Surfaces and Essences

by Douglas Hofstadter

  Let’s take another very simple sentence in English:

  The boy and the dog were eating bread.

  This sentence is nonzeugmatic in English; that is, it simply works, sounding neither strange nor humorous to the English-speaking ear. On the other hand, it sounds wrong in German, because different verbs apply to animal and human ingestion — “fressen” for the beasts, and “essen” for humans. In other words, German speakers split up what to us anglophones is the monolithic concept of eating, breaking it into two varieties, according to the type of creature that is carrying out the act.

  The “Natural” Conceptual Distinctions Provided by Each Language

  These examples might inspire someone to imagine a language (and culture) that has no verb that applies both to men and to women. Thus it would have one verb that would apply to eating acts by men and a different one that would apply to eating acts by women — say, “to wolf down” for men and “to fox down” for women, as in “Petunia foxed down her sandwich with relish, gusto, and pickles”. Speakers of this hypothetical language would find it jolting to learn that in English one can say, “My husband and I enjoy eating the same things” or “A girl and a boy were walking down the sidewalk.” To them, such sentences would sound nonsensical. A language like this may strike you as ludicrous, but many languages do make just such gender-based lexical distinctions.

  For instance, in French there is a clear-cut distinction between enjoyment partaken of by men and enjoyment partaken of by women, which shows up in, among other venues, the standard adjective meaning “happy”: whereas a joyous man or boy will be “heureux”, a joyous woman or girl will be “heureuse”. And thus, a curieux French male might well wonder what it feels like to be heureuse — but he would do so in vain! A man simply cannot be heureuse! In like manner, a curieuse French woman might wonder what it feels like to be heureux — but her efforts, no matter how valiant, would be doomed to failure. A Venusian might as well try to imagine what it feels like to be Martian!

  Does all this sound far-fetched to you? Well, consider that there is a famous Russian poem centered on what the poet, a man named Il’ya L’vovich Selvinsky, considered a very strange fact: namely, that every act of his lover — every single one of the mundane verbs that described her actions — was graced, when in the past tense, by a feminine ending (often the syllables or bisyllables “la”, “ala”, or “yala”). The poet describes various completely ordinary actions on her part (walking, eating, etc.), and then expresses wonderment at his own feeling of disorientation, since he, being a male, has never once performed a single one of these “uniquely feminine” acts, nor experienced a single one of these “uniquely feminine” sensations, and, alas, will never be able to do so. In making such observations, is Selvinsky expressing something deep, or is he merely playing with words?

  One can easily enough imagine a language that, with a panoply of verbs, distinguishes between a vast number of different ways of eating — the eating of a famished boy, of a high-society lady, of a pig, a horse, a rabbit, a shark, a catfish, an eagle, a hummingbird, and so forth and so on. Such a fine-grained breakup of a concept that seems to us completely monolithic is perfectly imaginable, because we understand that there are genuine differences between these creatures’ ways of ingesting food (indeed, if there weren’t any, we would not have written “genuine differences”). Each language has the right and the responsibility to decide where it wishes to draw distinctions in the zone of semantic space that includes all of these distinct activities. After all, there are not, on earth (and never have been, and never will be) two creatures that eat in an exactly identical fashion, nor even two different moments in which a single creature eats in exactly the same manner, down to the tiniest detail.

  Every act is unique, and yet there are resemblances between certain acts, and it is precisely these resemblances that give a language the opportunity to describe them all by the same label; and when a language chooses to do so, that fact creates “families” of actions. This is a subtle challenge to which every language reacts in its own fashion, but once this has been done, each group of people who share a common native language accepts as completely natural and self-evident the specific breakdown of concepts handed to them by their language. On the other hand, the conceptual distinctions that are part and parcel of other languages may strike them as artificial, pointlessly finicky, even incomprehensible or stupid, unless they find some interest in the subtleties of such distinctions, which may then make them see their own set of concepts in a fresh light.

  Wordplay with the Word “Play”

  The verb “to play” affords us a delightful sampler of zeugmas, or else, depending on a person’s native language and on their own personal way of perceiving the actions involved, non-zeugmas. For example:

  Edmond plays basketball and soccer.

  This sentence, on first sight, might seem about as natural as they come, and very far from zeugmaticity, and yet the two activities involved, although they both belong to the category of sports, are different in numerous ways from each other. For instance, one involves a ball that is primarily in contact with the feet (and on occasion with the head), while the other involves a ball that is primarily in contact with the hands (and virtually never with the head). Certain speakers of English might therefore hear a trace of strangeness, albeit only very slight, in the application of the same verb to two rather disparate activities.

  If essen (which is what people do when they eat food) and fressen (which is what, say, pigs and rabbits do with their food) are seen by German speakers as activities that belong to two different categories, then there is nothing to keep us from imagining a language in which one would say:

  Edmondus snuoiqs basketballum pluss iggfruds soccerum.

