Innovations

These categories, rough as they are, suggest that denominal verbs might be accounted for by derivation. Several linguists have argued just that. As Marchand says ‘Denominal verbs are verbalized sentences.’ McCawley, for example, suggests [...] that 21 be derived from 22:

(21) John nailed the note to the door.

(22) John caused a nail to hold the note on the door.

In this derivation, the capitalized words in 22 conflate to form the verb nail in 21. Green [...] proposes a similar analysis. Although McCawley and Green have not investigated the full range of denominal verbs, presumably they would derive the rest of them from sources not too different from our paraphrases.

But is this the right approach? For many common denominal verbs, derivations lead to problems. First, the noun origins of many verbs have been completely lost. How many people go back to Captain Boycott, Judge Lynch, and writing slates on hearing boycott the store, lynch the prisoner, and slate the event? These verbs have become opaque idioms. Second, even the more transparent verbs have interpretations that, strictly speaking, don’t contain the parent noun. If land and park truly meant ‘put onto land’ and ‘put into a park’, how could one land on a lake and park in a garage? Third, denominal verbs usually have semantic idiosyncrasies. Why should land the plane mean ‘put down’ and ground the plane ‘keep down’, instead of the reverse? That is, most common denominal verbs seem to be full or partial idioms. Their meanings have become fully or partially specialized, and are not fully predictable by an across-the-board process of derivation.

Innovative denominal verbs, however, do not have these problems. By definition they are not idioms; therefore they must be accounted for by some productive mechanism. But what is the mechanism like? We will be in a position to offer an answer to this question once we have considered the special properties of innovative verbs.

2.1. contextual expressions. Most semantic theories distinguish what we will call purely denotational expressions (man, blue, walk, day, bachelor) from indexical or deictic expressions (he, over there, yesterday, the bachelor). For an expression to be purely denotational, it must have a fixed sense and denotation. Bachelor, for example, has a fixed sense, say ‘unmarried man’, and denotes unmarried men in every real or imaginary world. Most English nouns, verbs, adjectives, and adverbs are of this type. For an expression to be indexical, however, it must have a fixed sense and denotation, but a shifting reference. He, for example, has a fixed sense, say ‘male person’, and denotes male people in every real or imaginary world. But the particular person it refers to — its referent — changes with the time, place, and circumstances of its utterance. The referential shifting is critical. Note that while bachelor is purely denotational, the bachelor is indexical because its referent will change from one use to another. What about innovative verbs? We will argue that they are neither purely denotational nor indexical, for they have a shifting sense and denotation. They constitute a new category that we will call contextuals.

To identify contextuals, one must be able to distinguish shifting from fixed sense and denotation. For that, we suggest three interrelated criteria.

(a) Number of senses.

Purely denotational and indexical expressions normally have more than one fixed sense and denotation. Bachelor, according to Katz & Fodor 1963, for example, has four: ‘unmarried man’, ‘young knight’, ‘person with baccalaureate degree’, and ‘mateless breeding fur seal’. In expressions like this, the number of senses is always small. Ambiguity of this type does not constitute shifting sense. To get some idea of what does, consider shifting reference. He, depending on the context, can be used to refer to any of an indefinitely large number of male humans — past, present, and future. Its referents cannot be enumerated. A distinguishing characteristic of something that shifts, then, is that it has an indefinitely large number of possibilities. So contextuals should possess not a small finite number of potential senses, but an indefinitely large number of senses.

(b) Dependence on context.

When expressions have a fixed sense and denotation, ‘these do not change with context (except for disambiguation). What about expressions with shifting sense and denotation? Once again we get a clue by considering shifting reference. The referent of he doesn’t merely change with the context. Listeners cannot hope to identify that referent without consulting information provided by the context — facts about the time, place, and circumstances of the utterance. This is a logical requirement for most uses of he. Because their referents cannot be identified from the sentence alone, they must be identified from the other facts associated with the utterance of that sentence, namely the context. So if contextuals have a shifting sense and denotation, these should depend on any, occasion on the context too.

(c) Cooperation between speaker and listener.

Shifting reference places a much greater obligation on the speaker and listener than does fixed sense and denotation. To use bachelor, the speaker must merely make certain that its denotation is correct — that the class of things it is intended to denote consists, say, of unmarried men. To use he, however, the speaker must also rely on the close cooperation of the listener, who must normally take note of such things as the speaker’s gestures, the people he has just mentioned, esoteric or private allusions, and other momentarily relevant facts about the conversation. That is, shifting reference requires a moment-to-moment cooperation that fixed sense and denotation do not. If contextuals with their shifting sense and denotation make similar demands, then we should expect moment-to-moment cooperation to be essential to their interpretation too.

Using these criteria, we will argue that innovative denominal verbs are contextuals. They have an indefinitely large number of potential senses; and their interpretation depends on the context, especially the cooperation of the speaker and listener. Once this is granted, innovative verbs must be dealt with differently from both purely denotational and indexical expressions.

