Davidson, radical interpretation and universality

Why Davidson should have been a quietist

1. Reflective Hints

I want to introduce a new kind of post, which I will call Reflective Hints. “What is this supposed to be?” You may ask. I felt the need to do something different and maybe putting this under a label is just me giving myself permission to post what I want. These little pieces are not concerned with giving answers to specific problems nor are they closed off essays which definitely address a singular problem. At the same time, these essays are “bleeding edge” battlegrounds in my thinking, where I am constantly groping forward on uncertain ground. They are Hints: They are small lights in the sky that disclose certain directional movements through the thicket of thought. This little paragraph by Callum Hackett prompted me writing this:

Alternatively, you can go the whole pragmatist or Wittgensteinian hog and say that linguistics is full of pseudoproblems, but I think “why do certain sentence forms not occur in human languages?” is a legitimate question and these other frames just don’t have the tools to even begin answering it.

Found here:

This is the interface I have been wrestling with for the past few months. Do linguistic theories imply philosophical points? What is a philosophical point after all? Let me be a rascal and take this for granted: by philosophical point is meant some kind of universal and context independent validity for specific utterances. ¹ Just to give an example of how it was viewed historically, even by the founder of Pragmatism, here is a quote from C. S. Peirce.

To erect a philosophical edifice that shall outlast the vicissitudes of time, my care must be, not so much to set each brick with nicest accuracy, as to lay the foundations deep and massive. […] The undertaking which this volume inaugurates is to make a philosophy like that of Aristotle, that is to say, to outline a theory so comprehensive that, for a long time to come, the entire work of human reason, in philosophy of every school and kind, in mathematics, in psychology, in physical science, in history, in sociology, and in whatever other department there may be, shall appear as the filling up of its details. The first step toward this is to find simple concepts applicable to every subject.(Peirce et al., 1992, p. 246)

So taking this view of Philosophy for granted is debatable but a necessary methodological step here in order to get to a point of tension I’m wrestling with, because this is in the forefront in my Master Thesis about Davidson.

2. Davidson as a linguist

Davidson, read as a linguist, in regards to his project as a whole as a triangulative non-representationalistic theory of interpretation, crafts a theory which makes empirical-descriptive claims about verbal behavior that can be tested. This is why he holds that Tarski style truth theory exapted as a theory for meaning is a necessary formal device. It is necessary because it allows for the formulation of empirical hypotheses whose particular instantiation he calls T-sentences.

A Tarski-style theory for meaning, in short, is based on Convention T. Convention-T requires that the truth predicate of a language is defined to entail for every sentence in a language a theorem of this form:

“s” is-true-in-L iff p

where “s” is the name or systematic description of s and p is the translation of s into the language of the theory. The most well known example of a closed sentence is:

“Snow is white” is-true-in-English iff snow is white.

Davidson calls such theorems T-Sentences.

Davidson treats such T-Sentences as empirical hypotheses. Whereas Tarski took meaning for granted in order to define truth (at least its extension) as the satisfaction of a closed sentence by all possible entities, Davidson presupposes the concept of truth in order to craft an empirical theory of meaning. This is presupposed by Davidson’s intention of constructing an empirical theory of interpretation – we have to be able to decide the truth of such a T-Sentence, otherwise the theory would be completely trivial and the concept of truth would be merely disquotational. The general concept of truth provides us the ability to judge T-Sentences concerning their truth value. Davidson introduces this explicitly as the “truth axiom.” Davidson puts it thus:

The trick is just to add to Tarski’s definition of a truth predicate for a language L (say, ‘s is true;’) the remark that Tarski’s predicate holds for all and only the true sentences of L. Here, of course, the word ‘true’ expresses the real-life, substantive, undefined concept we need for serious semantics. Let us call this remark the truth axiom.(Davidson & Journal of Philosophy Inc., 1990, p. 292)

Davidson is very explicit that the general concept of truth cannot be reduced further. So he writes:

We should apply this obvious observation to the concept of truth: we cannot hope to underpin it with something more transparent or easier to grasp(Davidson, 2006, p. 21).