  The speakers of this hypothetical language would see the actions of basketball players — or rather, of basketball snuoiqers — as being just as different from the actions of soccer igg fruders as the sounds “snuoiq” and “iggfrud” are different from each other.

  If this example’s zeugmaticity seems too weak, then we can try another avenue of approach to the same issue:

  Sylvia plays tennis, Monopoly, and violin.

  This sentence involves a musical instrument and two types of game that are much more different from each other than are basketball and soccer. If one tried to measure the distances between these three concepts by asking people to estimate them, it’s likely that most people would place violin quite a long ways from tennis and Monopoly, and those two games, though not extremely near each other, would be much closer than either of them is to violin. And finally, not too surprisingly, this matches the collective choice of Italian speakers, who would translate the above sentence as follows:

  Sylvia gioca al tennis e a Monopoly, e suona il violino.

  It would be unthinkable, in Italian, for anyone to play (in the sense of giocare) a musical instrument; the mere suggestion is enough to make an Italian smile. The kind of scene that such a phrase would conjure up is that of people playing catch with a Stradivarius, for instance. While it is natural for English and French speakers to see violin-playing as belonging to the same category as soccer-playing and basketball-playing, the idea would seem downright silly to Italian speakers.

  In French, the verb jouer is used both for musical instruments and for sports, but it is followed by different prepositions in the two cases. Thus one plays at a sport but one plays of a musical instrument. Does this syntactic convention split the concept of jouer into two quite clear and distinct sub-meanings? In English, there is no similar syntactic convention that would create a mental division of the verb “to play” into two separate pieces; rather, it simply feels monolithic.

  Playing Music and Sports in Chinese

  The distinction made in Italian between “giocare” (for sports) and “suonare” (for musical instruments) might seem a bit precious. After all, not only English but plenty of other languages are happy to use exactly the same verb for both kinds of
activities — thus French uses “jouer”, German uses “spielen”, Russian uses , and so on. What about Chinese?

  It turns out that Mandarin speakers are considerably more finicky in this matter than Italian speakers: they linguistically perceive four broad types of musical instruments, each type meriting its own special verb. Thus for stringed instruments there is the verb (pronounced “lā”), meaning roughly “to pull”, while for wind instruments one says (“chuī”), which means “to blow”. Then for instruments such as the guitar, whose strings are plucked by the fingers, or the piano, whose keys are pushed by the fingers, the verb is (“tán”) — and finally, for drums, which are banged, what one says is (“dǎ”).

  Curiously enough, it’s possible to apply the verb that means “to play” (as in “play with a toy”) to any musical instrument (it is , pronounced “wán”); unfortunately, however, the meaning is not what an English speaker might expect: it’s essentially the idea of fussing around with the instrument in question, and moreover this usage of is extremely informal, indeed slangy.

  One might naturally wonder how a Chinese speaker would ask a more generic question, such as “How many instruments does Baofen play?” But the best translations of this perfectly natural English sentence elegantly bypass the problem by making use of very broad verbs such as (“xuéxí”) or ("huì "), which means, respectively, “to study” and “to be able; to know”, and which have no particular connection with music. In short, there is no general verb in Mandarin that corresponds to the musical notion of playing, even though to us English speakers the concept seems totally logical, even inevitable; but the fact is that speakers of Chinese have no awareness of this lacuna in their lexicon, no matter how blatant it might seem to us.

  Well, all right, then. But what about playing games and sports — surely there is just one verb in Chinese for this monolithic concept? To begin with, one does not, in Mandarin, play board games and sports with the same verb. For chess, one engages in the activity of (“xià”), which one does not do with any kind of ball. And for a sport that uses a ball, it all depends on the kind of ball involved. For basketball, it’s (“dǎ”), the verb that applies to playing a drum (the connection may seem a bit strained to a non-Chinese), whereas for soccer it’s (“tī”), which means “to kick”. Thus one might say, “I prefer kicking soccer to beating basketball.” Once again we see that in a domain that strikes an English speaker as monolithic — everything is played, and that’s all there is to it! — distinctions are not just rife but necessary in Chinese.

  For English speakers, despite our use of the single verb “to play”, it’s not terribly hard to see that this verb conflates two activities that are quite different — namely, making rhythmic noises and having fun — and that the conceptual union thus created is not inevitable, and might even be seen as being rather arbitrary. On the other hand, within each of these two domains, it’s harder to see a lack of natural unity. If someone were to ask us if playing dolls, playing chess, and playing soccer are all really “the same activity”, we could of course point out differences, but to focus on such fine distinctions would seem quite nitpicky. And when we learn that in Mandarin, playing soccer and playing basketball require different verbs, it is likely to strike us as really overdoing things, rather as if some exotic tongue insisted on using two different verbs to say “to drink”, depending on whether it involved drinking white wine or red wine. But then again, this is an important distinction for wine-lovers, so it’s conceivable that some of them would very much like the idea of having two such verbs.

  Zeugmas and Concepts

  Our brief excursion to Zeugmaland will come to a climax in the following bold prediction:

  You will enjoy this zeugma as much as a piece of chocolate or of music.

