Chapter 3



Word Meaning, Sentence Meaning,

Speaker Meaning






The distinction between these two subdisciplines of linguistics has standardly been founded on the context of discourse: pragmatics has been claimed to study the contribution of the context (that is linguistic and situational context) to the meaning.  Hence, both semantics and pragmatics make use of context to a smaller or greater degree: the two fields are not disjoint. 

       ‘To mean’ is a notoriously vague predicate. For example, it can figure in the following expressions:


(1)   Dark spots on the leaves mean a deficiency of iron.

(2)   The bell means the end of the lesson.

(3)   ‘Sprachwissenschaft’ means ‘linguistics’.


These meanings of ‘meaning’ are of little interest to semantics. In (1) we have a natural sign of a disease, in (2) a conversational symbol and in (3) a translation from one natural language to another. Instead, in semantics we are interested in the relation world. In other words, we are interested in how sentences of natural language, such as English, reflect reality, be it objects, people, states, events, processes – or, according to a slightly different view, how sentences relate to our mental representations or reality.

       The meaning of a sentence is a proposition. Propositions are descriptions of states of affairs, for example (4) said by me and (5) said to me by someone else express the same proposition.


(4)   I am tired

(5)   You are tired.


Similarly, (6) and 7 express the same proposition.


(6)   The dog ate the chocolates.

(7)   The chocolates were eaten by the dog.


This is so in spite of the fact that a sentence in the active voice and its passive equivalent are not identical in meaning. The proposition does not exhaust the meaning of the sentence. It constitutes the core but there is more to meaning than the proposition expressed. In other words:


A proposition is that part of the meaning of the utterance of a declarative sentence which describes some state of affairs.

Hurford and Heasley (1983:19)


We can distinguish two types of semantics: linguistic semantics , or semantics in a narrow sense, and philosophical semantics , or semantics in a broad sense. Linguistic semantics studies meaning as an output of grammar. Philosophical semantics aims at establishing what proposition (or thought) has been conveyed by the speaker’s utterance of a sentence. For example, when the speaker says (8), linguistic semantics will not produce a full proposition. We have to establish (in philosophical semantics) who ‘the boy’ refers to and what it is that he is not old enough for.




(8)   The boy is not old enough.


In this way we relate the sentence to the world, we establish what it  says about a situation in the world. We need both linguistic and philosophical semantics to study meaning.





In the course of this book we shall be concerned with word meaning, sentence meaning and utterance meaning, trying to establish connections between these three levels of meaning and unify their extant analysis into one coherent view of meaning in language and discourse.  First, let us consider several approaches to word meaning. We can surmise that to know the meaning of a linguistic expression is to be in a position to provide the definition of the meaning of words. This view is  known as meaning as definition. By combining words into sentences by means of grammatical rules, one could arrive at meanings of sentences as definitions of these sentences.

       Finally, words  are used differently  in different contexts, in (9), the speaker does bit normally mean that he or she is going to die but rather that the task is particularly difficult and demanding.


(9)   This essay is killing me.


Instead of placing all these various uses of the word ‘kill’ in its definition, it is more plausible and economical to resort to the study of the contribution of context to meaning. The role of context in defining meaning is standardly approached by  dividing literal, conventional, context-independent aspects of meaning and all those aspects which depend on the situation of utterance.  But, separating context-free from context dependent aspects of meaning amounts to a mere redefining of the problem rather than solving it: the task is as equally difficult as the original one.

       Let us return to the view that meaning is reference. This view states that the meaning of a word is derived from the way it describes the world. When the speaker utters (10), ‘I’, ‘the black cat’ and ‘the roof’ refer to a person, an animal and an object respectively.


(10)   I saw a black cat on the roof.


But the meaning of words is not exhausted in their reference. Perhaps reference is not eve very important. As one influential theory has it, part of the meaning is also derived from relation words enter into among themselves in the language system. For example, we know what ‘cat’ means because we know what slot the concept CAT OCCUPIES in the English language.

