeming those claims. On the speaker’s part, the speaker must be willing to clarify, if questioned, and be able to supply reasons to support, if challenged, the validity of the claims. On the hearer’s part, the hearer must be able to identify the validity claims and to supply reasons, if questioned, for acceptance or rejection of the claims.
A communicative action is conceptually linked with the justification of the validity claims it raises. To make a validity claim is to assume an obligation to justify it through argumentation; and to accept or reject a validity claim is to be prepared for providing reasons for the decision. Where validity claims are disputed, attempts can be made to settle the dispute by force, by appeals to authority or tradition, or by giving good reasons. Habermas argues that the rationality internalized within our communication practice (communicative rationality) dictates that disputed validity claims be settled through justification only. More importantly, Habermas emphasizes that the justification of the validity claims cannot be done monologically, either from the speaker’s side or from the hearer’s side, without the participation from the other side. Instead, the justification should be done discursively through argumentation in dialogue between the participants.
Why does the justification of validity conditions have to be dialogical and discursive? M. Cooke (1994) identifies the two basic arguments from Habermas’s writings for his intersubjective concept of justification through argumentation. According to the first one, validity claims cannot be justified independently of discussion with others. This is simply because justification, by its very nature, is discursive. ‘We understand the expression of “justify” when we know the rules of the argumentation game within which validity claims can be redeemed discursively’ (Habermas, 1982, p. 231). But Habermas has admitted now that the thesis of the internal connection between validity justification and discursive argumentation cannot be held universally. For example, for a significant category of truth claims, such as some straightforward empirical assertions, discursive argumentation seems not to be necessary under usual circumstances. Even so, the thesis still holds for most theoretical truth claims, such as the truth claims raised by our scientific languages. The second argument is from the fallibilist perspective of validity and justification. No validity claims, especially theoretical truth claims, can ever be conclusively verified as final ‘truths’. ‘Truths’ (as universal validities, such as propositional truth and normative ‘truth’), for Habermas, always have a moment of ‘unconditionality’ that transcends all spatio-temporal contexts. Thus, no validity claims are, in principle, immune to critical evaluation in argumentation. Similarly, no justification is, in principle, immune to possible revision either. Therefore, no justification is ever conclusive; it is essentially open. What counts as a good reason or a good justification can only be determined through critical evaluation with other participants in dialogue. A good justification can only emerge from and during such a process, not before. And what counts as a good reason now may not be good enough later if new evidence and methods emerge. Habermas concludes that, since fallibilism has an inescapable dialogical dimension, the justification of validity claims is dialogical in principle.
Since reciprocal justification of validity claims is essential to any communicative action and the justification is discursive in nature, communicative action is, in essence, discursive. In an ‘ideal speech situation’, communication is free from domination and distortion: Every validity claim is open to dispute; every participant has the same rights to bring forward reasons for or against the claims; both participants fully expect to engage in serious argument about the validity of the disputed claims; and a consensus between the participants on the disputed validity claims is eventually brought about solely through the strength of the better argument.