“This obvious observation” is that we cannot reduce primal concepts further. He is well aware of how this might be taken as a father telling a child who has asked one too many questions to keep their mouth shut and go play a little. He resolves this brilliantly by saying that we are still spellbound by the Socratic idea that we first have to find the essence of an idea in order to use it.(Davidson, 2006, p. 34) Obviously, Davidson drops this.

This, for now, leaves us with truth being a methodological presupposition for a theory of interpretation Davidson envisions.

3. Rorty’s Admission

3.1 Normativity of Charity

In the discussion of Davidson’s philosophy, Ramberg persuades Rorty to admit that the Davidsonian normativity of triangulation is inescapable. This inescapability means that in order to have descriptive vocabularies, we already need to move inside this normative situation. Normativity is just what we can’t get out of. Rorty puts it thus:

Ramberg suggests that we see the ability to ascribe rights and responsibilities (along the lines of Brandom’s “social practice” reconstruction of the vocabularies of logic and semantics) as (usually) a prerequisite for the ability to predict and describe.(Brandom, 2000, p. 371)

This normativity is not thick, it is very slim and contextual. I’d go so far and say that the normativity Davidson is urging for is completely without specific content. Methodologically speaking this just means that we have to apply on the speaker-interpreter relationship the principle of charity, which assumes that we need to interpret the speaker in the most rational way possible by aligning their talk with how we, the interpreter, speak in order to understand the speaker. Davidson writes:

Charity is a matter of finding enough rationality in those we would understand to make sense of what they say and do, for unless we succeed in this, we cannot identify the contents of their words and thoughts. Seeing rationality in others is a matter of recognizing our own norms of rationality in their speech and behavior. These norms include the norms of logical consistency, of action in reasonable accord with essential or basic interests, and the acceptance of views that are sensible in the light of evidence.(Davidson, 2006, p. 319)

Following this, Davidson rejects that understanding is constituted by following a specific rule. Substituting the idea of rule-following as ingrained rules that dictate language use, we need only one additional norm: the speaker intends to be understood in a certain way.(Davidson & Philosophy Documentation Center, 1992, p. 261) ² The actual fulfillment of an intention can be judged as false or true.

Charity is therefore the presupposition that others too have similar propositional attitudes as we do, navigate in simlar ways the conceptual landscape, have similar goals and aims. We wish that something happens, we are afraid that something will happen, we regret that something has happened. The intention to be understood in a certain way opens then the space for the concept of truth, because we can be mistaken, others fail to understand what we are saying.

3.2 Acknowledgement of Causality

On the speaker-world side we have to acknowledge the causal prompting of world and speaker. This is roughly what Quine had in mind when discussing the indeterminacy of translation and ontology. He used the infamous example of “Gavagai.” In brief: when encountering a completely new language, it is impossible for a field linguist to determine whether the person being interpreted is saying “rabbit,” “it is rabbiting,” or “there are undetached rabbit parts” with “Gavagai.” This leads Quine to conclude that ontology is underdetermined, meaning the translation is never uniquely determined by empirical data.

Davidson has no problem with that. He has only a problem with Quines idea of reference. Whereas Quine argues for a proximal theory of reference (the pattern of nerve firings), Davidson assumes a distal theory of reference. And we can see why: If the causal acknowledgment of speaker-world is necessary for an interpreter to interpret the speaker, then the interpreter has to acknowledge the distal object which prompted the speaker to utter something as a common cause. Therefore Davidson writes:

The relevant stimuli are the objects or events we naturally find similar (tables) which are correlated with responses of the child we find similar. It is a form of triangulation: one line goes from the child in the direction of the table, one line goes from us in the direction of the table, and the third line goes from us to the child. Where the lines from child to table and us to table converge “the” stimulus is located.(Davidson & Philosophy Documentation Center, 1992, p. 263)