2.2. proper nouns. Denominal verbs based on proper nouns are common, although most are virtually complete idioms. There are agent verbs based on people’s names: diddle, dun, finagle, fudge, lynch, pander, philander — all current; and balb, bant, bishop, burke, dido, hector, marcel, nap, swartout — all obsolete. There are recipient verbs from names of people who met defeat, hanging, or similar fates: boycott, and the now obsolete burgoyne, cornwallis, dewitt, job. There are verbs from place names, most with complicated histories: the current charleston, meander, saunter, shanghai; and the obsolete barbadoes, chevy, copenhagen, dunkirk, japan, levant, maffick, rotterdam, stellenbosch. And there are instrument verbs based on company names: the current hoover, scotchtape, and xerox, and the obsolete archie, baby, and roneo.

Proper nouns, however, are also an excellent source of innovations, as in these attested examples: to Luchins out (to get stuck in problem-solving because of set as discovered by Luchins); to Shylock £ 2700 from the £ 17, 000 raised; The wind Bernoullis around the building (speeds up according to Bernoulli’s Law); We then Kleinschmidted the DNA (used a method of visualizing DNA developed by Kleinschmidt); a conductor simply Elvira Madigans the movement to death (conducts soppily, as in the film Elvira Madigan); I wanted to Rosemary Woods out that conversation (erase as Woods is alleged to have done); you’re in danger of being Hieronymus Bosched (putin a nightmare setting); He is Svengaling her to death; She wasn’t Krishna’d out, she was only hippied out (affected as a member of the Krishna sect); the perils of Don Juaning; She seemingly malapropped; we all Wayned and Cagneyed — to buy breathing space from the guy who really did like to fight; and I WalterMitty’d. They are also easy to create: to Ralph Nader the insurance industry, to Valentino the woman, to Bonny and Clyde one’s way through the West, to TWA to New York, to Ajax the sink, and The canoe Titanicked on a rock in the river.

How are these innovations to be accounted for? In semantic theories as different as those of Mill 1843, Katz 1972, 1977, and Kripke 1972, 1977, proper names have reference but no sense. Harry Houdini is intended to pick out a specific historical individual, but not by virtue of a set of properties all or some of which that individual must satisfy — being an escape artist, an early airplane pilot, and an exposer of mediums. Rather, it is intended to pick him out by virtue of the fact that, at least for Kripke, the name rigidly designates him — picks him out in all possible worlds.

But if proper names are assumed to have no sense, where does the sense of the denominal verbs come from? In My sister Houdini’d her way out of the locked closet, uttered in the right context, Houdini has the sense ‘escape by trickery’. If the proper name Houdini has no sense, it provides no source for ‘escape by trickery’, let alone any other sense. It is as if ‘escape by trickery’ had been Houdini’d out of thin air. So Houdini the verb, although clearly the morphological child of Houdini the proper noun, cannot get its meaning from the parent noun in the expected way. Under these assumptions, it cannot be handled by such proposals as McCawley’s and Green’s, which derive the child’s sense from the parent’s.

Moreover, under the McCawley and Green proposals, Houdini the verb would be a purely denotational expression; but actually it is a contextual, a clear case of shifting sense and denotation. It depends for its interpretation on the context, and on the cooperation of the speaker and listener. For Sam to tell Helen My sister Houdini’d her way out of the locked closet, he must believe that they mutually know that Houdini was an escape artist. Mutual knowledge is used here in the technical sense of Lewis 1969 and Schiffer 1972 to mean that Sam and Helen each knows this particular fact about Houdini, each knows that the other knows the fact, each knows that the other knows that the other knows the fact, and so on. If Sam believed that Helen didn’t know about Houdini’s escape artistry (even though everyone else did), he couldn’t have used Houdini cooperatively on that occasion with the sense ‘escape by trickery’. Yet if he believed she knew about Houdini’s manner of death and his investigations of fake mediums (even though most other people didn’t), he could have expected her to understand Joe got Houdini’d in the stomach yesterday (‘hit hard without warning’) and I would love to Houdini those ESP experiments (‘expose as fraudulent by careful analysis’). In short, Houdini’s sense and denotation on each occasion depends on the time, place, and circumstances of its utterance. It depends not merely on one’s knowledge of English, but also on one’s knowledge of particular facts about Houdini the historical figure.

If Houdini is a contextual, it should have an indefinitely large number of potential senses; and it does. Indeed, it has as many senses as there are facts that speakers and listeners could mutually know about Houdini. In theory, that number is indefinitely large. In context, of course, the number is narrowed down to one; it must be, just as the indefinitely large number of possible referents for he is narrowed down to one. That is accomplished through the speaker’s and listener’s judicious use of contextual facts. We will consider how that is done later.