       Let us have a closer look at reference. Speakers use language to talk about the world. When the speaker utters (11), he or she uses the proper names ‘Ludwig Wittgenstein’ and ‘Cambridge’ to refer to an individual and a location respectively.


(11)   Ludwig Wittgenstein lived in Cambridge.


Referring is a technical term in semantics. The reference (or referent) is the entity to which the expression refers. Words refer to entities but also speakers refer to entities in the world  by using these expressions. Hence, we have to distinguish reference from the act of referring , just we have to distinguish sentence meaning and speaker meaning. There is another term in semantics that has a similar role to play, namely denotation . A word refers to an object or person, but it also denotes a class of entities. In some cases, this class can have one member or can be empty. For example, ‘dog’ denotes all dogs in the world (that is, describes anything that can be called a dog), but refers to a particular dog, or to the property of doghood if no particular dog is meant. So, an expression denotes a class and refers to an entity or a property. The class denoted is called expression’s extension. In the case of proper names, the class contains one member, that is the particular person, city and so on. Denotation does not depend n the context, it is stable, whereas reference is context-dependent: ‘the dog’ refers to different dogs on different occasions.

       It is necessary at this point to establish which expressions in language are referring expressions. Proper names (‘Ludwig Wittgenstein’), definite descriptions (‘The author of Philosophical Investigations’) – although not all definite noun phrases – some pronouns (‘he’) including demonstratives (‘that’), and demonstrative noun phrases (‘this man’) are referring expressions. This reference can be either constant, context-independent (as in ‘Ludwig Wittgenstein’), or context-dependent (as in ‘that man’ or ‘he’)

        Similarly, pronouns and demonstratives refer by directly picking out the object. Definite noun phrases also refer, but it need not mean that the speaker refers to an entity by using them. In (12), the definite noun phrase ‘the architect of this church’, used as a definite description, has a referent, a particular person, but the speaker may have just said that whoever the architect was, he or she was insane.


(12)   The architect of this church was mad.


In other words, the definite description may have been used referentially, let us say referring to Antonio Gaudi, or attributively, attributing the property of madness to whoever satisfies that description.

       According to a more popular example, the ancients thought that what they called the Morning Star and the Evening Star were two different celestial bodies while, unbeknownst  to them, it was one and the same planet, Venus. Hence, the ancients would presumably consent to (13) as a tautology (a sentence that is always true, due to its form), but deny (14). Clearly, these two sentences in turn differ in meaning from (15).


(13)   The Morning Star is the Morning Star.

(14)   The morning Star is the Evening Star.

(15)   Venus is Venus.


In other words, the Morning Star and the Evening Star are coreferential, that is refer to the same object, and yet they differ in meaning, as the failure of substitutively of one for the other in sentences such as (16) demonstrates. The substitution of the asterisked expression results in a false statement in this situation:


(16)   The ancients believed that the Morning STAR/ *THE Evening Star is the Morning Star.


The substitutivity is not truth-preserving ( not salva veritate ) and the context such as (16) are called intensional contexts , that is context in which intensions matter for the meaning. In (16)   we have an instance of a so-called propositional attitude report , a sentence reporting on an attitude, such as belief, thought, fear, doubt, etc., towards a proposition.





There are two core issues to remember from the discussion so far: (i) the distinction between semantics and pragmatics and consequently between the sentence, proposition and utterance, and (ii) the role of reference and sense in defining word meaning.



The meaning of sentence (17) includes, or entails, the meaning of (18).


(17)   Tom bought tulips.

(18)   Tom bought flowers.


Entailment is an automatic commitment to the truth of the other sentence. Under some conditions entailment is reversed: (19) entails (20) in spite of the inclusion of the meaning of ‘flower’ in the meaning of ‘tulip’ (note, the class membership is opposite to  meaning  inclusion: the class of flowers includes the class of tulips).


(21)   All animals are forbidden.

