SORITES, ISSN 1135-1349

Issue #14 -- October 2002. Pp. 63-69

A Trio on Truth

Copyright © by SORITES and Herbert Hrachovec

A Trio on Truth

Herbert Hrachovec


HENRY: Yesterday, looking at the log-files of an electronic discussion group I am running, I was in for a surprise. The president of the United States and his First Lady had subscribed. It was a very brief moment of elation, obviously. The notifications I had received could not have been true, even though they contained the correct adress of the White House and seemed perfectly ordinary messages. This hoax was just too obvious. But imagine Mr. Frank Costello subscribing as Mary Cooper. I would not have noticed at all. It is, in fact, well known that there is a lot of identity switching in Internet Relay Chats and MUDs. Chances are that Selma, your exiting contact in Cyberspace, is some male teenager making fun of you. Truth seems just off the point under those circumstances. Perhaps Baudrillard is right: it has been substituted by makebelieve and we should stop worrying and love the bomb.

THOMAS: Don't let yourself become confused. Baudrillard is but a hacker within theoretical discourse. You should trust him just as much as you rely upon those pranksters that faked President Clinton's subscription. Look at your instinctive reaction: You recognized more or less immediatly that someone was cheating here. This presupposes some reliance on the notions of correctness of procedures and well-foundedness of utterances on your part. One has to start from an understanding of truth to be able to complain about a lie. Simulation theory is wilfully exploiting our intuitions about truth in order to turn them into paradoxes. This move looks convincing under certain circumstances, but it is a cheap trick nevertheless. Deception derives its meaning from knowledge. Take away the possibility of knowledge and deception loses all its content. There is no way to avoid a theory of truth if you want to explore the philosophical implications of your experience as a listowner.

JUDY: Maybe Baudrillard went a little too far, but Thomas is certainly much too conservative. Let us agree on his claim that some account of truth is called for. This does not automatically imply the grasp of an elaborate set of foundational concepts. Thomas is talking about «correctness», but how can this be spelled out? I hope he doesn't believe that there are facts «out there» and truth consists in «getting them right». This kind of intuition is quite common, I agree. But it rests on extremely dubious theoretical foundations. How do you pick out states of affairs prior to describing them using true sentences? It seems that truth is just an honorary name we attach to utterances that serve a certain purpose, i.e. to guide us reliably through whatever environment we find ourselves in.

HENRY: Do you want to say that a sentence can be true on the Internet even though it contains an incorrect reference? Imagine I never find out about the spuriousness of Mary Cooper. All linguistic actions directed towards this construct turn out to work in one way or the other. Would I be entitled to the claim that statements relating to Miss Cooper are true? This sounds extremely odd. I agree with Judy that there has been a lot of ideological distortion of the notion of truth and we have admittedly not yet agreed on any of its features in this discussion. But wouldn't it be fair to just abandon the use of this term rather than twisting it into a completely counter-intuitive construction? If I understand Judy, sentences can be more or less successful -- and that is it. One might complain that this is not the whole story, but it is at least a coherent position. This does not seem to be the case with Baudrillard and his followers. They are invoking the whole set of classiscal intuitions only to subvert them. I do not deny that our language permits this strategy. Puzzles and paradoxes are sometimes instructive and amusing. But if there is any argumentative force in Baudrillard's utterances it has to come from their being true on some level. One should rather avoid the predicate if one is not prepared to own up to this.

THOMAS: Actually, big concepts are often disposed of in this somewhat disingenious way. After having discovered that there is no God, Freedom or Subject philosophers tend to propose substitute uses for those empty expressions. You can, for example, retain the wisdom of the Bible and yet refuse to follow its prescriptions. Turning «true» into a predicate on a par with «convincing» or «successfull in the long run» is an example of the same strategy. I regard this as conceptual cosmetics. There has, of course, to be a mechanism to challenge unquestioned metaphysical notions. But one should not disregard the basic intuitions at work here. Using the term «true» I want it to have a distinctive, non-derived meaning. Donald Davidson's scenario of «Radical Translation» is extremely helpful here. He envisages a person confronted with a group of strangers that speak an entirely foreign language. How would such a person react to their prompts? How, in fact, do we proceed with Maya languages or encrypted code? Davidson's answer: By tentatively reconstructing the chain of signifiers into an ensemble of sentences which are then checked against our own perception of reality and our interpretations of the foreign idiom. We employ our mastery of language, including an ability to proclaim a sentence as true, in order to understand foreigners or crack a code.

