Umberto Eco takes on Worf and Quine

In his introduction to The Search for the Perfect Language, Umberto Eco writes:

"Since language expresses the modes which organize the way we categorize and classify reality, natural languages must be considered holistic systems. They organize the totality of our vision of the world. It has sometimes been suggested (Whorf 1956, Quine 1960 for example) that there are experiences, recognized by other cultures and capable of being expressed in their languages, which are neither recognized by our own, nor even capable of being expressed in our languages. Although this is a rather extreme view, we will continually be finding ourselves faced with it as we examine the criticisms leveled at the various projects for a perfect language."

I am still reading the book. I don't know if Eco will take a final position on the idea that different natural languages might refer to entities with which the users of that language are comfortable and take for granted (and fit in with the vision of the world that the language organizes) but which are impossible to translate into (or to talk--clearly, very clearly?--about in) other languages, a position which he here calls "rather extreme". (And he may not have to, since his self-described task is to document and bring alive the long series of failed attempts to create a perfect language, and the attacks made against those efforts, one of which is the extreme claim he here refers to :)

But this is the direction that I am headed with this blog: to see to what extent we can sensibly claim that parts of used Khmer language may be impossible to translate into English. There is a whole lot of brush that needs clearing first before we can ease ourselves toward this question. We need to do a lot of work to say what we are interested in and what we are not, and, in the best of all possible worlds, to develop a sound methodological basis on which to make concrete our findings.

But a short sketch of what I have in mind is perhaps useful. I am interested in connecting three things within the Khmer culture, and then to see to what extent that can be carried over into English. I will start with the visible, the empirically verifiable: action and the self-descriptions of that actions and entities. The physical action may be easily described, one might think of casting a ballot in the national elections (we do that in the United States, Cambodians have that right and privilege also). There may be entities which are named in Khmer which are not named, of have no existence according to American English, though they may often be amenable to description (a kind of interpretation if not translation), by indicating distinctions that are made one language and not in another (for example the word k'moaic in Khmer is used to talk about ghosts and corpses). Where we may run into problems is with the self-descriptions and the cultural surround (beliefs and commitments of the culture) that make the actions (or named entities) meaningful.

For example, a Khmer woman goes to the temple. She carries lotus buds and incense sticks and prepared food. What is she up to? She can describe her actions and motivations in a way that allows us to get the gist in English (or maybe not):

"We are in the middle of an important annual festival period, Bon Kathin. I am bringing food to appease the hunger of my dead relatives (some of which the monks will eat and for which I will receive merit). My dead relatives have been released for a short time from the Buddhist hell where they have been starved. I hve fourteen days to do this in and if I can make it to seven different temples, each on seven different days, I will have a very, very good chance, of giving sustenance to my dead relatives by scattering some of the food around the temple in the pre-dawn hours."

That is what she might say. The normal English translation of Bon Kathin is singularly unhelpful: 'Ghost Festival' (and of the relatives: 'Hungry Ghosts'), but that is not the central issue here. Just because some English-language re-descriptions are unhelpful, does not mean that there are none that are helpful:)

In this example we have pointed to two empirically available entities: the actions that this lady engages in (preparing food and herself, her travel to the temple and its timing, the scattering of food on the ground) and the descriptions she might give of her motivations and her own actions (the language that she uses).

In turn, these words and phrases, have a larger cultural and religious surround, they are connected, logically and by semantic deduction from and to a set of beliefs and doctrines, doctrines that she may not herself be able to articulate in their fullness (how the issues of merit, reincarnation, the cycle of birth and death might play into what she is doing, she may do it simply because she used to do it with her own mother, now dead) but which we can gain some access to, over time, by talking to her, to others engaged in the same activity, and to the monks.

Similarly, a woman lighting candles in memory of her mother in a Catholic Church might not be able to call to mind all of the Catholic doctrine that is implicated in that single act. But the surround is a necessary backdrop and context to the language that uses (and of the behavior that she displays) when we ask her what she is up to, and without which backdrop the act which she engages would not have the meaning that it does for her, however well or poorly she is able to articulate it.

Even if we had what she would say was a full and complete description of what she was up to in this Catholic sanctuary, how easily would she be able to explain this to a Southern Baptist (for example). The reaction might not be simple belief or disbelief, agreement or disagreement, denial or accord. It might be astonishment, a complete inability (even if coupled with a willingness to try) to throw oneself into that doctrinal world to see what sense of it one might make and how well and poorly it might accord with the religious world which one inhabits as an engaged and fulfilled Southern Baptist. One might still find a very poor one-to-one mapping of doctrines and beliefs.

(You could be asking: "Would that matter?' Or, "Who cares?" (Or maybe: "Hasn't this been done before?" -- "Like in Anthropology?" I won't answer that ;) But I take the first two questions as a challenge not only to see what sense might be made of these investigations, but to do so in a way that might compellingly show that it does matter.)

I don't think we have to look at differences between U.S. English and Cambodian Khmer to see these kinds of translation problems. In this modern world, despite its apparent growing homogeneity (which on surface seems so compelling, my dark vision is an unending sequence of strip malls, fast food restaurants and gas stations, coast to coast :) is increasingly fractured, with powerful communities tied together in sub-cultures with their own languages (scientists, academicians, political conservatives and liberals, hobbyists of every stripe), in some sense unable to talk to each other about issues and questions central to them across growing divides.

I don't think that this is just a culture to culture thing, where we think of only cultures that built and sustained on the backs of particular natural languages. I think that it is a growing fact of a modernizing world that growing cultural homogeneity continues to collapse into a mosaic of speech communities who share the same major language, but which on some issues find it impossible to talk to each other (one might think that an obvious example of this is the inability proponents of abortion and their opponents have of talking to each other but I am not so sure, because I am not so very sure that they live in such ethically and morally different universes) .

I still remember the shock I received when I first understood the phrase in economics "sunk cost." I think that my first naive reaction to the phrase was that it was similar in meaning to "investment" but I did not understand the implications for (rational) economic behavior of the meaning of this word to (classical) economists.

My professor asked on a test: "True or False: A rice mill owner pays $10,000 dollars for his rice mill. Treating that as a 'sunk cost,' he should count that against the money he makes running the mill (that is he should calculate his profit taking into account this cost)."

I agreed and was marked wrong. My teacher said that a sunk cost should never be counted against profit or be part of the basis for any future decision making (he never used the phrase, as far as I remember, "recoverable costs"). At that moment, I felt I had entered another world. I had passed from a world where if I build a lemonade stand, I work hard to recover that investment and then to make a profit on top of that to a world where, ostensibly, the cost of the lemonade stand was had no bearing on how I thought about my business and whether I should stay in it or not. Still don't get it, actually. But I recognize that this word, the entities and actions it refers to, help to consitute a branch world of economic activity that would be difficult to reconcile with the more naive one that I still inhabit.

I think that what I am after is a way to talk about and think about differences between langauges and the world's which they constitute where straight-forward translation or redescription still do not get us to a point where we are satisfied we know what is going on, what is stake in a particular usage, where our questions are exhausted. We will want to see what issues are raised for the possibility of translation if we take natuaral languages in Eco's sense: They express "the modes which organize the way we categorize and classify reality". We must consider them "holistic systems".

Read more...