(22)   All dogs are forbidden.


While on the level of sentences we talk about the relation of entailment, on the level of lexical items entailment is often called hyponymy: ‘tulip’ and ‘rose’ are hyponyms of ‘flower’. A hyponym includes the meaning of a more general term (here ‘flower’). The more general word is called a superordinate or a hypernymy. Hyponymy can be a many level-relation, creating whole networks, such as ‘ostrich’ is a hyponym of ‘bird’; ‘bird’ is a hyponym ‘animal’; and so on. The relation of hyponymy is transitive:


(23)   for all x, y and z, (if R (x, y) and R (y, z), then R (x, z)),


Where R stands for the relation and x, y, and z for the juxtaposed terms. Hyponymy is a principle on which we can build whole taxonomies based, for example, on species and kinds of animals and plants. Words on the same level in the network, such as ‘rose’, ‘tulip’ and ‘poppy’ as hyponyms of ‘flower’, are called sisters. Now, when we look at some sisters, such as in the male-female pairs ‘dog’-‘bitch’,  ‘hog’-‘sow’, we can see that the superordinate is not always available. ‘Hog’’ and ‘sow’   have a superordinate term ‘pig’, but ‘dog’ and ‘bitch’ have the superordinate ‘dog’. ‘Dog’ is called the unmarked member of the pair, whereas ‘bitch’ is marked semantically and distributionally: it is more informative and more restricted in meaning (distribution).

        A special kind of inclusion is meronymy , a part-whole relation between lexical items, for example ‘body’-‘hand’, ‘hand’-‘finger’, ‘finger’-‘nail’, ‘house’-‘chimney’.  Meronymy also forms a hierarchy, as can be seen from the  body to  nail example. But meronymy is an optional relation: the part is not a necessary constituent of the whole, for example not all houses have chimneys. In some languages, meronymy is distinguished grammatically as inalienable possessives .



3.2                SAMENESS


The relation based on sameness is called synonymy. Synonymy is mutual hyponymy: the first term entails the other and the second entails the first. Synonyms are words that are phonologically different but have the same meaning, for example ‘boy’ – ‘lad’, ‘hide’ – ‘conceal’, ‘police officer’ – ‘cop’, ‘lawyer’ – ‘attorney’, ‘toilet’ – ‘lavatory’. The constraint the same meaning seems very strict and it is contentious whether two different terms can ever have exactly the same meaning. The substitution renders different meaning and a less likely construction in (24a) and (24b), while leaving the meaning little affected in (24c).


(24a)   strong/’powerful tea

(24b)   powerful/’strong car

(24c)   powerful/strong argument


‘Deep’ and ‘profound’ trigger similar reflections. Collocations are even  more useful in fossilized phrases, such as in (25), where both adjectives mean ‘old and unsuitable for eating’.


(25)   rancid/*addled butter

         Addled/*rancid eggs


‘Herd’, ‘flock’ and ‘pack’ of animals give rise to similar observations. So, it seems that the test for synonymy is substitutivity in all contexts. According to this test, the word ‘pen’ is ambiguous between an enclosure and a writing tool.


(26)   Mary bought a pen and so did John.


Sentence (26) clearly means that they both bought the same type of  item rather than one a writing instrument and the other a fenced area of ground, unless there is deliberate word play involved. In the case of vagueness, two different uses of an expression can be juxtaposed in a ‘do so’ sentence as in (27):


(27)   Mary is a dancer and so is John,


where ‘a dancer’ is unspecified as to male/female features. Collocations also help decide between ambiguity and vagueness: if two uses of a word have different collocational links, then there is an ambiguity.

        There is no clear-cut boundary between homonymy and polysemy: the two senses of a phonological word can be remotely related. For example, the senses of ‘gay’ are related historically (etymologically) but are felt to be an example of homonymy rather than polysemy. ‘Flower’ and ‘flour’ are also etymologically related. However, the senses of ‘corn’ are not, although the word seems polysemous. Considering the opposites can help in making the decision, as in (28).


