For a postscript to this post, go here.
I believe there is much to be learned philosophically from the study of languages that are spoken by only a small number of people, who lack a high degree of political self-determination and are relatively powerless to impose their conception of history, society, and nature on their neighbours; and who also lack much in the way of a textual literary tradition or formal and recognisably modern institutions of knowledge transmission: which for present purposes we may call “indigenous” languages.
This is of course going to be a hard sell, given that the great majority of Anglophone philosophers do not even recognize the value of learning German, Latin, Arabic, Sanskrit, or Chinese, and believe that they can penetrate as deeply as one might possibly go into fundamental philosophical questions from a standpoint of monolingualism. German, Latin, and the others are cosmopolitan languages, and historically all cosmopolitan languages, rightly or wrongly, have functioned as vehicles of what most discerning people are prepared to recognise as philosophy. But there is a significant difference even among the five cosmopolitan languages I’ve listed, which can begin to point us towards the even greater difference between all five of these, on the one hand, and, say, Yanomami, Ainu, Ket, or Sámi on the other: the first three cosmopolitan languages may be grouped together, as having a long legacy of shared and standardised terminology such that problems of translation between them are relatively small; by contrast, while there is to some extent a legacy of translation from Sanskrit towards Chinese, often via Tibetan, for the most part philosophical terminology has developed in these languages independently and without any felt need to establish cross-linguistic equivalencies.
Now, Yanomami, Ainu, Ket, and Sámi are generally held to be non-philosophical languages, languages from which one must depart if one wishes to start doing philosophy, and one of my contentions is that this judgment of them goes together with their status as non-cosmopolitan languages: that is, philosophy, throughout its long history and very much still today, is presumed to be an activity that may be pursued only in languages that may pretend to universality.
What is it like to speak a language that may not pretend to universality? Most philosophers in the English-speaking world have never considered this question. There are of course degrees of cosmopolitanness and indigeneity, and if over the past, say, two centuries, Arabic and Chinese speakers have been compelled to think somewhat more about the relative non-universality of their languages compared to English (as illustrated for example by the fact that if you want to operate at the highest institutional levels in global academia, you must do so in English), they nonetheless have a foundation and an orientation in native languages that in the past have been, and perhaps in the future will be, deemed adequate for expression of any philosophical idea whatsoever.
The same is not true of Yanomami, Ainu, Ket, and Sámi. From the very first moment of these languages’ contact with the cosmopolitan or imperial languages of the dominant cultures that engulf or abut them (Spanish, Japanese, Russian, and Scandinavian respectively), there is an immediate and total separation between the sort of things that can and should be expressed in the indigenous language, and the sort that can and should be expressed in the cosmopolitan language.
I myself have been studying Sakha, a Turkic language of Northeast Asia, for the past two years, and have occasionally written about my progress here. Before starting to learn Sakha, I had studied, with varying degrees of success: French, German, Russian, Latin, Greek, Sanskrit, Old Church Slavonic, and Turkish (as well as dabbling in many other languages while making no real progress). All of these languages are paradigmatically cosmopolitan (and all but one are Indo-European). Never, before beginning to study Sakha, did I have much insight into the way contingent features of the natural languages we are using condition the manner in which we enter into and seek to resolve philosophical questions.
Sakha presents a complicated case, and much could be said, and disputed, as to its indigenous, non-cosmopolitan status. I would call it ‘relatively non-cosmopolitan’: if you are a native Yukaghir speaker in the Sakha Republic, you will likely have to learn both Russian and Sakha in order to make your way through the world, and in this respect from a certain linguistic starting point Sakha has the quality of a lingua franca. Unlike a paleo-Siberian isolate such as Yukaghir moreover, Sakha is a geographically remote fragment of a major world language family that also includes the lingue franche of the Ottoman Empire and the modern Turkish Republic. And it is a language that was significantly conditioned by past interactions with the Mongol Empire, and, since the 16th century, with the Russian Empire (and this latter is an empire in its Asian extension, even if few people today think to question the territorial integrity and sovereignty of the Russian Federation: this is the virtue of pursuing your colonial expansion without an ocean separating the metropole from the colonies).