Calling the above undistorted communicative action an ‘ideal speech situation’ implies that our actual communications do not fully comply with those ‘strong idealizations’. Nevertheless, they do somehow make reference to, and attempt to bring about at least an approximation to it. In fact, the forms of argumentation involved in everyday communication could be widely different, ranging from very rudimentary—in which what counts as a good reason may be given by authorities, cultural norms, traditional rules or codes, or simply shared assumptions, many of which are not open to dispute in a certain society and culture—to open-ended and critical argumentation. Those different forms of argumentation can be distinguished based on what kinds of ‘idealizing suppositions’—which are rooted in the very structures of communicative action—are materialized within certain forms of argument. Some of the important idealizing suppositions are:
(1S1) |
No relevant argument is suppressed or excluded. |
(1S2) |
No force except that of the better argument is exerted. |
(1S3) |
All the participants are motivated only by concern for the better argument. |
(1S4) |
No validity claims are exempt from scrutiny. |
(1S5) |
All participants have equal rights to query any claims and to participate in debate. |
(1S6) |
All participants are motivated to reach a consensus (Verständigung: reaching-an-agreement) on the validity claims involved. |
(1S7) |
A rationally motivated consensus (Einverständnis: a well-grounded agreement) on two types of the universal validity claim (i.e., the truth claim and the Tightness claim) is in principle possible discursively (through argumentation in dialogue). |
Habermas maintains that all forms of argumentation involved in communicative action, even the very rudimentary ones, are based on some of those ‘idealizing suppositions’, such as ISl, IS2, and IS3, but only more advanced forms of argumentation satisfy other more stringent ones, such as IS4, IS5, and IS6. The form of argumentation that satisfies all the available idealizing suppositions, including IS1 to IS6 and maybe others, would be an ideal form of argumentation (‘ideal speech situation’). Forms of argumentations coming sufficiently close to satisfying the above listed suppositions, especially IS7, could be called ‘discourses’: They are either theoretical discourses that thematize universal truth claims—the claims to propositional (empirical or theoretical) truth—and practical discourses that thematize universal claims to normative Tightness. Be aware that discourses, for Habermas, are still sort of ideal forms of argumentation, which rarely exist in their pure forms.
Accordingly, we can categorize communication actions into different forms based on the forms of argumentation involved, such as basic communicative actions in which rudimentary forms of argumentation are utilized and advanced forms of communicative actions in which more demanding forms of argumentation are present, something close to discourses. What we are interested in here is the theoretical discourse since that is where scientific communication would fall. If any communicative action is a theoretical discourse at all, scientific communication is, I think, perhaps a paradigmatic theoretical discourse in action around us everyday. Scientific discourse has all the features of the theoretical discourse defined by Habermas. Since many of the following statements about the discursive feature of science are straightforward, I will state them without getting into details.
First, scientific discourse consists of scientific claims, such as ‘The association of the yin and rain makes people sleepy’, ‘The spatial location of the body b at time t1 is different from its location at time t2’. Scientific claims are oriented toward scientific understanding in two senses: They are external toward nature (to understand and explain nature), and they are internal toward other participants (to communicate to or to be understood by other scientists and the general public).
Second, scientific claims aim at validity and objectivity. In other words, any scientific utterance raises truth claims (empirical or theoretical truths), both about what is explicitly claimed and about what is implicitly presupposed by the claims. Those truth claims need to be substantiated and validated in order to be understood and to be carried out. To justify the truth claims raised with a scientific utterance, one is supposed to avoid any non-argumentative means, such as tricks, force, propaganda, psychological pressure, brainwashing, appeals to authority or emotion, personal attacks, and so on (what Aristotle called ethos and pathos, as different from logos). Instead, only argumentation is admitted—to argue for it, to present one’s best arguments, to compare one’s arguments with those of one’s critics, and to try to convince them by putting forward the best reasons available. In addition, no matter how strong and how convincing one’s arguments and reasons for one’s validity claims are, no reasons or arguments can be sustained without being subjected to counter-reasons and counter-arguments. In science, everything is fallible in principle; nothing is final, conclusive, or immune to revision, including the best evidence, reason, method, and justification at the time. Those determine that scientific justification cannot be monological, but rather dialogical, i.e., right or wrong through debate. Therefore, scientific discourse is by nature discursive.
Third, in an ideal situation, the form of scientific argumentation should be based on all essential ‘idealizing suppositions’ identified by Habermas, including IS1 to IS7. For example, it is an ethical duty of both parties involved in genuine scientific argumentation or debate (that is, the motivations of the argumentation are genuinely communicative) with good intention to reach agreement. To advance scientific understanding through possible consensus is, at least implicitly, inherent telos of scientific discourse. Of course, this does not mean that the agreement between two parties in a scientific debate could be reached in reality. But the history of science has shown us that possible convergence of different competing scientific theories is, in the long run, still possible. In fact, many scientific disciplines have developed and established some institutionalized, specialized forms of argumentation (‘expert culture’) which are fairly close to an ‘ideal speech situation’. Be aware that to say that scientific discourse has specialized forms of argumentation does not mean that those forms have to be formal logical rules, such as modus ponens or modus tollens. Formal logic rules are only parts of the forms of argumentation used in scientific discourse. Other parts consist of informal rules, constraints, and binds, such as the idealizing suppositions given above. We can cover all those formal logical rules and informal material rules established in scientific discourse under the name, following Aristotle, ‘dialectics’. In this sense, scientific discourse is not only dialogical and discursive, but also dialectic.
