Ambiguity, Polysemy and Vagueness
Ambiguity, polysemy and vagueness are terms used in cognitive semantics referring to different instances of plurality of meaning. According to Deane (1988) these three phenomena “form a gradient between total semantic identity” (vagueness) “and total semantic distinctness” (p. 327) (ambiguity). Therefore, polysemy is a case somewhere in between these two extremes. The borders between the categories of ambiguity, polysemy and vagueness are fuzzy. Thus, there are lexical examples that can be assigned to more than one category.
Ambiguity is a term used to characterise phenomena that have more than only one meaning. These meanings are distinct from each other and have no close schema in common. That is why a single expression may lead to multiple interpretations. In natural language many words, strings of words and sentences are ambiguous, simply because of the fact that numerous words cover several distinct meanings, or specific structural elements give rise to different readings. That means that “an expression or utterance is ambiguous if it can be interpreted in more than one way” (Löbner 2002: p. 39). However, disregarding puns (see 1.5), in every linguistic situation only one meaning of an ambiguous expression can be used. There are several forms of ambiguity to be distinguished – according to their trigger:
Lexical ambiguity is concerned with multiple interpretations of lexemes. A word is ambiguous if it involves two lexical items that have identical forms, but have distinct, i.e. unrelated meanings. There are numerous examples of lexical ambiguity. A clear-cut one is the lexeme ball. This word may either denote the round object which is used for several sports, like football, volleyball or basketball, or it can be used to refer to a large formal dancing party. Both forms are identically written and pronounced but just accidentally share the same form: ball in the sense of the round object originates in the Old Norse word ‘ballr’, whereas ball as the formal event comes from Greek ‘ballizar’ (meaning ‘to dance’) and was first attested in the English language in the 1630s being introduced through Old French. (Online Etymological Dictionary) Another example by Chierchia & McConnell-Ginet (1993) is the following sentence: You should have seen the bull we got from the Pope. The sentence is ambiguous, because the word bull may stand for several distinct things – either a male animal of different kinds, a swearword or an official order or statement from the Pope. As ambiguity is context dependent lexical disambiguation (knowing which word meaning has been used) is quite easy in most cases when considering previous expectations and context (p. 32).
The most classical example of lexical ambiguity is bank, which may either denote an organisation providing financial services, or the side of a river – just to mention two of the lexeme’s possible meanings.
Further examples of lexical ambiguity are:
bright - a bright (intelligent) person vs. bright (sunny) weather
file - arranged collection of papers vs. metal tool
Structural ambiguity is a result of two or more different syntactic structures that can be attributed to one string of words. That means that a sentence is structurally ambiguous not because it contains a single lexeme that has several distinct meanings, but because the syntactic structure of the sentence causes multiple interpretations.
(1) Young boys and girls love the adventure playground.
This sentence is syntactically ambiguous, because the reference of young is unclear. There are two possible interpretations of the subject. It may either be that [young boys] and girls love the adventure playground, or young [boys and girls] love [it]. The structural analysis shows that the sentence may be interpreted in a way that young only refers to the boys, or it may be understood as characterising the boys as well as the girls.
(2) Flying planes can be dangerous.
This often quoted example of structural (also called syntactic) ambiguity comes from Noam Chomsky. Sentences that contain lexemes that change their word form or even word class depending on the sentence’s interpretation are part of this category. Flying planes in this example sentence may be understood as “to fly planes” as well as “planes, which fly”. Therefore, the lexeme flying can be interpreted as the gerund form of a verb in a verb phrase, or as an attribute of a noun phrase.
There are several linguists, who have come up with similar examples like the last one. Lyons (1975) (cf. pp. 212f.) as well as Langacker (1993) (cf. p. 432), for example, determine those instances as grammatically ambiguous. The terms structural, syntactic and grammatical ambiguity are basically interchangeable.
Lyons’ examples for grammatical ambiguity are the following:
(3) They can fish.
The ambiguous interpretation is a result of “the double classification” of fish (either intransitive verb or noun) as well as of can (either auxilliary or transitive verb).
(4) beautiful girl’s dress
This is an example similar to the first one of structural ambiguity. It is a question of reference and therefore, may either be understood as [beautiful girl’s] dress or beautiful [girl’s dress]. Is the adjective beautiful attributed to the girl or to the dress? That question makes the reading of the sentence ambiguous.
However, Lyons points out that these cases of free interpretation are hardly leading to misinterpreations in natural language use. It usually gets quite obvious which meaning is aimed at through the rest of the sentence and the context.
The category of transformational ambiguity is mentioned by Lyons (1975), who characterises its prototypes as “ambiguous constructions which depend upon the ‘deeper connexions’” (p. 249). Furthermore, he points out that these constructions are mostly only ambiguous out of context. One of his examples is the phrase the love of God. Isolated from any textual relations it is unclear whether God is the subject or the object in this noun phrase. Additionally, Lyons quotes Chomsky’s already mentioned example of flying planes, which he – in contrast to other linguists – counts as belonging to transformational ambiguity. Basically, neither categorisation is wrong as many linguists do not distinguish transformational from grammatical ambiguity. Thus, transformational ambiguity is a subcategory of grammatical ambiguity.