4. The district of the question

The description of a theory of interpretation itself rests on a normative play between agents and a causal relationship between agents and world. If this normative-causal triangle is really the inescapable situation we are in, so that we do not get “outside” in any shape or form – and we have given up representation and non contextual reference as bedrock descriptive tools which explicate “in itself” relations between word and world – then we have given up the quest of “getting at how things-in-themselfs (or any structure whatsoever) really are.” As Rorty puts it:

There is no such thing as Reality to be gotten right – only snow, fog, Olympian deities, relative aesthetic worth, the elementary particles, human rights, the divine right of kings, the Trinity, and the like. (Brandom, 2000, p. 375)

Why can we get “snow” right or wrong? Because we have the intention to be understood in a certain way, we stand in a causal relation to the world and we want to make ourselves understandable to others. Davidson second person approach drops correspondence completely without dropping the idea that we can get “things right.”

If we want to be completely rigorous with what we are doing right here, then the normative vocabulary ³ is not something that can be captured at all in a descriptive vocabulary. This is why Davidson still assumes irreducibility between mental and physical vocabularies. Even if Davidson would most likely rebel against this, I’d put it thus: the normative vocabulary is in itself a type of descriptive vocabulary just differing in purpose and aim from physical vocabularies. Both vocabularies are irreducible, that may be true, but they are irreducible not because they are inherently prioritized as Rorty suspected, ⁴ but because the methodological necessity for irreducible axioms implies irreducibility of vocabularies. We need to presuppose different foundational concepts for normative and non-normative vocabularies, but both are language games. Concerning irreducibility, Davidson writes, that a concept is only irreducible relative to specified resources.(Davidson, 2001, p. 85) Broken down that means that the color red is not reducible to geometrical concepts like “asymptote” or “triangle,” but that it is reducible to physical vocabularies like wavelengths. Red is irreducible relative to geometry. Davidson assumes that we can get “so foundational” that no reducibility is really imaginable. This is the case for normative and non-normative vocabularies.

The point of critique now comes in that way: The normative, like the non-normative vocabulary, can only be tested by the criteria and purposes we demand. We need different concepts, but those are only justified in yielding working theories for different kinds of enterprises. “What does ‘working’ mean in general?” you may ask, yet I won’t give an answer because this would imply that the validity of theories, utterances, and symbols is uniquely determinable by a single method. I take them to be contextual questions, needing to be asked and answered in the messy and frictional diverse enterprises by humans actually doing the work.

But, I take Rortys idea seriously that vocabularies are vertically (meaning epistemologically/metaphysically/ontologically) unextended; they all stand side by side horizontally. Wittgenstein read as a quietist enables us to say that finding any foundation or escape out of language by language is futile. We cannot get outside the game while in the game. The point is, that even the theory of radical interpretation by Davidson falls under this. The normative vocabulary is still a vocabulary, still a practice of humans. This is an antifoundationalist and contextualist perspective concerning vocabularies. There is no way of prioritizing any vocabulary independently of specifying a purpose. The description of the physical cannot be crammed together with the description of the mental. Both are differentiated by requiring different irreducible starting points. Davidson seems to agree on that point as well:

The point of the concept of a language, then, and its attendant concepts like those of predicate, sentence, and reference is to enable us to give a coherent description of the behavior of speakers, and of what speakers and their interpreters know that allows them to communicate. I do not suggest that speakers and those who understand them must themselves be able to provide such descriptions of their abilities and behavior.(Davidson & Philosophy Documentation Center, 1992, pp. 256–257)

5. The Error

Davidson was a brilliant mind. He was one of the great and I learned much from him and continue to do so. In my eyes he made this following error.