Based on Habermas’s theory of validity claims about the necessary connection between speech acts and validity claims and his discourse model of communication, we are ready to see how Habermas brings the role of truth in cross-language communication back to the central stage. To see this, the best place to start is with Habermas’s pragmatic theory of understanding.
We have seen that, for both Gadamer and Habermas, hermeneutic understanding is the process of reaching understanding [Verständigung], which is in turn oriented toward agreement/consensus [Einverständnis], not simply comprehension (in the sense of the English word ‘understanding’, which roughly corresponds to the so-called ‘propositional understanding’). For Gadamer, understanding is linked to reaching agreement through a fusion of horizons by engaging in genuine conversation. I have suggested that some notion of truth is indispensable for participants to engage in genuine conversation. Without a proper understanding of the role of truth in argumentative dialogue, Gadamer’s conversation model of communication is substantially incomplete. My criticism of Gadamer resonates with Habermas’s critique of existing pragmatic theories of meaning (such as the so-called use-theories of meaning as expounded by the later Wittgenstein). For Habermas, the main weakness of the existing pragmatic theory of meaning is that it loses sight of the connection between understanding and a context-transcendental concept of validity. By a context-transcendental validity, Habermas means a universal concept of validity that can potentially go beyond the conventional validity of a given form of life, tradition, or lifeworld. A valid claim is universal if the claim is agreeable to everyone involved, that is, everyone would agree that what is agreed upon to be valid is valid for everyone. The concept of truth would be such a universal concept of validity. Although each participant may disagree about the truth-value, even about the truth-value status of a truth claim, but truth is agreeable since all participants accept that a claim could be true or false and if it is true, it should be true for everyone. It seems to me that Gadamer makes the same mistake by removing propositional truth out of the universal conditions of understanding.
To overcome the weakness of existing pragmatic theories of understanding, Habermas turns his attention to formal semantics (from Frege through the early Wittgenstein to Davidson and Dummett), especially truth-conditional semantics, according to which to understand an utterance is to know its truth conditions. As I see it, there are two basic characteristics of the truth-conditional theory of understanding, which are at the heart of formal semantics. First, it identifies understanding with the knowledge of the conditions of understanding. By the conditions of understanding, I mean the conditions that make what is to be understood (assertoric sentences or utterances) possible or that determine the semantic contents (such as meanings) of what is to be understood. To understand a sentence S is to know what makes S possible by assigning S a meaning. Second, according to the truth-conditional account, the conditions that make an assertoric sentence S possible by assigning (literal) meanings to it are its truth conditions. Therefore, to understand S is to know its truth conditions (under what conditions a sentence is true). Both assumptions work together to specify a universal condition of understanding. Briefly, to understand is to know the conditions of understanding. In other words, to understand S is to know the meaning of S; and to know the meaning of S is to know the conditions that make S’s meaning possible. This basic ‘innocent’ intuition behind formal semantics, I believe, should serve as the foundation of any formal theory of understanding.
Habermas believes that these two basic assumptions of truth-conditional semantics should be preserved in his formal pragmatic theory of understanding. Corresponding to the first assumption of formal semantics, Habermas contends that ‘we understand an utterance when we know what makes it possible’ (1984, p. 297). Since what makes an utterance possible is, for Habermas, its validity claims, to understand an utterance is to understand the validity claims it raises, or more precisely, to understand the validity conditions (under what conditions an utterance is valid) as I will call it. More importantly, Habermas discerns from the second assumption of formal semantics what is missing from existing pragmatic theories, namely, a concept of universal validity—truth. It is necessary, Habermas argues, to re-establish the broken linkage, in both existing pragmatic theories and Gadamer’s hermeneutics, between ‘truth’ (in a broader sense, referring to a context-transcendent concept validity, including propositional truth, normative rightness, and subjective truthfulness) and understanding. By doing so, Habermas reestablishes the central role played by truth in understanding and communication.