JUDY: Both of you have taken me to task for advocating a thoroughly pragmatic approach to truth. According to your opinion truth should not depend on the passing fortunes of particular linguistic environments. So it is quite a surprise to have Thomas praising Davidson's account, which puts truth explicitly into the context of particular persons, speaking a particular language. Davidson has recently rejected classical correspondence theories of truth. His reasons are precisely the reasons I have given before. A term's reference cannot be ascertained independently from the sentences in which it occurs. We have to handle terms according to our given understanding of the environment and there cannot be any guarantee that we are parsing the sentences -- and consequently conceptualizing the world -- in one uniquely valid way. Inscrutability of reference and indeterminacy of translation combine to make classical truth impossible. Davidson is actually quite close to hermeneutics, which is a notoriously relativistic enterprise. Trying to develop a maximally consistent picture of manifestations of unfamiliar agents, given a set of restricted and frequently puzzeling inputs, can be regarded as the job description of any cultural anthropologist.

HENRY: Your summary of Davidson's position is correct as far ays it goes, but you forgot to mention a crucial feature. Let me bring this out by returning to our previous example regarding the internet as an alien life-form. Many confusing expressions are produced on terminals. There is no way to know, initially, whether this is the operating system's ordinary routine, an error report or an e-mail message. Starting from the presumption that at least some of those cyphers can be treated as names of real persons a person may begin to make sense of a number of strings. This, in turn, allows the agent to construct increasingly larger chunks of the output on the terminal as meaningful communication. This is standard Davidson so far. But now imagine an insidious hacker who has designed a program which constantly juggles around name-like expressions. The minute you begin to form a hypothesis concerning the existence of Frank Costello this name is changed to Mary Cooper in all the relevant contexts. And once you conjecture that there has been a substitution another switch occurs. You would feel frustrated, that's for sure. But there is a deeper lesson to be learned here.

TOMAS: Your point seems to be a methodological one. Just as truth without extralinguistic warrants is but a shadow of its former self, naming would be severely impaired under those circumstances. One has, in fact, to put it stronger. It would be entirely misleading to call such exertions «naming». There is a semblance of meaning here, but it derives from external contexts that allow for a language-game like «naming» to be established. In other words: Something has to work before we can even begin to think about discussing possible effects of those disruptive activities. That's why I am so weary about considerations exploiting the ordinary meaning of «truth» to turn it against itself. Davidson's way of putting this is to say that most of the utterances directed towards our environment must be true. I read this not as an empirical estimate but rather as some kind of transcendental claim. «Truth» is built on a propositional activity that cannot effectivly be relativized within the respective language. Removing this feature does not change or extend the concept; it just destroys it.

JUDY: And what would be so terrible about getting rid of truth-talk? I notice that none of you has come up with an account that supports «truth» in the traditional metaphysical or epistemological fashion. Thomas seems to be insisting on some foundational surplus of the truth predicate, but he has failed to show us how it differs from my pragmatic understanding of our communicative activities. We presume some sentences to be true, i.e. to provide reliable guidance for our dealing with the world. So what? Those sentences do their work for us and we might feel that we would not like to miss them. But this is a contingent decision nevertheless. Another person, or I myself, at a different point in time, may arrive at dramatically different conclusions. You cannot have it both ways, namely in effect abandon the classical theory by taking singular language games as your point of departure -- and retain the reassuring objectivity provided by a «God's eye» position.

HENRY: There is indeed some disagreement on how to classify Davidson's position. Richard Rorty makes him into a wholesale pragmatist whereas Davidson himself insists on being a realist at various places. I am not interested in exegetical matters here, so I shall propose my own reading of the situation. To put it in a nutshell: I think that one has to have it both ways, even if this sounds like a paradox. The concept of truth is such that it calls for an attempt to put together two lines of thought that seem to exclude one another. Davidson is right in pointing out the unconditional character of truth-talk and we have to square this with Judy's insight, namely that we are perfectly able to relativize it in every single instance. Something is already settled before we even begin to make assertions about Mr. Costello. And yet, none of the sentences referring to this person is cast in stone. There is no ultimate guarantee that he cannot turn out to be Mary Cooper after all. The challenge is to make those ends meet.

THOMAS: Here is an idea on how to approach this problem in the Tarski-Davidson tradition. Tarski taught us one extremely important thing about the conceptual location of truth: It has to be situated at the crossroads of object-language and meta-language. Forget, for the moment, all the elaborate details of Convention T and recursive definitions of the T-predicate. Everything hinges on his decision to operate within a two-languages framework. Truth is a predicate that cannot be treated adequately by restricting oneself to one of those levels. To treat it adequately you must not get stuck within the one language you happen to speak. If you do, the need to talk about truth will never arise. There will be no occasion to rise above your statements and proclaim them to have a certain relationship towards the world. But you cannot completely immerse yourself in a meta-language either, since this language is defined by reference to an outside idiom. You may attempt to incorporate the object language into the meta-language. There has, however, still to be a mechanism for distinguishing between the two languages. In the absence of such a distinction the metalanguage collapses into any language this side of the problem of truth. «To make ends meet», as Henry has called it, one would have to exploit the changeover made possible by Tarski's setup. Combining unconditionality and relativity begins to make sense if one reflects on the fact that switching languages has been built into the problem from the very start.

