But no language in fact does this, because all languages come from the key human need to have categories that apply at once to a vast number of superficially extremely different and yet deeply extremely similar situations. Such categories help us to survive and to have comfortable lives. To be sure, some language could, in principle, have separate words for red books and green books, or for books printed on butterfly wings, or for orange books of under 99 pages, or for puce-colored books about subtropical botany that contain between 221 and 228 pages (but not 225) and are in (Brazilian) Portuguese and are typeset in 13-point Bodoni — but it’s obvious that there comes a point of diminishing returns, and it’s nowhere near the absurdly fine distinctions just hinted at. There’s no reason for any culture to construct any of these categories, let alone to reify it via a word in its language, although the miracle of language — of every language on earth — is the charming fact that any of those odd and far-fetched categories could in theory be invented by someone, if they were needed or desired.
We should also point out that category refinement doesn’t always move in the direction of an ever-finer mesh. Sometimes refining one’s mental lexicon of categories means broadening through abstraction, in the sense of learning to perceive common threads in situations where people who lack the concept would simply see unrelated phenomena (for example, the commonality linking human mothers with animal mothers, den mothers, and mother companies, or the commonality linking female animals with female plants, or the commonality linking hubs of wheels with airports that are hubs, and so forth). The emergence of this type of broader category is also extremely useful for the development of a people or a culture.
There is thus a tension between the desire to make finer distinctions that cover very few cases and the desire to make broader categories that cover many more cases. Earlier, we saw that children’s perception of the world is quite coarse-grained relative to the perception of adults (this is why some young children uninhibitedly speak of “patching teeth”, “eating water”, “undressing bananas”, and so forth), and we saw that as children grow older, they acquire more and more refinements in their conceptual systems. This is a universal tendency, but at some point, adults stop refining their lexicon when it comes to ordinary objects, actions, relationships, and situations. Each language and culture has found its natural grain size for such entities, and in a kind of unspoken collective wisdom, it ceases to go beyond that, although of course experts are continually refining their technical vocabularies, and each society, as it makes new discoveries and inventions, collectively creates new concepts and new words for them.
Everyone in every culture is constantly refining their conceptual repertoire by acquiring ever more compound words, idiomatic phrases, proverbs, and new catch phrases that enter the language through books, movies, and advertisements; in addition, everyone is also constantly building up a rich repertoire of concepts that have no verbal labels. In the next two chapters, we will turn our attention to these two key ways in which our conceptual storehouse continues growing as long as we live.
CHAPTER 2
The Evocation of Phrases
Categories Vastly Outnumber Words
What is it that links words and categories? To be sure, words are often the verbal counterparts of categories. We can describe and refer to categories with them, but that does not mean that categories should be equated with words — not even with the broader notion of lexical items — for categories are mental entities that do not always possess linguistic labels. Often words are names of categories, often they can be used to describe categories, but sometimes they simply are lacking. All in all, the connection between categories and language is complex. A single word can of course bring a category to mind — “mother”, “moon”, “chair”, “table”, “office”, “study”, “grow”, “shrink”, “twirl”, “careen”, “thanks”, “ciao”, “much”, “and”, “but”, and so on — but the correspondence is somewhat lopsided, because in fact we all know many more categories than we know words.
Coining a word is cognitively costly, and our mental categories are so numerous and constantly changing that it would take an astronomical repertoire of words if we wanted to have exactly one word per category. As a consequence, humans have figured out how to economize with words. Thus, there are many words that have multiple meanings, depending on the context. Such words cover a variety of categories (consider the multitude of meanings of a simple word like “trunk”, for instance). Another word-saving device is that many categories have verbal labels that consist of a string of words rather than just one word, and that idea will be the central focus of the present chapter. And then there are myriads of categories that simply have no verbal label at all, and the goal of the next chapter will be to shine a bright light on those.
In sum, whereas Chapter 1 focused on categories whose labels are just one word long, this chapter is concerned with categories whose linguistic labels are more complex; thus compound words, idiomatic phrases, proverbs, and fables are among the scenic spots we shall visit.