Of course, formal semantics is still not the formal pragmatic theory of understanding that Habermas wants because it has been guilty of ‘the three abstractions’: the semantics abstraction (only focusing on the meanings of sentences in isolation and ignoring the use of sentences), the cognitivist abstraction (reducing all meanings to the propositional content and ignoring both the illocutionary and expressive contents), and the objectivist abstraction (truth conditions are objective as to what makes a sentence or an utterance true independent of any subjective comprehension or knowledge). As I have argued all along, as far as our P-languages is concerned, which are basically consisting of constative utterances or assertoric sentences, we can only focus on the propositional content and put other dimensions of meanings aside.2 For this reason, among the three abstractions, what we should be concerned about here is the so-called objectivist abstraction.
In fact, Habermas is not the first person who notices the objectivist abstraction. Not all formal semantic theories are guilty of it. The most notable truth-conditional semantics that commits itself fully to the abstraction is Davidson’s truth-conditional theory of understanding. According to it, there is a necessary conceptual connection between the knowledge of Tarskian semantic truth conditions and understanding. Such truth conditions are objective in the sense that there are semantic procedures available for effectively deciding what the truth conditions are for every assertoric sentence and when those truth conditions are satisfied, no matter whether the interpreter is entitled to know those conditions or not. The problem with such ‘objective truth conditions’ is, as Dummett (1993) and many others within the formal semanticist tradition have pointed out, that the truth conditions of many sentences are either unknowable or simply not available for the interpreter to know under certain conditions, such as the truth conditions of many scientific theoretical claims. To remedy it, Dummett introduces a so-called ‘epistemic turn’ within formal semantics and replaces Davidson’s ‘objective’ truth conditions with his ‘intersubjective’ assertability conditions (under what conditions a sentence is assertable), a verificationist version of ‘truth conditions’ which the interpreter is entitled to know when the conditions are satisfied.
Habermas applauds Dummett’s ‘epistemic turn’. Nevertheless, Habermas blames Dummett for not making the ‘the turn’ sharp enough because Dummett’s assertability conditions are monological in nature, that is, the interpreter is able to know the assertability conditions (that is what Habermas desires) without engaging in dialogue with the speaker (that is what Habermas tries to avoid). In contrast, for Habermas, to understand an utterance is not just to know the conditions of understanding, namely, the conditions that make the utterance possible—such as Dummett’s assertability conditions or Habermas’s validity conditions—which Habermas calls the ‘conditions of satisfaction’, but also to be able to justify the satisfaction conditions discursively in dialogue with the speaker, which Habermas calls the ‘conditions of validation’. In Habermas’s hand, the traditional one-layer conditions of understanding, i.e., the conditions of understanding as the conditions of satisfaction, is expanded into two-layer conditions of understanding, namely, as both the conditions of satisfaction and that of validation. Therefore, to understand an utterance, we have to know both the conditions of satisfaction and the conditions of validation.
By adding the conditions of validation as a condition of understanding, Habermas wants to achieve at least three goals. One is to make the conditions of understanding (the conditions of satisfaction) knowable to the interpreter. That the conditions of satisfaction have to be knowable is actually presupposed by the conditions of validation since if the former are not knowable for the interpreter, then the interpreter cannot justify them. The second goal is to make the conditions of understanding dialogical. The conditions of understanding need to be justified in argumentation oriented toward reaching agreement. Closely related to the second goal is Habermas’s attempt to substantiate his basic conviction shared with Gadamer, that is, understanding is a process of reaching understanding and agreement through argumentation in dialogue. In this sense, we can say that Habermas’s notion of understanding no longer simply focuses on comprehension as does propositional understanding, but rather becomes an integrated part of his formal pragmatic theory of communicative action.