Against this analysis, however, one could argue that the verb Houdini is derived from a Houdini, a common noun, rather than Houdini, the proper noun. Since a Houdini can be purely denotational, as in He is a Houdini (the argument would go), the verb Houdini could be derived with a fixed sense and denotation. For one class of innovative denominal verbs, however, this is not a possible explanation. Consider the report by Herb Caen, in the San Francisco Chronicle, that a woman had been Jarvis-Ganned out of her Convention and Visitor’s Bureau job, where Jarvis-Gann is the name of a California tax-cut initiative that resulted in the axing of many jobs throughout the state. Caen didn’t mean that the job was eliminated by a Jarvis-Gann initiative (the common-noun interpretation), but by the Jarvis-Gann initiative. Another example appeared in the title of a 1966 Paul Simon song, A simple desultory Philippic, or how I was Robert McNamara’d into submission, where the person responsible for the submission was Robert McNamara himself. Similar innovations are easy to create: Richard M. Nixon was John Deaned right out of the White House; General Motors was Ralph Nadered into stopping production of the Corvair; Napoleon was Waterlooed in 1813; and The medium Margery of Boston was Houdini’d into disgrace in 1924. If these proper nouns are assumed to have a reference but no sense, then the previous arguments apply, and these verbs must be contextuals.

Even under the common-noun proposal, however, the verb Houdini must be a contextual. Note that a Houdini is itself an innovation, whose sense and denotation on each occasion depends on the speaker’s and listener’s mutual knowledge about Houdini and the context. So if the verb Houdini is formed from a Houdini, it too must be a contextual. At present, we know of no evidence favoring either Houdini’s direct formation from the proper noun or its indirect formation via the common noun. We will assume the direct route, although nothing critical to our argument hangs on that assumption.

2.3. common nouns. Unlike proper nouns, most common nouns have a sense that could conceivably serve as the basis for the sense of innovative verbs. To loaf the dough, for example, might be derived from something like to cause the dough to come to be like a loaf. The sense of the lexical constants cause, come, be, and like would be conflated with the sense of loaf, to form the sense of to loaf. This is the essence of Green’s and McCawley’s approaches.

The trick is to find the right lexical constants. Thus Green argues that instrument verbs (like hammer and radio) are derived from ‘as by using NP (on) in the usual manner, for the purpose for which it was designed’. Given the hammer’s design and usual manner of use, hammer gets roughly the right interpretation in hammer the nail into the board and hammer on his head with a shoe. The representation also accounts for such innovations as unicycle down the street, autoclave the scalpels, and keypunch the data.

Counter-examples, however, are easy to find. On the BBC in 1976, a demonstrator complained, We were stoned and bottled by the spectators as we marched down the street; and the (London) Observer noted that battered wives may be stabbed or bottled as well as punched. Bottles, of course, are designed for storing liquid, as reflected in bottle the beer; yet both innovations are perfectly interpretable. Most objects can be used for purposes for which they weren’t designed, and denominal verbs can reflect those purposes. This is also shown in such innovations as celluloid the door open (‘use a credit card to spring the lock open’, from the San Francisco Chronicle), hairpin the lock open — and, from a Time Magazine article on pie-throwing, pie the woman in a local doughnut store.

What is critical for innovative instrument verbs is not normal function or usual manner, as Green claims, but the speaker’s and listener’s mutual knowledge, along with certain other criteria. Imagine that Ed and Joe have an odd mutual acquaintance. Max, who occasionally sneaks up and strokes the back of people’s legs with a teapot. One day Ed tells Joe, Well, this time Max has gone too far. He tried to teapot a policeman. Joe arrives at teapot’s sense, ‘rub the back of the leg with a teapot’, not by using Green’s normal function, but by finding a situation that is consistent with Ed’s and Joe’s mutual knowledge about Max and teapots — the situation Joe thinks Ed intended to denote. If Ed hadn’t believed that Joe knew about Max’s compulsion, he couldn’t have meant what he did, nor expected Joe to see what he meant.

So teapot, like Houdini, is a contextual, with a shifting sense and denotation. First, teapot has an indefinitely large number of possible senses. It has as many senses as there are Max-like stories that one could contrive, and that number is without limit. Second, its sense and denotation on each occasion depend on the context. If teapot the policeman had been uttered under different circumstances, it could have had the sense ‘bash a teapot over the head of’, ‘offer a teapot to’, ‘turn by sorcery into the shape of a teapot’, etc. Note, incidentally, that in each of these senses, the verb teapot ‘relies’ in part on the fixed sense of the noun teapot in denoting teapots. It is just that, of the many situations in which teapots can play a role, the one intended can only be determined at the time of utterance. Third, the sense of teapot intended on each occasion depends critically on the cooperation of the speaker and listener. They must assess their mutual knowledge at the moment, and use other constraints that we will take up later. What holds for teapot appears to hold for every other innovative denominal verb, as well. Thus innovative verbs formed from common nouns appear to be contextuals too.

Although Green’s representation for instrument verbs makes them appear to have a fixed sense and denotation, that isn’t really so. Note that her representation contains the phrases ‘the usual manner’ and ‘the purpose for which it was designed’ — both indexical expressions whose referents change with the time, place, and circumstances of the utterance. Both phrases refer to facts that presumably lie outside one’s linguistic knowledge. Since these indexicals depend on context, so must the senses that contain them; and the same obviously holds for many of our initial paraphrases. The one for butcher ‘do the act that one would normally expect a butcher to do’ hides an assortment of similar indexicals, as does the one for bicycle. Treating innovative verbs as contextuals makes quite explicit what up to now has always been implicit in such paraphrases.