From the ALWS archives: A selection of papers from the International Wittgenstein Symposia in Kirchberg am Wechsel

1. Complicated forms of feeling.¹

Let us first turn our attention to Wittgenstein’s suggestions about the constitution of our inner world – our repertoire of feelings, emotions and inner experiences. Those subjects able to learn and master a language are also able to develop that repertoire in many ways.

A baby, for example, is a subject who feels pains as well as throbbing pains, although the specific form of pain we describe as ‘throbbing’ is something whose expression requires the use of a language. Matters are different, say, with the concept of ‘fear’. While we should be often right in stating that this baby here fears the stranger in front of her, it would not seem quite acceptable to say that she fears an ancient mausoleum just at the end of the street. In learning a language we not only discover how to express and describe our pre-existent kinds of fear; rather, the things we fear and the ways we fear are largely expanded – up to the point of our acknowledging a speaker as someone in position to claim the discovery of an unsuspected fear within herself. But things go still another way when it comes to feelings such as ‘hope’ or ‘grief’, for here we must say that the very feeling of these experiences requires the learning of a language. It seems that nothing like hope or grief could correctly be ascribed to a sentient subject unless the subject were at the same time described as a language user. That is the position Wittgenstein assumes when he describes “the phenomena of hope” as “modifications of this complicated form of life” (Wittgenstein 2001, p. 148).

Wittgenstein’s remarks, as I read them, leave us enough room to doubt whether modifications of the talkative life-form of subjects come about in the same way by the users of very different languages. The point is reinforced by analogy with the idea that “if a concept refers to a character of human handwriting, it has no application to beings that do not write” (Wittgenstein 2001, p. 148). For it is also the case that different concepts of characters of writing apply to different human systems of writing.

If I am right in keeping your imagination open to the possibilities of modification of the complicated language user’s form of life, I should be also right in asking whether at least some part of our inner experiences may take the form of concepts that vary according to one’s cultural background – and therefore whether the mental repertoire of two different subjects could be other than identical.

2. Setting the stage to the radical interpreter.

Imagine a field linguist or anthropologist trying to explain the behavior of subjects whose language and habits remain unknown to her. The first problem in such situations is that thought, speech and action always suppose each other when it comes to human agency. So, before ascribing any meaning to the utterances of a subject, the scientist needs some grasp of the subject’s beliefs and intentions; on the other hand, beliefs and attitudes are only finely discriminated under the assumption that she understands the subject’s utterances. Thus, according to Davidson, “we must have a theory that simultaneously accounts for attitudes and interprets speech, and which assumes neither” (Davidson 2001, 195).

The radical interpreter relies on two indispensable instruments to enter into the circle of interpretation. One is the basic evidence available in the form of general attitudes towards sentences, while the other is the interpreter herself; this second point concerns us more directly and deserves some clarification.

One of the basic tenets of radical interpretation is the so called principle of charity: the idea that the interpreter, in interpreting thought and talk of others, must methodologically ascribe to her subjects as much true beliefs as she can (according to her own views). Given the starting point of the radical interpreter, there seems to be no other theoretical procedure available: “Since knowledge of beliefs comes only with the ability to interpret words, the only possibility at the start is to assume general agreement on beliefs” (Davidson 2001, 196). In this sense, the interpreter herself is her main instrument of work.

Besides, Davidson’s suggestion seems to be that the interpreter secures a base of understanding from out of which disagreement becomes first identifiable and accountable for. To make for the point of methodological charity, therefore, we must keep in mind that “widespread agreement is the only possible background against which disputes and mistakes can be interpreted” (Davidson 2001, p. 153). Such a heuristic approach gives way to the idea that in ascribing attitudes to subjects and meanings to utterances, the interpreter should assume neither: for insofar as agreement is seen as a mere platform that allows tracing those threads of disagreement between interpreter and interpreted subject, any ascription of attitude to a given person or of meaning to one of that person’s utterances may be revised and corrected while interpretation is on its way.

3. Further Complications.

Suppose we present to the radical interpreter the results of our reflections on the constitution of our own mental states and its relations to acquisition of language. We concede to her that attributions of feelings like pain, and of mental attitudes such as belief, desire or fear, must be held not just as basic ingredients of interpretation, but of anything we may intelligibly call a mental life. We concede that the interpreter who refuses to describe a subject’s behavior by means of these basic concepts is never likely to get a foot in the door. Besides, we add the interesting argument, based on observation of our own forms of life, to the effect that feelings and attitudes correctly ascribable to pre-linguistic infants might be extended, expanded or improved, but never annihilated, by language acquisition. So making sense of forms of life according to which people do not feel pain, do not desire things nor hold some of them true is an utterly dim possibility.

But what are we going to say about those occasions when a subject’s behavior suggests she’s in grief, or when we are inclined to interpret one of her utterances as manifesting a feeling of hope? The Davidsonian radical interpreter will naturally hold on to the idea that, no matter what the practices, language and cultural background are, the methodology of interpretation keeps its regular course: so what we should do is to remember that first-person present-tense statements are usually true (unless we have reasons to believe the subject is lying) and that what is being self-ascribed is much like the same things we regularly ascribe to ourselves; besides, we should keep in mind that our interpretation could be corrected at some later stage, and that we might come to realize the subject was not feeling what we at first figured she was.