To separate his version of the conditions of understanding from both Davidson’s truth conditions and Dummett’s asssertability conditions, Habermas names it the acceptability conditions. He claims,
In a distant analogy to the basic assumptions of truth-conditions, I want now to explain understanding an utterance by knowledge of the conditions under which a hearer may accept it. We understand a speech act when we know what makes it acceptable. (Habermas, 1984, p. 297)
As we have mentioned, the conditions under which a hearer would accept an utterance U are its validity conditions (conditions under which U is valid, or the conditions under which the validity claims raised with U can be satisfied). But the validity conditions, as the conditions of satisfaction, need to be supplemented by the conditions of validation, that is, under what conditions the validity claims can be justified with good reasons. To know the validation conditions is to know the kind of reasons that a speaker would provide in support of the validity claims raised. Therefore, for Habermas,
To understand an utterance is to know its acceptability conditions; and to know its acceptability conditions is to know its (a) satisfaction conditions (its validity conditions) and (b) its validation conditions.
For a constative utterance U, its validity conditions are its truth conditions. To know U’s validity conditions is to know U’s truth conditions, or to know U’s propositional content. In addition, to know U’s validation conditions is to know the conditions under which the speaker has convincing reasons for holding the truth claims TCs raised with and presupposed by U; and those reasons can only emerge in the process of intersubjective justification of TCs through argumentation in dialogue. However, if one knows how to justify the truth claims raised and presupposed by U, then presumably one already knows U’s truth conditions. That means, in the case of constative utterances, that we can reduce the satisfaction conditions to the validation conditions. Therefore,
To understand a constative utterance is to know how to justify, intersubjectively and reciprocally, the truth claims raised with and presupposed by it through argumentation in dialogue.
Recall that, in Gadamer’s hermeneutics, the relationship between understanding and communication is symmetric: The process of coming-to-an-understanding (agreement) is in essence the process of communication (conversation). The same is true for Habermas’s concepts of understanding and communication. The difference between Gadamer and Habermas seems to be this: Each starts with a different end of the equation. Starting with his notion of hermeneutic understanding, Gadamer ends with dialogical communication; while Habermas starts from his discourse model of communication, and ends with his notion of hermeneutic understanding. For Habermas, linguistic communication is an exchange of communicative speech acts oriented toward reaching understanding, i.e., toward reaching both comprehension and consensus/agreement. But understanding and agreement can only be reached through discursive argumentation in dialogue. More precisely, the process of reaching understanding and agreement is a process of reciprocal evaluation and justification of the validity claims raised through argumentation in dialogue, which is nothing but the process of communication in discourse. Therefore, for Habermas, dialogical communication turns out to be (hermeneutic) understanding.
We have seen the crucial role that truth plays in understanding of constative utterances. Since understanding is communication, based on Habermas’s discourse model of communication, the connection between truth and cross-language communication is evident. This is especially true if the cross-language communication between two typical P-languages, i.e., scientific languages, is concerned. Habermas distinguishes the cognitive use from the interactive use of language. In the cognitive use of language, with the help of constative speech acts, the thematic emphasis is on the propositional content of the utterance. In contrast, the thematic emphasis in the interactive use of language, with the help of regulative speech acts, is on the relationship that the utterance establishes between the speaker and the hearer. A P-language, a scientific language in particular, is primarily used cognitively. Although a P-language could be used interactively to regulate human relationships, it is certainly not its primary function. That means that the sentences used by a P-language are primarily constative utterances (such as ‘Element a contains more phlogiston than element b’) with propositional contents as their primary contents (literal meanings).
When used in cross-language communication, a constative utterance of a P-language directly raises explicitly and presupposes implicitly one type of universal validity claims—propositional truth claims, or claims to either empirical or theoretical propositional truth (such as ‘It is true that element a contains more phlogiston than element b’, ‘There are two elements named “a” and “b” ‘, and ‘Phlogiston, a chemical component which is measurable in quantity, exists’). Using our terminology, part of those truth claims presupposed implicitly by an utterance of a P-language is actually what we have identified as the semantic presuppositions underlying the utterance. Among them, the most fundamental presuppositions are so-called M-presuppositions, the absolute presuppositions taken for granted by the user of the language. While Habermas seems to focus only on one type of truth claims or presuppositions, i.e., the existential presuppositions, I think that he would agree with me that other types of semantic presuppositions, at least universal principles, if not categorical frameworks, can also (and should) fit in his schema of the truth claims raised with a speech act.