I mean to say that there are other languages that are much more local and non-cosmopolitan than Sakha; I chose Sakha in part because the majority of the world’s languages lack resources for self-directed learning; if I wanted to learn Yukaghir, in contrast with Sakha I wouldn’t have the option of learning from news reports or ingenious music videos such as this on YouTube. But still, the relative non-cosmopolitanness of Sakha is sufficient to reveal to me an enormous epistemic inequality between the languages in which philosophy is by default conducted, on the one hand, and the great majority of the world’s languages on the other. This inequality, I believe, generates a bias so enormous that all other concerns about the lack of diversity in philosophy seem trivial by comparison. The deepest failure to incorporate diverse philosophical viewpoints is the failure to notice the existence of indigenous life-worlds, and to pay attention to what distinguishes them epistemically from the cosmopolitan ones.
Let me try to work my way into why this is so important with a few examples that might appear mundane. It was recently pointed out that Timothy Williamson --whom I take, perhaps unfairly, to be an example of an Anglophone philosopher who does not even recognise that you would do well to know German and Latin, let alone Sanskrit and Chinese, let alone, in turn, Yanomami and Ainu, in order to do philosophy better-- considers the sentence ‘Vixens are female foxes’ to be a paradigmatic analytic truth. In part prompted by examples such as these, I recently wrote a short post explaining why gendered animal terms in fact require complicated knowledge of the social saliency of domestic breeding, which is something that only pertains to a limited number of animal species, and to different species in different human cultures. Given this social dimension to such terms, it is by no means a simple fact of the world outside us, I argued, that ‘vixen’ denotes the female of the various members of the Vulpes genus.
So let us try to find a truly non-controversial example of a truth-claim, expressed in English, about the world. We may as well turn, as a first stab, to that old staple of analytic philosophy: The cat is on the mat. I’m not sure who first used this example, though I’m sure this would be easy to find out, but I know I saw it written on countless chalkboards throughout my time in graduate school. It is not an analytic proposition, like ‘Vixens are female foxes’, but is rather deployed in discussions of denoting: “The cat is on the mat” is a claim that seems to involve at a minimum existential commitment to a particular cat and a particular mat. But let’s leave aside though the problem of the definite articles, which not every language has (Sakha does not have them, and neither does Latin or Russian: I once saw a funny translation of Bertrand Russell’s “On Denoting” into the latter language, in which his nuanced philosophical analysis of the definite article simply left the English word ‘the’ untranslated. What other choice did the translator have?).
Leaving aside the ‘the’ problem, this sentence, chosen as paradigmatic in view of its supposedly non-controversial character, in fact becomes laden with problems the further we stray from the language in which it was first deployed as an example, most likely by a monolingual Anglophone philosopher. In order for this sentence to begin to make sense, for one thing, presuming that by ‘mat’ we mean ‘doormat’ (as I’ve always understood it), we must suppose that its speaker or hearer is someone who lives in a culture whose members wear their shoes when they enter into a house, and must therefore wipe the soil off of them as they enter. But a comparative anthropological perspective reveals to us that this is in fact a very strange and deviant thing to do, not quite as rare as, say, communal defecation, but up there in the general neighbourhood. Now we might translate with the English word ‘mat’ a given word in another language for something that has the bare form of a Euro-American ‘doormat’, but we will almost certainly find that its range of functions in the culture from which the word comes would strongly compel us to simply retain the native word, if we did not wish to allow our understanding of the object in question to be distorted by all the connotations we allow to attach to the English word in question.