The theory by Davidson is a description of what interpreters must know in order to speak. It may be obvious now that we are on the verge of substituting an ancient duality for a modern one. The difference between appearance and reality is discarded and faces us again as norm and description. If there is no bedrock which “gets reality right,” then we have to give up the idea that we find any kind of conditions of possibility. What Davidson’s approach can provide us is a new resting and starting place, which makes ancient problems impossible to ask. Davidson – seen as making negative points – allows us to just stop asking those questions in the same way as we stop attempting to call someone with a calculator once we realize the attempt is futile. But we cannot make positive statements, statements about the conditions of something, unless the condition for the respective vocabulary is meant. We cannot find conditions for something if by “something” we mean “the thing-itself out there.” The point where I stub my toe therefore is when Davidson writes things like:

For the most part, the concepts philosophers single out for attention, like truth, knowledge, belief, action, cause, the good and the right, are the most elementary concepts we have, concepts without which (I am inclined to say) we would have no concepts at all.(Davidson, 2006, p. 20)

And then also writes:

For when we ask where the norms come from that each of us applies in understanding others, the answer is that they cannot be derived from a source outside ourselves, for any attempt to check with others drives us back to the process of interpretation in which we necessarily employ our own norms.(Davidson, 2006, p. 319)

If the theory had genuinely and purely descriptive aims, as he himself wrote, and as such rests on the primal normative situation which is not reducible to the descriptive (otherwise he would not be an anomalous monist), then it does follow that the descriptive necessary conditions generalize beyond the theory and allow Davidson to say that without the concept of truth we would have no concepts at all. What he should instead have said is that those concepts are concepts without which we would not be able to describe verbal behavior.

To turn back to the beginning quote by Callum: The tension between linguistics and philosophy lies in the former’s contextual and provisional validity and the latter’s demand for universal validity, (linguists – of course – can also experience the universalistic urge) to say something about the “conditions of there being x” not only in methodological enterprises, but independently of them. Those urges are still found in people taking intuitions as epistemologically valuable sources of knowledge.

To misconstrue the contextual results of linguistics for philosophical universalistic aims is an analytical remnant of the hope that if we find some way of specifying the conditions of language, we can make universal philosophical points. I do not demand more in order for universalistic claims, but that we should drop that urge entirely.

Davidson should have stayed quietistic about truth being a general and primitive concept, that it is a methodological presupposition for the theory of radical interpretation he envisioned, as the last frontier of metaphysical urges lies in the aim for stance independent universality. Not only normativity is inescapable, contextuality is as well.

References

Brandom, R. (Ed.). (2000). Rorty and his critics. Blackwell Publishers.

Davidson, D. (2001). Subjective, intersubjective, objective (Reprinted). Clarendon Press.

Davidson, D. (2006). Truth, language, and history (Reprint). Clarendon Press.

Davidson, D., & Journal of Philosophy Inc. (1990). The Structure and Content of Truth. Journal of Philosophy, 87(6), 279–328. https://doi.org/10.2307/2026863

Davidson, D., & Philosophy Documentation Center. (1992). The Second Person. Midwest Studies in Philosophy, 17, 255–267. https://doi.org/10.1111/j.1475-4975.1992.tb00154.x

Peirce, C. S., Houser, N., & Kloesel, C. J. W. (1992). The essential Peirce: Selected philosophical writings. Indiana University Press.

---

1. You can of course disagree with me on that! Terming this specific utterances already discloses me as a nominalist concerning universals. The indeterminacy of the word red is not that it references an universal upon which the particulars take part in order to be what they are, but that “red” has no clear boundaries, forged by and through practical demands.↩︎

2. But is that justifiably called a norm? Do norms not require that we can explicitly specify them?↩︎

3. I’m unsure if we should call this “vocabulary” though, as this implies that we enter into different discussions in kind.↩︎

4. Rorty so worte, that we need to have a normative vocabulary in order to have a descriptive one. Rorty writes “A normative vocabulary is presupposed by any descriptive vocabulary - not because of any inferential relations between sentences in the one vocabulary and those in the other, but pragmatically. We could not deploy the descriptive vocabulary unless we could also deploy the normative one, just as we could not employ a screwdriver if we did not have hands. As Ramberg says”Describing anything, if Davidson is right, is an ability we have only because it is possible for others to see us as in general conforming to the norms that the predicates of agency embody.” ↩︎