But if we rely on Wittgenstein’s reflections on the mental, then it seems we should proceed with caution when it comes to interpreting some thoughts and behavior of subjects remote from our common cultural background. Perhaps we should avoid making use of concepts such as ‘hope’ or ‘grief’. Perhaps applying these concepts we lay ourselves open to the threat of projecting our provincial perspectives and experiences onto subjects who not necessarily share them. For it may be the case that mental concepts entirely language-dependant, such as hope or grief, be modifications of human experience linked with specific languages in which they are used.

Can we make good sense of the above suggestion? I think we get a clear answer by aligning the problem with the standpoint of the radical interpreter.

4. Back to the radical interpreter.

Now suppose we have reason, as interpreters, to say that some mental concept c as used by a speaker of language L in a large range of sentences had no clear counterpart in our own language; or else that in giving the truth conditions of L-sentences we never made use of something like our own concept of ‘hope’. Here, some people will urge, there’s little point in accounting for the relevant differences in terms of explicable disagreement; rather, the facts must be accounted for by saying that each concept, c and ‘hope’, is relative and necessarily referred to its respective (kind of) language. So what we get here is a case of full-fledged conceptual difference between two languages. And this is the best reason we could ever find for the belief that “the phenomena of hope” are local.

It seems to me that the above account depends on a mislead construal of the expression ‘conceptual difference’. I will maintain, in any case, that the inference is wrong from ‘lack of conceptual equivalence’ to ‘linguistic relativity’, ‘difference in mental repertoire’, or any other exciting suggestions. The point will be made first in connection with Davidson’s and then with Wittgenstein’s ideas.

1. Allow concept c to be a peculiarity of L in the above sense. In this case, the idea that concept c of language L lacks a good counterpart in our language is tantamount to saying that we could not find appropriate truth conditions for some sentence, or some range of sentences, uttered by speakers of L. But that amounts to no more than recognizing we did not arrive at a correct interpretation of some or all of the sentences in which c occurs. According to radical interpretation two conclusions are therefore available: either we do not have enough evidence to support that the alien sentences were, to begin with, a sample of speech behavior; or else we must improve our platform of agreement with the speakers of L until we get at a plausible interpretation of the aforementioned sentences.

The story goes like this: i) in order to be in a position to judge that others had thoughts different from our own, expressed by mental concepts peculiar to their language, we must be in a position to interpret their utterances; ii) in order to interpret people’s utterances we must find the appropriate truth-conditions of the uttered sentences; iii) the truth conditions of an alien sentence must be formulated in the interpreter’s language; iv) therefore, nothing could count as a case of mental or conceptual difference between ourselves and others that did not require an alien piece of speech behavior to be translatable into our own language. We may also repeat points i)-iv), so as to fully attend the initial demands, conceiving a speaker of L as the interpreter and ourselves as the interpreted subjects. Translatability expresses a symmetrical relation.

Of course point iii), in the above sketch, is the most crucial to the argument: it shows that the concepts of truth and translatability go always together in interpretation. What seems important here is to acknowledge that the interpreter would never be able to tell an alien speaker’s utterance was true unless she were convinced that she herself might have held one sentence of her own language as true under the same conditions of the speaker’s utterance. In other words: the interpreter could never be in a position to say she understood the meaning of some alien sentence if she had failed in making her understanding linguistically explicit; and of course her only means to make it explicit is using some language she already masters.

Now we have a clearer grasp of our problem. Suppose one is right in assuming translatability as a criterion for the linguistic. Then how are we to make sense of claims about conceptual difference? I think the only way out is to show we can make good sense, but no big deal, of these claims. The sentence “Concept c cannot be rendered in our language” only states the very natural fact that c has no straightforward vernacular counterpart in our language. But when someone says that a given word has no counterpart in another language we should not think she is saying more than she really is. She could not be saying, for instance, that only the speakers of L had some mysterious grasp of c, or that c constituted some mysterious object or state accessed only by those speakers. Radical interpretation precludes radical excitement.

1. Wittgenstein asks: “Can only hope those who can talk?”; and his answer is: “Only those who have mastered the use of a language” (Wittgenstein 2001, 148). The question we have been trying to answer was: “Can only hope those who master given languages?”. And I think the answer is no, at least if we rely on the ideas connected with Davidson’s account of radical interpretation.

If we accept the standpoint of the radical interpreter, perhaps we can come to understand why Wittgenstein describes the phenomena of hope as modifications of this complicated form of life. That is, why he describes the ability to talk and master whichever languages as one single form of life – instead of forms of life in the plural. I think we should understand the criteria of learnability and translatability as a natural consequence of language’s publicity. We could not make sense of some thought or feeling that was both thought or felt only by certain speakers and that was linguistic all the same. Yet this is precisely what is required of us in order to make sense of the sentence “Only those who master given languages can hope”. What we must realize once and for all is that privacy and publicity fare badly together.

Of course the interpreter is bound to find all difficulties ordinary translators usually experience, and these difficulties are often correctly attributed to some tricky words of a language. Still, here is the place to train our capacity of resisting philosophical temptations. No interpretation could be so difficult as to prove our mental or linguistic lack of resources – instead of proving, like Wittgenstein’s attacks to Sir James Frazer, that imagination just needs some workout.