Now it might sound like I’m being facetious with this example, but the only way I know to defend against that accusation is to double down, and to continue to draw out this paradigm sentence’s further problems. We’ve already problematised mats, so what about cats? A first thing to note is that the Sakha word for ‘cat’, куоска / kuoska, is not like the Sakha word for ‘horse’ (ат / at) (of Turkic origin), ‘dog’ (ыт / yt) (Turkic), or ‘deer’ (таба / taba) (Tungusic). Куоска is rather a Yakutized form of the Russian кошка / koshka, which is cognate to countless other Indo-European words for felines, such as chat, gato, Katze, and cat. While horses, dogs, and deer constitute central features of Sakha social reality from the very beginning of their existence as a culture, quasi-domestic felines only arrived in Northeastern Asia with the arrival of Russian colonists, and so the word for ‘cat’ marks in Sakha a moment in the history of this culture’s political subjugation.
The Sakha translation of this paradigm sentence, then, is (perhaps):
Куоска муоста тэлгэҕигэр баар.
Kuoska muosta telgeğiger baar.
Cat floor['s] mat(on) is.
Not bad, for a start. We have, here, a quasi-domestic mammalian species spatially located on top of a human-made artefact, much like what is meant by the English sentence, “The cat is on the mat.” This quasi-domestic species however is an invasive one, and goes by a foreign name; it has not traditionally been subject to spaying and neutering, and so (unlike dogs, horses, and domestic reindeer), reproduces as it pleases, and often to the detriment of human food supplies. And it is sitting on top of something, a ‘mat’ as we have attempted to render it, the name of which implies woven grass, and thus something far closer to the natural world than a ‘mat’ made out of, say, petroleum by-products. Муоста тэлгэҕигэр is a compound phrase in the possessive dative (which is how we get the cat onto it); if we change this back into a nominative possessive, we have муоста тэлгэҕэ, ‘mat of the floor’ or, as we would say in English, ‘floor’s mat’, or, more usually, ‘floor-mat’ or ‘floor mat’ or even ‘floormat’.
Муоста by itself also means ‘bridge’, and in this meaning it is also a borrowing from Russian, namely the word мост. I am not sure --and my authoritative Sakha-Russian dictionary, by the great Russian linguist E K. Pekarskiï, does not say-- whether in its meaning as ‘floor’ the word муоста is a simple homonym of муоста as ‘bridge’ (like ‘bark’ and ‘bark’ in English), or whether in its meaning as ‘floor’ муоста also comes from the Russian мост. If Sakha had to borrow a word for ‘bridge’, it is possible that it also had to borrow a word for ‘floor’ in the special sense that this conveys in the broader Slavic and Indo-European world. Even something so apparently basic as a floor is not a pre-given feature of external reality; it is something that takes shape as a result of the way we live, and of the way we conceptualise our habitations and what lies beyond them.
The Sakha ‘equivalent’ of “The cat is on the mat,” then, involves at least one, maybe two, entities that are imported from another linguistic reality, while a third entity, the ‘mat’ without any special reference to the floor, bears a much closer relationship to nature than the pure artefact denoted by the English ‘mat’. There is, again, no way to reproduce the occurrences of the English ‘the’, and so in the end it is only context in Sakha that will tell us whether the speaker does not in fact mean that a cat is on a mat (but, again this is a problem we also have in Russian and Latin). And there is no preposition ‘on’, either, so instead we deal with that by means of nominal case endings. And as for ‘is’ being captured by the Sakha баар (cognate with the Turkish var, which is often described as functioning more as an existential quantifier than as a verb), well, that’s another enormous question I am not even going to touch.
Instead I want to dwell on the difficulties inherent in the search only for cross-linguistic equivalences in nouns. Of course, many words in many languages, even in cosmopolitan ones, have a similar legacy, and it is obviously no mark of the subaltern status of the United States relative to Italy that we refer to its form of government as ‘republican’, even though res publica may be traced back to the Italian peninsula. But in Sakha the lexical dependency on the language of a still-existing external political hegemon for any dimension of language dealing with politics, science, technology, economics, business, extraterritorial zoology and botany, global popular culture, and so on, and so on, makes the situation very different from one such as that of English, which shows the historical marks of past domination (by, e.g., Romans, and then Normans), without there being any pervasive awareness in the use of modern American English that we are relying on words derived from other empires.