Psychology does Not Recapitulate Etymology
No less than indivisible words, compound words designate categories. Thus the word “airplane” is no less the name of a category than are “air” and “plane”; the same goes for “airport”, “aircraft”, “airfield”, “airlift”, “airsick”, “airworthy”, “airhead”, “airbag”, “airplay”, “airtight”, and so forth. There are many words whose components are so tightly fused inside them that the individual pieces are seldom if ever noticed, since (in most cases) the wholes are not analyzable in terms of their pieces — for example, “cocktail”, “cockpit”, “upset”, “upstart”, “awful”, “headline”, “withstand”, “always”, “doughnut”, “briefcase”, “breakfast”, “offhand”, “handsome”, “cupboard”, “haywire”, “highjack”, “earwig”, “bulldozer”, “cowlick”, “dovetail”, and so on.
To be sure, in some of these cases — for instance, “cupboard” and “headline” — a little guesswork provides a plausible story about their origins, but the possibility of doing an intellectual analysis doesn’t mean that a fluent speaker conceives of the word — that is, hears it — as a compound word. For example, we don’t pronounce “The plates go in the cupboard” as if it were written “The plates go in the cup board”, and we don’t hear it that way. In fact, we never say “board” when we mean a storage location, even if it once had that meaning. What we say aloud sounds more like “cubberd” than like “cup board”, and virtually no one hears either part inside the whole. As for “airport”, although we can deliberately slow down and hear “air” and “port” inside it, who ever thinks about the atmosphere and about a harbor when picking up a friend at the baggage area, or when transferring between planes? Indeed, were someone to call an airport an “atmospheric harbor”, it would invite ridicule, if not sheer incomprehension.
Many compound words are positively mysterious if one starts to think about them. Why do we sometimes call a woman’s purse a “pocketbook”? It’s not a book by any stretch of the imagination, and it certainly doesn’t fit in any kind of pocket! Nor is it a book of pockets! And how can we understand the compound word “understand”? Understanding has nothing obvious to do with standing anywhere, let alone underneath something. Then again, in certain compound words, just one of the two components sounds strange and strained, such as the “body” in “nobody”.
An analysis of where words come from and how they came to mean what they now mean belongs to the classic discipline of etymology, and often it is truly fascinating, but it does not have much bearing on how words are actually perceived by a native speaker. In that sense, psychology does not recapitulate etymology (to tip our hat to the phrase “ontogeny recapitulates phylogeny”). Many compound words simply act like indivisible wholes; we learn them as wholes as children, and it is as such that we usually hear them.
Thus a toddler learns and uses the word “pacifier” without having any ide
a of the existence of the verb “pacify” or the suffix “-er” inside it (not to mention the Latin root “pax” and the suffix “ify” found inside the verb “pacify”!). For a toddler’s purposes, the sound “pacifier” is simply that concept’s arbitrary-seeming label, and the word doesn’t need to be broken down or analyzed. As for adults, they seldom need to decompose compound words, either; indeed, doing so would often be more confusing than helpful, as in “eavesdrop” or “wardrobe”, for example. Who ever thinks of a wardrobe as a place where one wards one’s robes? The standard pronunciation (“war-drobe”) would not suggest hearing it or thinking of it that way (and in any case, the verb “to ward” is quite a stretch when one is talking about storing clothes in a closet). As for “eavesdrop”, well, that’s just as opaque as “understand”, “handsome”, “cockpit”, and “cocktail”.
Often we hear one part of a compound word quite clearly inside the word, and the other part less clearly. Thus, people called “gentlemen” are always of the male sex (showing that the second component is heard loud and clear), but they are certainly not always gentle. It is perfectly possible to say, “Would the rowdy gentlemen in the corner of the room please pipe down?” On the other hand, a freshman is always fresh (in the sense of being new), but only about half the set of freshmen in a typical high school are men. A great-grandson and a great-grandmother are unlikely to be particularly great or grand, but the former is sure to be somebody’s son, and the latter to be somebody’s mother. A restroom is certainly a room, but seldom if ever is it a place to take a rest.