In Habermas’s view, communication is conceptually connected with reaching understanding and agreement through the intersubjective evaluation of validity claims in argumentation. Bear in mind that we have identified earlier two necessary conditions of successful cross-language communication:
(C1) Mutual recognition of the truth claims raised with and presupposed by the utterances used by the other language.
(C2) Reciprocal redemption of those truth claims through argumentation in rational dialogue.
Considering the conceptual connection between Habermas’s theory of communicative action and his formal pragmatic theory of understanding (two sides of the same coin), it should be unsurprising that those two conditions of communication between two P-languages are actually the two conditions of understanding (of constative utterances). That is, (a) the knowledge of truth conditions: recognition of the conditions under which the truth claims raised with and presupposed by the utterances used by the other language (the satisfaction conditions); (b) the capacity of justifying reciprocally those truth claims through argumentation in dialogue (the validation conditions).
But there is still one more crucial necessary condition of communication (understanding) missing from the list. What is the purpose of justifying reciprocally the truth claims involved discursively? Remember that Habermas, following Gadamer, contends that communicative acts are oriented toward reaching understanding; and reaching understanding is reaching consensus, not simply comprehension. For Habermas, the very idea of engaging in serious discussion and argumentation with one another with regard to the validity of a claim makes sense only if both sides have good intention to reach consensus on the validity claim; otherwise why should we even bother to argue with one another? Habermas thinks it is absurd to assume that participants in genuine argumentation could aim at disagreement. To facilitate communication (understanding), all participants are rationally motivated to reach a consensus on the validity claims involved (idealizing supposition IS6). Furthermore, communicative rationality inherent in everyday communicative actions requires that the consensus on the validity claims can only be reached rationally and discursively through rational argumentation in dialogue, nor through other non-rational means. This seems uncontroversial. What is debatable is whether a rationally motivated consensus (Einverständnis: a well-grounded agreement) on two types of the universal validity claims (i.e., the truth claim and the Tightness claim) is in principle achievable through argumentation in dialogue. Habermas is not very clear on this. It certainly cannot be achieved in most cases of communication. I would prefer to interpret Habermas as adopting IS7 as an idealizing supposition, which is achievable only in discourse, a sort of ‘ideal speech situation’. Thus, we have the third condition of dialogical communication:
(C3) In a discourse, reaching agreement in principle (maybe not in practice) on the validity of the truth claims in dispute discursively.
Under normal situations, especially normal intra-language communication, communication can continue smoothly so long as participants ‘suppose that the validity claims they reciprocally raise are justified’ (Habermas, 1979, p. 3). While all communication presupposes a background consensus that is typically taken for granted, the consensus quite often breaks down when the validity claims involved are not recognized or are questioned and disputed by the participants. Whenever this happens, argument is called for. In our example given earlier, the assistant could dispute one or two of the truth claims raised with the professor’s request by questioning or challenging them. For example, ‘I don’t see any refrigerator nearby’, or, ‘Do you really expect me alone to carry that tube?’ Alternatively, he can challenge the authority of his boss established by social norms: ‘Hi, I am just your academic assistant, not your servant’, or ‘I really do not feel like doing anything at this moment’. The professor can then defend her request by either clarifying the facts (‘See, the refrigerator is right on the corner behind the bookcase’, or ‘It only weights about five pounds’) or restating her authority status (‘Remember we are doing an experiment in the lab now. I didn’t ask you to clean my house’). Of course, the assistant could be stubborn and keep disputing. Hopefully, the muddleheaded assistant could be convinced by the professor’s arguments so that the communication could continue smoothly.