My point here is not a simple reheating of Sapir and Whorff. It is not, or not just, that different social realities will make different features of the world salient enough to be captured in words, and thus that one culture will have a thousand words for different qualities of snow, and so on. Sapir and Whorff (not to mention Quine) were not particularly interested in the political dimensions of untranslatability, and did not spend much time reflecting on the vast difference between the political orders served by English and Inuktitut respectively, nor on the way this service shapes the sense of what may be said in these languages.
It is well known that a core feature of modern nationalisms, beginning in Europe in the 18th century and radiating out through the 19th and 20th centuries, was to raise up national languages, by means of dictionaries, academies, and the often artificial elevation of revered national poets, into the vehicles of expression of any thought a human being is capable of having, that is, into languages that may pretend to universality. Thus in the 14th century one could not describe military strategy or the immortality of the soul in Lithuanian, but by the 19th century, one could. Yet this has mostly been a failed project, as today, notice, again, that one cannot really enter the global discussion of, say, genetic engineering or the sovereign debt crisis, in Lithuanian. In the case of Sakha, there was a brief efflorescence of top-down, imported Soviet linguistic modernisation, in which attempts were made to create a Sakha-language version of the Homo sovieticus complete with daily newspapers and dull socialist-realist novels and plays. But returned mostly to the forces of global market ideology and official indifference to matters of culture, the future vitality of sub-state, nominally “autonomous” national languages such as Sakha within the former Soviet Union is very much in doubt.
And Lithuanian and Sakha are doing pretty well, relatively speaking. The vast majority of languages in the world are such that their native speakers must switch to another language in order to talk about any of the things that interest the sort of people who pass through airports, have credit cards, or go to philosophy conferences. And this is the default form of linguistic competency in the world, and has been ever since ancient empires arose and put some people's cultures under the political domination of others. To have to reach outside of your native linguistic resources in order to engage with ideas that are supposed to be accessible in a linguistically neutral way is an effort that English speakers par excellence, and speakers of any cosmopolitan language to a somewhat lesser degree, do not have to make. And the lack of any need to do so makes them systematically unaware of the cost of having to do so.
Language is not neutral, and “The cat is on the mat” is not a claim whose meaning is by any means obvious. This is one, and only one, of the lessons that the study of indigenous languages affords: to expose the unconsciously provincial character of philosophy done in languages that have the luxury of taking themselves for universal.
I feel like you had a great argument to make here and then by reducing it to “the cat is on the mat” it all boiled down to “but not every culture has independently encountered every noun” which I don’t know is that valuable? This comes up fairly commonly even between the major languages of today (guanxi is an easy example of a concept, kangaroo an easy example of an animal). Perhaps I’ve just missed out on the experience of formal philosophy classes and thus don’t have an axe to grind against “the cat is in the mat”.
If you want to understand if the word “cat” matters in a philosophical sense then go talk to some indigenous Australian tribes who encountered invasive species independently of European contact.
I’m surprised you didn’t explore the possibilities of the word “on” as my experience in learning languages (much more limited than yours) is that it is in these types of words where the real ambiguity often occurs.
Posted by: BK | February 9, 2020 at 12:14 PM
Isn't this just Wittgenstein's language game? He manage to figure this out speaking only German & English.
"The cat is on the mat" can mean the suspect has arrived at our hideout, if the language you're speaking is spy-thriller cliche English which is even further from regular English than Куоска муоста тэлгэҕигэр баар.
You can talk about military strategy in the 14th century Lithuanian as long as you're willing to set up the language game.
Posted by: Kieron George | February 9, 2020 at 02:31 PM