(28)   fair - dark

         fair – cloudy

         fair – unfair


Since the opposites (antonyms) are very different, this is an example of homonymy, unrelated senses, rather than polysemy. Cognitive semantics, an approach that focuses on the relation between language and the mind, has a more practical explanation of polysemy:


Polysemy arises from the fact that there are systematic relationship between different cognitive models and between elements of the same model. The same word is often used for elements that stand in such cognitive relations to one another. Lakoff  (1987a:13)


        Homonyms and polysemous words have   disambiguated in context. There have been some theories put forward suggesting how rules of discourse constrain this lexical disambiguation. Asher and Lascarides (1995) founded this process on the (1) avoidance of incoherence and (ii) reinforcing the rhetorical connections, such as Narration which says that the order of sentences in the text (or utterance in the discourse) mirrors the temporal order of events. For example, the word ‘bar’ will normally be interpreted as a courtroom bar when it follows such words as ‘defendant’, ‘bailiff’, ‘judge’, as in (29):


(29)   The judge asked where the defendant was. The court bailiff found him slumped underneath the bar .



3.3         OPPOSITION


In addition to inclusion, sameness and sameness of form, it is necessary to introduce opposition or antonymy . Antonyms are words opposite in meaning. There are various types of opposition. There are pairs where the truth of one member as a description of a thing, event, etc. necessitates the falsity of the other, as in ‘dead’ – ‘alive’, ‘male’ – ‘female’ (where applicable). They are called complementary opposites, binary opposites or contradictories . So, (30) is a contradiction.


(30)   The male kitten we have bought is female.


There are also opposites where the gradation of the property is possible, as in ‘hot’ – ‘cold’, ‘old’ – ‘young’, ‘long’ – ‘short’. In gradable antonyms the inapplicability of one term does not imply that the other applies: if something is not hot, it need not be cold; if somebody is not poor, he or she need not be rich. There are terms in-between, such as ‘warm’, ‘cool’. Some pairs  exhibit markedness. We normally use unmarked (31a) and (32a) rather than (31b) and (32b).


(31a)   How old is he?

(31b)   ˀ How young is he?

(32a)   How long is it?

(32b)   ˀ How short is it?


Gradable antonyms are also called contraries . When predicated about a certain object they cannot both give rise to true propositions. However, they can be both false about an object. It has to be noticed that the meaning of adjectives is relative to the whole noun phrase: a big hamster is still smaller than a small crocodile.

       Linguists also distinguish other types of opposites. There are converses , also called relational opposition , where the relation is reversed: ‘husband’ – ‘wife’, ‘parent’ – ‘child’, ‘buy’ – ‘sell’, ‘employer’ – ‘employee’, ‘above’ – ‘below’. So, in converses, if a is below b , then b is above a . formally the relation can be captured as follows:


(33)   for all x and for all y (if R(x,y) then R’(y,x))


Some relational opposites are transitive: if a is above b and b is above c, then a is above c. they are never symmetric because symmetric relations hold both ways as in: ‘a is married to b and b is married to a’. formally, a symmetric relation can be captured as follows:


(34)   for all x and for all y (if R(x,y) then R(y,x))


Further, there is also directional opposition as in ‘up’ – ‘down’, ‘come’ – ‘go’, ‘left’ – ’right’; antipodal apposition as in   ‘east’ – ‘west’; and incompatibility , non-binary contrasts (also called taxonomic sisters), such as cycles (days of the week, months of the year), scales (‘excellent, …, good, …., bad, …’), ranks (‘one, two, three,….’, ‘field marshal, general, …., private’) and other taxonomies, such as colour terms, breeds of dogs, etc.

Mon, 16 May 2011 @11:43




Melawan Kemustahilan










Twitter Facebook Instagram Google Plus Youtube Channel




Copyright © 2019 bejo sutrisno · All Rights Reserved