The English and French languages certainly don’t agree on how the world should be broken up into categories, even for the nouns of the highest frequency that exist, let alone for categories labeled by verbs, adverbs, prepositions, and so forth. For example, those incorrigible French speakers, they irrationally distinguish between two kinds of “in” — namely, “dans” and “en”. What could make less sense than that? Whereas we clear-sighted English speakers, we distinguish (most rationally, of course) between two kinds of “de” — namely, “of” and “from”. What could make more sense than this?
These kinds of discrepancies are totally typical of how different languages carve the world up differently from each other, and between any given pair of languages there are myriads of such discrepancies. How, then, do people ever communicate at all across language boundaries?
Spaces Filled Up with Concepts
To help answer this question, we would like to offer a simple visual metaphor for thinking about the words of a language (and more generally about lexical expressions) and the concepts that they represent. We begin by suggesting that you imagine a two-dimensional space or a three-dimensional one, as you prefer; next, we are going to start filling that space up, in our imagination, with small patches of color, using a different color for each different language that we are interested in — say green for French, red for English, blue for German, purple for Chinese, and so forth. It is tempting to think of these concept-blobs as something like rocks or jelly beans — odd little shapes having very well-defined edges or boundaries. The truth is far from that, however. While each blob is intensely colored in its center (deep red, deep green, whatever), as one approaches its “boundaries” (which in truth don’t exist), it grows lighter in shade — think of pink or chartreuse — and then it simply fades out, passing through lighter and lighter pastel shades as it does so. This image of blobs with hazy contours of course echoes our metaphor likening concepts to very dense cities that gradually turn into suburbs and then fade into countryside.
We will call the space itself, before the insertion of any colored blobs (somewhat like a house without furniture), a “conceptual space” (there are many such, which explains the indefinite article). At the very center of each conceptual space are found the most common kinds of concepts — those for very common tangible objects, intangible ideas, phenomena, properties, and so forth — the concepts whose instances are encountered all the time by people who belong to a particular culture (or subculture) and era, and which those people must be able to categorize quickly and effortlessly in order to survive, or simply to live.
The core items in a typical conceptual space include, quite obviously, the concepts for various entities such as the main parts of the human body; general classes of common animals, such as bird, fish, insect, and a few farm animals; general classes of plants, such as tree, bush, and flower; things to eat and drink; common feelings, such as being cold or hot or hungry or thirsty or sleepy or happy or sad; common actions, such as walking and sleeping and eating and giving and taking and liking and disliking; common properties, such as big and small, near and far, kind and cruel, edible and inedible; common relationships, such as belonging to, being inside or outside, being above or below, being before or after; common degrees, such as not at all, not much, slightly, medium, very much, totally — and so forth. Every language has words for such notions, because all humans require these concepts in order to live. This list merely scratches the surface of the core of a typical conceptual space, of course, but it gives the general idea. In any case, these concepts, all residing at or very close to the dead center of a typical conceptual space, are quasi-universals that most humans deal with constantly, and they are thus bases that are well covered, and necessarily so, by every language.
The idea of conceptual spaces will help to make more tangible and concrete some ideas about the words and expressions of a given language and the concepts used by its speakers. One of the most important ideas that it helps one to think about is how different languages cover, or fail to cover, certain concepts. Between the conceptual spaces of distant cultures there will be large discrepancies. But what about cultures that are bound together by geography, history, traditions, and so forth? In such cases, the conceptual spaces will be very close to each other.
In what follows, we will focus mostly on contemporary Western cultures, simply because we ourselves feel more competent in that context, and we assume that many of our readers (at least those who are reading this book in one of its two original languages) would also feel more comfortable that way. However, our general points have nothing to do with the specific concepts that we will discuss.
Looking at Two or More Languages within a Conceptual Space
How is a conceptual space filled up with sets of blobs of different colors? For instance, how do the repertoires of concepts possessed by French and English speakers who share essentially the same culture compare?
According to our visual metaphor, regions near the very center of a conceptual space are densely filled in, no matter what language we are speaking about. If, as suggested above, French is represented by green, then there is a green blob near the middle of conceptual space that covers the area occupied by the concept hand. And if English is represented by red, then fairly much the same area is covered by a red blob of similar size and shape to the green (French) blob. Each different language will cover that same area of a conceptual space fairly well, so there will be blobs of many different colors right there, all closely overlapping with one other.