Compound words whose components are still at least blurrily heard in the whole can be a bit tricky when it comes to pluralization, because one isn’t sure to what extent one hears their parts resonating inside them. Thus when we sit in a café, do we gaze at the passersby or the passerbys as they stroll before us? And how many teaspoonsful of sugar do we add to our coffee? That is to say, how many teaspoonfuls? Are we thinking of giving our children jacks-in-the-box when Noël rolls around, or contrariwise, jack-in-the-boxes at Yuletide? On the golf course, do we aspire to make holes-in-one, or would we prefer the glory of hole-in-ones? And in golf tournaments, do we beam if we are runners-up or are we disappointed to be mere runner-ups? As married folk, are we fond of our mothers-in-law while finding our father-in-laws rather stuffy? And turning the tables, how do those respected elders feel about their sons-in-law and daughter-in-laws?
Looking at the statistics of the rival plurals for compound nouns of this sort gives one a sense for where those nouns lie along the slippery slope on which the parts slowly “melt”, over time, into the whole. But once the parts have truly been absorbed into the whole, then the whole becomes truly a single unit, and no one hears the pieces any longer. Thus “handsome” might as well be spelled “hansim”, “nobody” might as well be spelled “gnobuddy”, “cupboard” “cubberd”, and so on — and of course we have all seen “donut”, “hiway”, and “hijack”, which show the parts as they make their way towards absorption (much like the vestigial “five” and “ten” inside “fifteen” and the vestigial “two” inside “twelve” and “twenty”).
Often compound words have drifted so far from their etymological roots that native speakers can easily miss what is right in front of their eyes. Thus in German the word for “nipple” is “Brustwarze”, which, broken up into its parts (the two nouns “Brust” and “Warze”), means “breast-wart”. Once again in German, the word for “glove” is “Handschuh” (“hand-shoe”), and the French word for “many” is “beaucoup”, which, decomposed, is “beau coup” — that is, “beautiful blow”. But no native speaker would hear these words in the way that they strike us — namely, as ugly or strange — because over time, they have melted together to make category names that are seamless wholes and which therefore feel completely bland.
How could the native speakers of these languages possibly fail to see (or hear) something that is so blindingly obvious? Is it really possible? Well, yes — it’s just as possible for them as it is for us anglophones to fail to see or hear the “dough” and the “nut” inside “doughnut”, or the “break” and “fast” inside “breakfast”, or the “under” and “stand” inside “understand”. And keep in mind that no one flinches at the overtly sexual allusions in the common terms “male plug” and “female plug”.
Opening the Door Doesn’t Require Taking the Lock Apart
“Front door”, “back door”, “doorknob”, “door knocker”, “dog door”, “dog dish”, “dish towel”, “dishwasher”, “washing machine”, “dining room”, “living room”, “bedroom”, “bathroom”, “bathtub”, “bath towel”, “towel rack”, “kitchen table”, “tablecloth”, “table lamp”, “lampshade”, “desk chair”, “hair dryer”, “grand piano”, “piano bench”, “beer bottle’, “bottlecap”, “toothbrush”, “toothpaste”… Here, without our once setting foot outdoors, are some compound words or phrases that designate familiar household sights. Some are written with a space between their components, and some are not. Fairly often it takes a trip to the dictionary to find out which ones take a space and which do not, and at times the official word handed down from on high runs against the grain or seems totally arbitrary, and moreover the official spelling frequently changes as one traverses the Atlantic or the decades. Indeed, from a psychological as opposed to an etymological point of view, the presence or absence of a space (or sometimes of a hyphen) makes no difference to the typical language user (or language-user), who is unaware of such fine points and will usually just improvise in writing such things down. One’s point will be made equally well whether one writes “door knob”, “door-knob”, or “doorknob”.