The intra-language communication in the above case sounds simple enough even when the validity claims are in dispute. However, for cross-language communication, especially in abnormal discourse, the process of communication would become much more complicated. Use our old friends Ms Jones (suppose this time that she is an assistant to Mr Wong for the time being) and Mr Wong as an example again. When asked why Jennifer, the patient, has a painful spleen, Mr Wong diagnoses that it is caused by an excess of the yin within her spleen (an asthenic spleen) by uttering:
(41) Jennifer’s painful spleen is caused by an excess of the yin within her spleen.
He further asks Ms Jones to administer some herbs to Jennifer to make up for the insufficiency of the yang. Remember that our Ms Jones is a bi-medical, but refuses to accept the M-presuppositions behind the Chinese medical language and for some unknown reason does not practice it. Somehow she decides to challenge the truth claims raised with Mr Wong’s request in front of Jennifer, such as (7) and (5):
(7) There is a fundamental element, force or principle in the universe, namely, the yin, and there is a pre-established connection between the human body and natural forces.
(5) All diseases are due to the loss of the balance between the yin part and the yang part of the human body.
When challenged, Mr Wong is obligated to redeem those truth claims with reasons (at least in front of Jennifer). Of course, Ms Jones could push the debate further.
Could the two participants eventually settle their disagreement by reaching consensus? This brings us back to our primary question again: Could two P-languages with incompatible M-presuppositions communicate with one another successfully according to Habermas’s discourse model of communication? To answer the question, we need to see whether the three necessary conditions of communication identified above (Cl, C2, and C3) could be satisfied in abnormal discourse.
Let us consider two P-languages, PL1 and PL2 (such as the Western medical language Lw and the traditional Chinese medical language Lc), with two sets of incompatible M-presuppositions. For the speaker (such as Ms Jones) of PL1 (such as Lw) to understand an utterance (such as (41)) of PL2 (such as Lc), she has to be able to identify and comprehend3 the truth claims raised with and presupposed by the utterance (such as (41), (7) and (5)). In other words, she needs to know under what conditions the utterance could hold to be true (to be valid). Notice that the truth claims to be identified and comprehended here not only include what is raised explicitly with (41), that is, the truth of (41), but also include what are presupposed implicitly by the utterance to be understood, i.e., (5) and (7). (7) is one of the M-presuppositions of the P-language to be understood, i.e., Lc. Therefore, one condition of communication specified by Habermas (Cl)—that is, mutual recognition of the truth claims—is exactly the conclusion that we have reached in chapter 12 when we discussed effective (propositional) cross-language understanding: One can understand the sentence of an alien P-language only if one can conceptually recognize and comprehend its underlying M-presuppositions. Failure to identify and comprehend them would disturb mutual understanding.
Nevertheless, we need to be cautious with the above comparison between Habermas’s conditions of understanding and the conditions of propositional understanding we identify in chapter 12. We have to realize that Habermas’s notion of understanding is not propositional. Following Gadamer’s innovation, Habermas moves away from the informative communication model, which is propositional by nature, by adding the second crucial condition of understanding or communication. For Habermas, only mutual recognition of the truth claims is not sufficient to understand them; in addition, those truth claims (presuppositions) have to be reciprocally justified in order for understanding or communication to be successful (By the way, in our case, mutual recognition of the truth claims is not a problem anyway since we have assumed that Ms Jones is a bi-medical who comprehends Lc).
Based on Habermas’s second condition of communication (C2), for the communication between PL1 and PL2 to be successful, the truth claims have to be justified discursively through argumentation in rational debate between participants. Notice that in a discourse, what is in dispute is about the truth-value of the truth claims, such as the truth-value of (41). But as we have argued repeatedly, in abnormal discourse, what is at issue is not the truth-values of the claims in dispute, but rather their truth-value status. For example, (41) is neither true nor false when considered within the context of Lw. It is an established notion of validity, i.e., the belief that an utterance has to be either true or false, that is called into question in abnormal discourse. In this case, to solve the dispute, the participants have to trace the source of the truth-valuelessness to the level of more fundamental truth claims (such as (7) and (5)) presupposed by the truth claim (such as (41)). Eventually, they find themselves face to face with two radically distinct P-languages with distinct, and often mutually exclusive, cosmologies, modes of thinking and justification, and categorical frameworks.