Some of the different-colored blobs representing different languages’ coverages of a given extremely frequent concept will tend to have pretty much the same shape, but in the case of other blobs there will be discrepancies, some minor and some major. We’ve already seen a pretty major one, involving time corresponding to heure, temps, and fois, and temps corresponding to both time and weather. To provide another case, the red blob representing the extremely frequent concept expressed by the word “big” in English aligns quite well with the green blob for the French word “grand”, but by no means perfectly so, since some of the meanings expressed by our “big” are usurped by French’s “gros” (for instance, things that are large in thickness or width, as opposed to those that are large in height), and conversely, our word “great” usurps some of the meanings expressed by French’s “grand” (those that mean “highly accomplished, world-famous, and deeply influential”).
Some even more severe misalignments involve extremely frequent prepositions such as “in” (which in fact is covered in French not just by “dans” and “en”, but by many other prepositions, depending on the context), and by similarly frequent and enormously protean verbs such as “to get” (which sometimes is best rendered by “obtenir”, other times by “prendre”, other times by “chercher”, other times by “recevoir”, other times by “comprendre”, other times by “devenir”, other times by “procurer”, and on and on). Of course, the story is symmetric; that is, each of the just-mentioned high-frequency French prepositions and verbs is likewise covered by all sorts of different English verbs, depending on the context. There’s no clean one-to-one alignment between blobs of different colors, although there’s a great deal of overlap.
On the other hand (and quite luckily!), for a very large number of truly important concepts — say, finger, water, flower, smile, weight, jump, drop, think, sad, cloudy, tired, without, above, despite, never, here, slowly, and, but, and because, to give just a few examples — there is generally quite good agreement between French and English, and, for that matter, among all the languages that we are familiar with.
Thus, the center of this conceptual space is inhabited by red and green blobs that often coincide quite well, and when they don’t coincide, then there are all sorts of overlapping blobs, each with its own curious shape. Luckily, though, despite the fact that the green blobs covering a certain concept and the red blobs covering the same concept are often shaped rather differently,
the central zone of the overlapping space is extremely densely covered both by red blobs and by green blobs (and also, if we want to throw in other languages, blue blobs and purple blobs, and so forth).
Furthermore, there aren’t going to be any gaping holes in the linguistic coverage of concepts residing near the dead center of the conceptual space of some other culture (such as the Nepali or the Navajo culture); there won’t be blank zones where a human language totally lacks a lexical item labeling a concept that is universally part of the human condition. Any language spoken by more than a tiny, isolated group will easily be able to talk about, for instance, sleeping poorly, or seeing a friend after a long time, or breaking a stick, or throwing a stone, or walking uphill, or feeling sweaty, or being very tired, or losing one’s hair, though each one will have a unique way of doing so.
Rings or Shells in Conceptual Space
Let us now imagine moving outwards from the core towards slightly less frequently encountered concepts, such as, for instance thanks, barn, fog, purple, sincere, garden, sand, star, embarrassing, roof, and although. If these concepts are of comparable importance to one another within the culture, then their distance from the center will be about the same, and we can say that they constitute a ring (or a shell, if you are envisioning a three-dimensional space). These concepts are still important in the conceptual space — and so, once again, we expect that this region of conceptual space, though not belonging to the most central core, will still be quite densely filled with blobs of every color. On the other hand, let’s zoom outwards a considerable distance further from the core of our conceptual space, to a different shell where we will encounter (let us say) the concepts frowning, cantering, fingernail-biting, tap-dancing, welcome home, income tax, punch line, corny joke, sappy movie, vegetarian, backstroke, chief executive officer, wishful thinking, sexual discrimination, summit meeting, and adverb (just to give a tiny sampling of the hypothetical shell). This latest outward leap has clearly carried us into more rarefied territory, and so we would not expect all the cultures of the world (and of all different historical epochs) to share all the concepts in this shell, nor would we expect all the earth’s languages to have words or phrases to denote all the concepts in this shell.
What is Monolithic is in the Eye of the Beholder
Let’s take any shell of this conceptual space. Since languages differ enormously, we can easily find a red blob that no single green blob covers precisely. However, a small set of green blobs will collectively do a pretty good job of covering all the territory of the red blob (although they will inevitably also cover areas outside the red blob). And of course, what’s sauce for the goose is sauce for the gander, meaning that we can easily find green blobs that no single red blob covers precisely.
To make things concrete, let’s take an example. English speakers fluently and effortlessly use the word “pattern” to describe regularities, exact or approximate, that they perceive in the world. However, if they wish to talk about such phenomena in French, they will soon learn, to their frustration, that there is no French word that exactly covers this very clear zone of conceptual space. And thus, depending on details of what they mean, they will have to choose among French words such as “motif”, “régularité”, “structure”, “système”, “style”, “tendance”, “habitude”, “configuration”, “disposition”, “périodicité”, “dessin”, “modèle”, “schéma”, and perhaps others.