  This sentence has a couple of zeugmatic aspects. Firstly, it plays on two senses of the noun “piece”. In some readers recognition of this contrast will evoke a smile, even though there’s no denying that both usages of the word are completely standard. Secondly, it plays on three senses of the verb “enjoy” — one involving a gustatory experience, another involving an auditory experience, and yet another involving the savoring of a linguistic subtlety. Each reader will of course have a personal feeling for how large the distinction between these three senses of the word is.

  Aside from making us smile, zeugmas offer us the chance to reflect on the hidden structure behind the scenes of a word or phrase — that is, on the concept associated with the lexical item, or more precisely, on the set of concepts associated with it — and since most words could potentially be used to form a zeugma (including very simple-seeming words such as “go”, as we saw above in the discussion of German and Russian), the phenomenon necessarily increases our sensitivity to the miracle of the human brain’s ability to spontaneously assign just about anything it encounters to some previously known category. After all, despite the inevitable and undefinable blurriness of the “edges” of each one of our categories, and despite the enormous number of categories, our brains manage to carry out such assignments in a tiny fraction of a second and in a manner of which we are totally unaware.

  The Nature of Categorization

  The spontaneous categorizations that are continually made by and in our brains, and that are deeply influenced not just by the language we are speaking but also by our era, our culture, and our current frame of mind, are quite different from the standard image, according to which categorization is the placing of various entities surrounding us into preexistent and sharply-defined mental categories, somewhat as one sorts items of clothing into the different drawers of a chest of drawers. Just as one can easily stick one’s shirts into a physical drawer labeled “shirts”, so one would easily assign dogs to the mental drawer labeled “dog”, cats to the nearby mental drawer labeled “cat”, and so forth. Every entity in the world would fit intrinsically into one specific mental “box” or “category”, and this would be the mental structure to which all the different entities of the same type would be assigned. Thus all bridges in the world would be unambiguously assigned to the box labeled “bridge”, all situations involving motion would be assigned to the box labeled “move”, and all situations involving things standing still would be assigned to the box labeled “stationary”. This mechanism of “boxing” everything in the world would be both automatic and completely reliable, the raison d’être of mental categories being to assign entities objectively to their proper conceptual label in an objective, observer-independent fashion.

  Such a vision of the nature of categorization is very far from what really goes on, and in the pages to come we will do our best to show why this is so. But hopefully, already from Chapter 1 onwards, readers will feel persuaded that mental categories are anything but drawers into which clear-cut items are automatically sorted, and this idea will be reinforced ever more strongly as the book proceeds.

  What, then, do we mean in this book by “category” and “categorization”? For us, a category is a mental structure that is created over time and that evolves, sometimes slowly and sometimes quickly, and that contains information in an organized form, allowing access to it under suitable conditions. The act of categorization is the tentative and gradated, gray-shaded linking of an entity or a situation to a prior category in one’s mind. (Incidentally, when we use the term “category”, we always mean a category in someone’s mind, as opposed to mechanical labels used in computer data bases or technical labels used in scientific taxonomies, such as lists of the names of biological species.)

  The tentative and non-black-and-white nature of categorization is inevitable, and yet the act of categorization often feels perfectly definite and absolute to the categorizer, since many of our most familiar categories seem on first glance to have precise and sharp boundaries, and this naïve impression is encouraged by the fact that people’s everyday, run-of-the mill use of words is seldom questioned; in fact, every culture constantly, although tacitly, reinforces the impression that words are
simply automatic labels that come naturally to mind and that belong intrinsically to things and entities. If a category has fringe members, they are made to seem extremely quirky and unnatural, suggesting that nature is really cut precisely at the joints by the categories that we know. The resulting illusory sense of the near-perfect certainty and clarity of categories gives rise to much confusion about categories and the mental processes that underlie categorization. The idea that category membership always comes in shades of gray rather than in just black and white runs strongly against ancient cultural conventions and is therefore disorienting and even disturbing; accordingly, it gets swept under the rug most of the time. Since the nature of mental categories is much subtler than the naïve impression suggests, it is well worth examining carefully.

  A category pulls together many phenomena in a manner that benefits the creature in whose mind it resides. It allows invisible aspects of objects, actions, and situations to be “seen”. Categorization gives one the feeling of understanding a situation one is in by providing a clear perspective on it, allowing hidden items and qualities to be detected (by virtue of belonging to the category person, an entity is known to have a stomach and a sense of humor), future events to be anticipated (the glass that my dog’s tail just knocked off the table is going to break) and the consequences of actions to be foreseen (if I press the “G” button, the elevator will go down to the ground floor). Categorization thus helps one to draw conclusions and to guess about how a situation is likely to evolve.

  In short, nonstop categorization is every bit as indispensable to our survival in the world as is the nonstop beating of our hearts. Without the ceaseless pulsating heartbeat of our “categorization engine”, we would understand nothing around us, could not reason in any form whatever, could not communicate with anyone else, and would have no basis on which to take any action.