Another Chomskian example mentioned by Lyons is: the shooting of the hunters. This is the same case as the God-example: it is unclear whether the hunters are subject or object in this phrase.
A further type of ambiguity called scope ambiguity is discussed by Chierchia and McConnell-Ginet(1993), among others. It is a subcategory of structural ambiguity. However, scope ambiguity can be distinguished from structural ambiguity through its “single surface syntactic structure” (p. 33). Chierchia and McConnell-Ginet’s example for this subcategory is: Someone loves everyone. This sentence can either mean ‘Everyone is loved by (at least) one person.’ or ‘There is a person and this person loves everyone.’ Note that in the former case someone may denote several different individuals, whereas it is only one single individual in the latter case.
A pun is a special form of ambiguity (mostly lexical) that is consciously used to create statements with ambiguous – distinct – meanings. That means that a pun is a play with words involving and creating double contexts (cf. Tuggy 1993: pp. 168, 178). Puns in spoken discourse make use of homophones, and puns in written discourse utilize homographs. Punning is a useful tool for jokes, creating at least two meanings – mostly a literal as well as a figurative one. Here is one example: After he ate the duck, the alligator got a little down in the mouth.
Tuggy (1993) offers a classical definition of vagueness. He characterises it as a linguistic phenomenon, where “two or more meanings associated with a given phonological form are […] united as non-distinguished subcases of a single, more general meaning” (p. 167). That means that vagueness involves “a lexeme with a single but nonspecific meaning” (ibid., p. 168).
Typical examples of vagueness are kinship terms, e.g. child, as well as lexemes with flexible boundaries, e.g. gradable adjectives like tall (cf. Löbner 2002: p. 45). An utterance like “It’s my child’s birthday tomorrow.” is vague, because the lexeme child is vague. There are two possible instantinations, namely [a female human under 18] and [a male human under 18]. When receiving such an uttereance it is much more likely that the common schema [a human under 18] is activated instead of its instantiations.
The lexeme tall in the statement “He is a tall man.” is vague as well. Tall belongs to the group of gradable adjectives. Therefore, the boundaries of the category tall are flexible depending on the context it is used in. From the perspective of a little child, a 5-foot-tall man will be considered tall. But the same man will not be considered with the same quality – tall – from the perspective of a 6-foot-tall woman. Thus, categorisations of this kind are always matter of norms (cf. Löbner 2002: p. 195).
On a scale of meaning variance ambiguity and vagueness are the two extremes, whereas polysemy is in between the other two. It shares features with both and is a common phenomenon in everyday language use. Polysemy involves lexemes that are clearly united (share a common schema) as well as clearly seperable at the same time. Polysemous words are the result of lexemes gaining new usages over time which share the same phonological form and appear to have separate meanings to non-etymologists. Foot is one example of polysemy. There are distinct usages of the word – either as a body part or as a scale unit (1 foot = 30.48 centimetres). Even though it appears that these are two very different meanings, both forms go back to the Old English word fot. Other examples of polysemy are earth (the planet vs. ground/ soil) and brother (kinship term vs. term of address for a male person living in a cloister).
Polysemy is sometimes mixed up with homonymy. However, as e.g. Langacker (1991) points out, there is a clear-cut distinction between these two (p. 268). Polysemous lexems always share the same etymological background and/or are conceived of as being semantically related by speakers, whereas homonymous words just happen to end up with the same phonological form. Therefore, homonymy may be seen as a subcategory of lexical ambiguity. Some linguists, like Löbner (cf. 2002: p. 39), claim the same for polysemy.
Metaphorical uses of words are instances of polysemy as well. The foot of a mountain, for example, makes use of the polysemous lexeme foot and denotes the bottom part of the mountain, just as a person’s foot refers to the bottom part of the human body.
 Online Etymological Dictionary, 23. January 2010.
Chierchia, Gennaro & McConnell-Ginet, Sally. 1993. Meaning and Grammar: An introduction to Semantics. Cambridge, Messachusetts [a.o.]: MIT Press.
Deane, Paul.1988. Polysemy and Cognition. Lingua 75.
Langacker, Ronald W. 1991. Concept, Image, and Symbol: The Cognitive Basis of Grammar. Berlin [a.o.]: Mouton de Gruyter.
Langacker, Ronald W. 1991. Foundations of Cognitive Grammar. Volume II. Descriptive Application. Stanford, California: Stanford University Press.
Lyons, John. 1975. Introduction to Theoretical Linguistics. Cambridge [a.o.]: Cambridge University Press.
Löbner, Sebastian. 2002. Understanding Semantics. London [a.o.]: Arnold [a.o.].
Tuggy, David. 1993. “Ambiguity, polysemy, and vagueness.” Cognitive Linguistics 4-3.
Spencer, Andrew. 1991. Morphological Theory: An Introduction to Word Structure in Generative Grammar. Oxford: Blackwell.
Langacker, Ronald W. 1987. Foundations of Cognitive Grammar. Volume I. Theoretical Prerequisites. Stanford, California: Stanford University Press.