At the outset, this lacuna in the French lexicon strikes English speakers as a rude violation of common sense, since the concept of pattern strikes them as being self-evident and objective, and therefore something that should be universal to all languages. It seems obvious that there “should” be just one word for all those notions that the English word unites; after all, it feels like just one thing rather than many. But in French and in fact in most other languages, there simply isn’t such a word. Nonetheless, other languages manage to cover the zone of conceptual space labeled “pattern” in English pretty completely, although somewhat less efficiently, by using a bunch of smaller blobs each of which corresponds to a limited facet of the notion, or else a set of large blobs that intersect partly with the English one.
For the sake of fairness, we should point out that French, too, has words of quite high frequency that have no counterpart in English — for instance, the adverb “normalement”, which certainly looks like it means what we anglophones mean when we say “normally” (and sometimes it indeed does), but which a large part of the time means something rather different. Here are a few examples that show typical uses of the word, and that give a sense for the wide variety of translations it needs in order to be rendered accurately in English:
Normalement, Danny doit être arrivé à la maison maintenant.
Hopefully, Danny’s back home by now.
Normalement, on va courir à 7 heures ce soir, non ?
Unless we change our plans, we’ll be taking our run at 7 this evening, right?
Normalement, nous devions passer deux semaines en Bretagne.
If there hadn’t been a hitch, we would have spent two weeks in Brittany.
French speakers will be just as puzzled by English’s lack of a single word for the obvious, monolithic-seeming concept expressed by the word “normalement” as English speakers are puzzled by French’s lack of a single word for the obvious, monolithic-seeming concept that is embodied in the word “pattern”. What is monolithic is in the eye of the beholder.
In cases such as these, where one language has a single word that covers a set of situations that another language needs a variety of different terms to describe, we are dealing with linguistic richness and poverty. Thus in the case of “pattern”, English is richer than French, and in the case of “normalement”, French is richer than English. More generally, we can say language A is locally richer than language B if language A has a word (or phrase) denoting a unified concept — that is, a concept that native speakers feel hangs together tightly, and that seems to have no natural internal cleavages — and if language B lacks any single word covering that same zone of conceptual space. We can thus speak of a local “hole” or “lacuna” in language B’s coverage of conceptual space, even though language B manages to cover the zone by resorting to a set of words.
On the other hand, when a certain area of conceptual space is finely broken up by a given language, and when speakers of both languages agree that this fine break-up is warranted, then a language that doesn’t offer its speakers such a fine break-up has to be considered poorer. Take the English word “time”, for instance. To native speakers of English, whereas the word “pattern” feels unitary and monolithic, the word “time” does not have that monolithic feel; native speakers readily and easily see (at least if it’s brought to their attention) that there are several very different meanings of “time” (for instance, those corresponding to the French words “heure”, “temps”, and “fois”). Thus in this case, it’s the French language that is richer and the English language that is poorer, for the English lexicon doesn’t break that large zone of conceptual space into smaller separate zones, as the French language does. An example where French is weaker is the word “beaucoup”, which corresponds to both “much” and “many” in English. For us anglophones, it’s obvious that these are separate concepts, one having to do with a large quantity of a substance, the other having to do with a large number of similar items. The French word that blurs this distinction thus seems rather crude. Thus in this case, the English language appears to be richer, and French poorer.
In summary, when language A has a word that strikes its speakers as representing a natural and monolithic concept, and language B has no corresponding word, then language B is poorer and language A is richer, because speakers of language B are forced to cobble different words together in order to cover the zone of conceptual space that language A covers with just one word. Conversely, when language B has a set of words that cut up a zone of conceptual space that is covered by just o
ne word in language A, and when the distinctions offered by language B seem natural to speakers of both languages, then it’s language B that is richer and language A that is poorer.
The Need to Stop Subdividing Categories at Some Point
When one studies various languages, one discovers that many concepts that one had at first naïvely taken as monolithic, because of one’s native language, are in fact broken up into subconcepts, and often with excellent reason, by other languages. And if one studies enough languages, one often discovers numerous different ways of subdividing one and the same concept. Seeing a concept being broken up into all sorts of subconcepts that one hadn’t previously dreamt of suggests that it would in theory be possible to continue carving the world up into tinier and tinier blobs, thus making an ever finer mesh of very small, extremely refined concepts, without any end.
Surfaces and Essences Page 15