Could we truly engage in rational debate about those basic M-presuppositions between two P-languages, such as Lw and Lc? To engage in rational debate on the validity of the M-presuppositions of a P-language, such as Lc, both sides have to follow certain rules of debate and argumentation. But what count as acceptable appropriate rules of justification or forms of argumentation themselves are, for many P-languages, determined by a P-language. We have argued in chapter 10 that Lc is associated with a unique mode of reasoning, which we call the mode of ‘associated thinking’. It was a dominant rule of rational justification in the premodern Chinese tradition. The rule determines what facts count as evidence for justification and even what states of affairs count as permissible facts. For instance, the states of affairs about the interaction between the yin-yang parts in the human body and the yin-yang forces in Heaven count as permissible facts. For the same reason, the interaction provides good evidence for the justification of the claim that sleepiness is caused by rainfall. Similarly, I. Hacking identifies a distinct style of reasoning within the Renaissance tradition, which admits the states of affairs dealing with the mutual interaction among the human body, celestial bodies, and other chemical elements as alleged facts. This further determines that the associations between mercury, the planet Mercury, the marketplace, and syphilis count as evidence for the claim that mercury salve is good for syphilis. A Paracelsan sentence like ‘Mercury salve is good for syphilis’ is rationally justified based on this style of reasoning.
When those different forms of argumentation inherent in distinct P-languages are in conflict, can we locate some underlying common forms of argumentation? Habermas (1993) has admitted that there is no metadiscourse to which we can refer in order to justify a choice between different forms of argumentation. There is no forum for deciding when we should bring to bear arguments from other spheres of validity or for deciding which kinds of arguments are relevant. ‘Habermas raises the possibility that this might be a matter for the practical judgment of individuals, only to dismiss it immediately as unacceptable. ... Nor does he give any hints as to how we might solve the problem’ (Cooke, 1994, p. 42).
However, the problem we face in abnormal discourse is more serious than the problem of conflicting forms of argumentation. Logically speaking, the propositions expressed by each P-language have to be in some normal bivalent logical relation (logical consequence, logical consistency, or logical compatibility) in order for both sides to engage in a rational debate. Such logical relationships between two languages can be established only if each side adopts, not just recognizes, the M-presuppositions of the other P-language. Otherwise, apparently same sentences (such as (41)), which express propositions within the context of one language (say, Lc), do not express any propositions or thoughts when considered within the other (say, Lw). In this case, there is no way to establish any logical contact between two sets of propositions expressed respectively by the two P-languages. This is the reason why Mr Wong and Ms Jones cannot engage in a rational debate. This is, in effect, also the conclusion drawn by Kuhn based on his argument from the law of non-contradiction (chapter 7). Recall that Kuhn argues that the law of non-contradiction prohibits the occurrence of actual truth-valueless utterances in discourse. The utterances used in communication have to be actually true or false for both the speaker and the interpreter. The utterances under consideration are actually true or false for both the speaker and the interpreter only when both hold their underlying M-presuppositions to be true (chapter 9). This is possible only when the M-presuppositions of the two languages are either shared or at least compatible. We end up with the same conclusion as we have drawn repeatedly before.
Lastly, let us turn to the third condition of communication (C3). Can the speakers PL1 and PL2 (such as Lw and Lc) ever reach agreement, even in principle, on the validity of the truth claims (such as (41), (5), and (7)) in dispute through rational argumentation in debate? I am not optimistic at all since both sides have lost their basic means to reach such an agreement, i.e., through rational argumentation in debate.
In conclusion, according to Habermas’s discourse model, full cross-language communication in abnormal discourse is doomed to failure.
1 The following exposition of Habermas’s theory of communicative action makes extensive reference to M. Cooke, 1994.
2 I do not intend to deny the existence and importance of other dimensions of meaning at all.
3 ‘Comprehension’ used here should be understood as ‘propositional understanding’, namely, to know the propositional content, which is not yet the genuine ‘understanding’ in either Gadamer’s or Habermas’s sense.