Although the types of words (and phrases) shown above have visible, hearable inner parts, these expressions are every bit as much the names of mental categories as are “simpler” nouns, such as “chair”, “table”, and “door”. These longer words and phrases are, just like the things that they name, wholes that are built out of parts. And yet, no more than we need to understand a physical tool in order to use it do we need to take apart a compound word or phrase in order to use it. We use our dishwashers and our loudspeakers as wholes or “black boxes”, undismantled and unexamined, and much the same holds for their names.
This observation has important consequences. Contrary to what one’s intuition might suggest, using a compound noun or phrase rather than a “simple” word does not mean that more cognitive activity is required to understand it, or that the named category resides at a higher level of sophistication. When we hear “living room”, for instance, it doesn’t mean that first we activate the most general concept of room (which includes dining rooms, bedrooms, bathrooms, restrooms, waiting rooms, etc.) and then maneuver inside the abstract space of room-ness until we locate the appropriate subvariety. Our concept living room enjoys the same status as do “simple” concepts such as room or bed. In other words, the fact that “living room” is a compound word doesn’t cast doubt on its status as the name of a stand-alone mental category. The same holds true for “bottlecap”. Understanding this word doesn’t require locating it among the subcategories of the concept cap, which include polar caps, yarmulkes, dental crowns, and lens covers. Cognitively speaking, bottlecap is no less simple a concept than are cork, plug, and lid, which, like bottlecaps, are devices for closing containers of liquid.
Jumping around from language to language helps make this idea clearer and more believable. Thus to express our simple noun “counter”, French uses three words — namely, “plan de travail” (“surface of work”) — while Italian uses just one — “banco”. Our two-word noun “dish towel” is merely the atomic “torchon” in French and the slightly molecular “strofinaccio” (“wiper”) in Italian; similarly, our “living room” is merely “salon” in French and “soggiorno” in Italian. Our compound noun “bedroom” is “chambre à coucher” (“room for sleeping”) in French but merely “camera” (not a compound noun) in Italian. And
our “camera” is “appareil photo” (“photo device”) in French and “macchina fotografica” in Italian. And oddly enough, our compound noun “video camera” is simply “caméra” in French and “telecamera” in Italian. The moral here is that what seems like a blatant compound in one language may perfectly well seem atomic — that is to say, unsplittable — in another language. (Speaking of atoms, the indivisible English word “atom” comes from a compound word in the original Greek — “a-tomos” — meaning essentially “without a cut” or “part-less”. Thus, as was wittily pointed out by David Moser, the word “atom” is an unsplittable etym in English despite not being so in the original Greek, and contrariwise, physical atoms are now known to be splittable despite what their etymology would suggest.)
In order to understand a compound noun, we do not need to break it down into its parts and then put together their “simpler” meanings in order to figure out what is being spoken of. To be sure, we are all aware that the words “bath” and “room” are found inside “bathroom”, and that “tablecloth” means a piece of cloth that one spreads out on a table, but we don’t need to take those words apart to understand them — no more than we do with “afternoon”, “psychology”, or “atom” — unless there is a special context that calls for it, such as explaining their meanings to a foreigner or a child.
By Concealing their Constituents, Acronyms Seem Simple
A widespread linguistic phenomenon that clearly illustrates the universal human tendency to represent complex concepts by short chunks whose parts are clearly “there” and yet are seldom if ever noticed is that of the creation and propagation of acronyms. Among the earliest-known acronyms are in Latin: “SPQR”, standing for “Senatus PopulusQue Romanus” (“The Roman Senate and People”) and “INRI”, standing for “Iesus Nazarenus Rex Iudaeorum” (“Jesus the Nazarene, King of the Jews”). For ages, letter writers have used “P.S.” (“post scriptum”), and mathematicians, not to be left behind, have for centuries used the classic Latin abbreviation “QED” (“quod erat demonstrandum” — “which was to be demonstrated”) to signal that the end of a proof has been satisfactorily reached. For centuries the British have used “HRH” (His/Her Royal Highness) and “HMS” (His/Her Majesty’s Ship), and of course there is the famous old call for help, “SOS” (Save Our Ship).
Surfaces and Essences Page 16