JUDY: This business about the unconditional still escapes me, but I find some of your remarks quite interesting. You are apparently moving in my direction when you describe the impact of truth by means of linguistic strategies. Proceeding from Tarski's assumptions your observations strenghten the case of playing down all the venerable claims usually raised in connection with Truth. One might deconstruct the classical intuition in the following way: unconditional truth is a certain kind of phantasma arising from the careless transition from one given language to a meta-language. Imagine someone who has taken one step up from a given discourse. We can see her holding on to an impossible postulate, expressing an untroubled confidence into some of her utterances, knowing quite well that this urge is triggered by going beyond the realm of unconditional confidence in the first place. It is like reflecting on mutual trust among partners. Once you make an issue out of it, there have been reasons for mistrust. So I agree that your two-language framework throws some light on how confusing truth-talk can turn out to be. But I reject the metaphysical approach to trust as well as truth. We are beyond any simple origins once we talk about those subjects.

HENRY: No doubt about it. I accept your last sentence as it stands. But certain essential parts of the story have again been skipped. You insist on our being at a distance from unconditional uses of language. And there is in fact a disturbing tendency for people to become dogmatic when their initial assumptions are put into question. They often lack the resources to adapt themselves to a more flexible view of the issues considered. Arguing against rightwing politicians I use precisely your kind of argument. But there is more to this, and it is important to bring it out for methodological as well as for political reasons. It's quite simple, really and we touched on it before. No amount of second-level considerations will spare you the burden of establishing first-level confidence. You can question each of your trusted assumptions at one time or the other. You can also question some of them all the time. But questioning all of them all of the time is not an option. It is insanity.

THOMAS: We have been here before. So let me offer some clarification before we go into an argumentative loop. Up to now we have compared intuitions about the unconditional and about relativity somewhat superficially. In arguing for the former I have to be more precise. What exactly is being claimed to be unconditional here? Reflecting on our conversation I find that I have not been clear enough about this. There are two quite different contexts to be considered. The first one is a transcendental argument (sort of). Truth demands unquestioned success of some communicative behavior, any other way lies madness. This seems irrefutable to me. There is, however, a second line of thought and things are more complicated there. Analyzing transcendental qualities of truth-talk should not make us forget that it is sentences that are pronounced to be true. And there is serious doubt whether there can be a transcendental argument for any particular sentence to be true. So we are facing a gap here. A rather speculative consideration on the one hand and a matter of actual linguistic practice on the other. There is no easy attribution of properties across the divide. It is a long way from spoken sentences to reconstructing the conceptual obligations incurred by truth-talk. To put it succinctly: How does one derive metaphysical content from formal considerations? I can, admittedly, not name an unconditionally true sentence, even though I have argued in favor of unconditionality.

JUDY: So what's the use of your excursions into transcendental philosophy? The divide you are talking about seems to separate old-fashioned foundational reflections from the investigation of how we actually use our language. And since I know of no proposition that remains true under any circumstances whatsoever I just can't see the point of your insistence on something exempt from the flux of time and place. You have actually admitted that it could not be a sentence. What sort of entity is it, then? There is small comfort in an in-principle-stability that does not render any tangible results. Our discourse is, of course, guided by many regularities and norms. But they are themselves subject to change. The best way to approach them is via pragmatics.

HENRY: O.k., I'll give you a more or less pragmatistic account of how to arrive at ultimatly true sentences. I agree on the following observation: There are no privileged sentences as long as our empirical involvement with the environment is concerned. So I concede, for example, that every single name on the Internet could be missdirected. And you are right to imply that no amount of external metaphysical or transcendental pressure will fix this problem. But we have by now established a crucial distinction in our handling of various discursive disturbances. Errors, disappointments and improvements are always preceded by language use within a status quo. You simply cannot call a sentence into question without having placed it within a given framework of understanding and practice. You cannot avoid holding that there is a Mr. Costello before discovering that you may be wrong. Now this initial conviction is obviously not ultimatly true. But note the following: the way this term's use might be challenged is very different from how a new name is introduced. In the latter case I can just add an expression to my lexicon, whereas in the former case my empirical findings might have repercussions on the language employed. I may have to attend to some malfunctions. This activity is quite different from straight empirical usage. «Mr. Costello has written» might be an identifiable string of letters across various contexts. But it triggers a very different response depending on its placement within the variety of discursive practices. Its employment as an actual proposition has to be distinguished from its examination within a corrective loop.

THOMAS: If I understand you correctly you want to mirror my transcendental detatchment into pragmatical language use. You seem to say that recognizing errors and improving on given knowledge demands at least a threefold view on sentences, depending on whether they are used in ordinary circumstances, or challenged, or scrutinized. Given these distinctions it does indeed make sense to insist on a more subtle treatment of truth-talk. In claiming truth for a sentence we would have to allow for various alternative settings. Is a given sentence part of our taken-for-granted linguistic attitudes -- or is it singled out for special scrutiny? It is indeed a matter of discussion to what extent this «sentence» remains the same across contexts. Now, this enables me to rephrase my Davidsonian account. No transcendental moves (in the strict sense) have to be invoked to arrive at the claim that the majority of our sentences must be true. It suffices to say the following: The language-game of error-correction can only be played within the more encompassing game of asserting how things are. And this provides us with ultimate truth in the following sense: For every case of doubt an overwhelming number of undoubted propositions have to be in place.

JUDY: Your use of «ultimate truth» is very idiosyncratic here. It just seems to amount to saying that a lot of sentences go unchallenged for lenghty periods of time. This does not give them special status nor does it delineate an area of language-use that is somehow exempt from change. Some things don't change an awful lot, this is all the comfort I get from your remarks. By the way: Do you realize that your argument has a reactionary touch? It can easily be directed against critical inquiry. You are doubtlessly familiar with a certain type of attitude suppressing innovation. Some people feel that things are the way they are just because they have to be that way. On innumerous occasions attempts to change the status quo have been rejected by appeals to tradition masquerading as self-evidence. I am concerned that your emphasis on the more persistent features of our use of language work against curiosity and experimentalism. So, while I am prepared to accept the point that there are several distinctivly different strata within language, I'd still resist any attempts to strenghten the immobility of the most unsophisticated ones.

HENRY: I've seen this coming, but rest assured, your criticism is unwarranted. It is, in fact, one-dimensional pragmatism that is in danger of supporting any given set of power-relations. Thomas' considerations seem much more challenging to me. Coming up with something new and unexpected is quite cheap nowadays. Its the stuff advertisments are made of. Novelty is part of the most entrenched socio-economic order. The line of thought I have been advocating sets different priorities. It explores sentences which carry the weight of some discursive formation. Assertions that cannot be moved around at one's fancy. Think of the following remark: «If `Costello' is no name I do not know what names are.» Or else: «If this is not a letter by Mr. Costello I don't know what you are talking about.» On such occasions a name or a sentence are doing double duty. They serve as tokens of ordinary expressions and they indicate a point of no return. It's been discussed by Wittgenstein: A certain stick in Paris is a measuring-standard and its length is exactly one meter. Throughout our idioms there are expressions functioning in this peculiar way. I consider it of the utmost importance to find out about this double deployment, testing its power and recognizing how it works in any given case. Substituting a different city: to what extent do the Maastricht criteria define the framework of a common economic approach of the EU -- and to what extent are they themselves subject to changing economic strategies within the community?

THOMAS: Truth occurs at the interface of two languages or two usages of a language, this seems to be agreed upon. Now, Judy takes this as an unspectacular event, like switching from face-to-face conversation to talking to a person on the phone. I think there is much more to it. Something goes largely unnoticed in the ascent to the meta-level. That simple talk is often insufficient is in itself a remarkable fact. It is amazing how we manage to develop complicated linuistic strategies. There are doubts, demands for reassurance, guarantees are offered. All of this is worked out between level one and two. And one remarkable question pervades those transactions: How can speakers communicate their straight intentions within a setup that overrules this straightness from the start. This is what the problem of truth boils down to. It has its origins in ordinary language. But communication sometimes breaks down, sending shockwaves through our entire cognitive arrangements. Sentences are stripped of context and seem to be in need of external validation. Quite often it simply works and they are re-integrated in an encompassing linguistic practice. These are the paradigmatic cases of disquotational theories. The explanation of truth for a sentence consists in giving an account of how it is successfully incorporated into another operative language. I'm in agreement as far as this story goes. But I want to add one more prospect. Duplicating languages misses one point: Why are there multiple languages in the first place? Holding on to a distinguished notion of truth is just a way to resist papering over this rupture. Mr. Costello wrote a letter. Let us pursue such sentences in their variety of uses. One of them amounts to saying: If you do not take this for granted we have nothing in common any more.

JUDY: Have you been around before the Tower of Babel broke down? I just want to cope with some of the confusion that its collapse left behind.

HENRY: For me the shock of Babel resonates through language.

THOMAS: I am told that the towers of medieval cathedrals are embellished with elaborate sculptures very high up. No one can see them from the ground. Those artists haven been risking truth.

Herbert Hrachovec
University of Vienna
Department of Philosophy