Different Kinds of Synonymy in Language

Table of Contents

I. Introduction

II. Problem
II.1 Absolute Synonymy
II.2 Complete Synonymy
II.3 Cognitive Synonymy
II.4 Plesionyms

III. Summary

IV. Literature cited

I. Introduction

I asked several of my friends what they thought synonyms were. Nearly everyone said that in their opinion, synonyms were at least two words which meant the same thing. This statement, nevertheless true, oversimplified matters in the extreme. Their definition would only refer to absolute synonyms (this term will be explained later). This papers intention is to give an overview of the problems we encounter when dealing with synonyms and to tackle the problems in determining synonyms.

II. Problem

To begin with, let us take two firm semantic insights. First is, that there are pairs of words which have a special semantic relation. They bear resemblance to one another in a way, so that they are interchangeable in several contexts. Usually, these words are called synonyms. The second is that there are obviously words that are more synonymous than others. For example let us take settee and sofa, which are more synonymous than cunning and savvy. So synonymy seems to be a gradable phenomenon; thus introducing at least two different kinds of synonymy, absolute synonymy and complete synonymy. I decided to focus on absolute synonymy first, because the problems of determining synonymy will become more obvious.

II.1 Absolute Synonymy

Absolute synonymy means, that a pair of lexemes is absolutely interchangeable in all imaginable contexts and that they have the same ratio of distribution. The first aspect means, that for two lexical items a and b, if they are to be recognized as absolute synonyms, a has to be normal in a context in which b is normal and b has to be completely abnormal in a context in which a is. This requirement is passed by very few pairs. The following test gives evidence for this (‘+’ indicates “more normal”; ‘-‘ indicates “less normal”):

1a smoke vs. fume

I’m going to smoke a cigarette. (+)

I’m going to fume a cigarette. (-)

1b big vs. large

He is a big boy. (+)

He is a large boy. (-)

1c knock vs. cream

He knocked him unconscious. (+)

He creamed him unconscious. (-)

1d almost vs. nearly

She looks almost Chinese. (+)

She looks nearly Chinese. (-)

This hints to the fact that there are presumably always contexts in which a lexeme is less common than the other. There are however words which could be accepted as absolute synonyms. For example airport and aerodrome which co-existed for a while in the English language. But due to the economy principle aerodrome has become obsolete altogether. The economy principle leads us to the idea that two words with the same reference object cannot co-exist for long in one language (except with cognitive synonyms which I will tackle later in the text), therefore weakening the idea of absolute synonymy considerably. Another possibility that could come into existence is that one word develops other nuances in its meaning which would mean that the lexeme pair is not synonymous in every context anymore.

If there should be a pair of lexemes that have exactly the same meaning in all contexts in spite of the economy principle (which I doubt heavily), there is always the second point that needs to be fulfilled to qualify them as absolute synonyms. The distribution ratio of the two lexemes must be the same. This requirement however is much harder to test, because language is not static. The methodic problems of grasping a developing test object (language) and the size of the sample that has to be taken to make such an analysis valid (the sample must cover every social group, in every part of the terrain that speaks the relevant language) are enormous indeed.

II.2 Complete Synonymy

Let us take a look at the pair almost and nearly again. As it was shown before, this pair can’t qualify as absolute synonyms. But in several contexts they seem to be completely interchangeable.

2a

I’ve almost succeeded in winning the race.

I’ve nearly succeeded in winning the race.

2b

He almost made me cry.

He nearly made me cry.

In this context both lexemes are equinormal, they are complete synonymous. This synonymy however is context restricted because, as we have shown earlier, there are contexts in which they are not completely identical in their meaning. Context restricted synonymy is relatively rare but it exists. But can we say why some lexemes are identical in one context and one of them is odd in the other? It is even hard to tell whether there is a difference or not, and to make out what the difference is, is not always clear. One is tempted to say, that it is due to subtle nuances that these differences exist. Let us try to pin down these subtle nuances. How can we determine what these nuances are? One way to find out is the use of componential analysis. In componential analysis we try to find characteristics that can be attributed to lexemes. The idea is that synonyms should have the same characteristics. If they haven’t they’re not absolutely synonymous. The more characteristics they have in common, the higher the semantic overlap is, the more synonymous they should be. But is this truly the case? The following example will give us some insights.

3a

Abbildung in dieser Leseprobe nicht enthalten

3b

Abbildung in dieser Leseprobe nicht enthalten

3c

Abbildung in dieser Leseprobe nicht enthalten

Example 3c has more characteristics in common than example 3a, so it should be more synonymous. But this isn’t the case. There is not one context in which bird can be interchanged with cat and that will keep its original meaning in spite of their high semantic overlap. No matter how finely we will sub-divide classes, the pairs will not become more synonymous. Why is that the case? The answer is that synonyms not only need a high semantic overlap, they also need a low implicit contrast. One function of the word bird is to exclude any other closely related items, in this example cat.

Another reason why componential analysis is not useful to determine synonyms is the choice of attributes that should be involved. What differences would still be allowed to have a high semantic overlap, what differences make a low implicit contrast? In other words, in what traits must the contrast be low in what traits the overlap high? The answer is as simple as unsatisfying. Complete synonyms are lexemes that are identical in central semantic traits and differ in minor and peripheral ones. This statement, intuitively true, is yet very hard to describe because there is no scientific evidence to determine whether a trait is a central or a peripheral one. There are however signals with which we can recognize complete synonymy. Let us take a look at some examples.

4a

She wore black trousers, or rather shorts.

We were eating cereals or, more exactly, oatmeal.

I was riding a steed, that is, a gelding.

When these specifiers are used, it is almost sure that we deal with synonyms of a kind. These specifiers can’t be used when we deal with words that differ in one central trait or are even antagonists.

4b

? He was beaten to death, that is, life.

? I was riding a steed or, more exactly, a bike.

All these examples show us, that there are pairs of lexemes which are obviously more synonymous than others. This leads us to the assumption of a scale of synonymy. But to determine a scale, we need at least one fixed end point. Since we already presumed that absolute synonymy is very hard to show if not impossible, the endpoint has to be zero-synonymy. But zero-synonymy is no unitary concept. The first point is c heap vs. expensive and white vs. fast are surely no synonyms but for different reasons. And the second thing is that it is impractical to draw a line between synonymy and zero-synonymy.

5a

creep vs. crawl; creep vs. cringe; creep vs. sneak; creep vs. grovel; creep vs. toady.

Where in the aforementioned example ends synonymy and begins zero-synonymy? The scale is therefore impossible to establish.

II.3 Cognitive Synonymy

Cognitive synonymy (sometimes called descriptive synonymy, referential synonymy or propositional synonymy) is a special relation between at least two lexemes. Examples of cognitive synonymy are: fade, die, decease, nibble off, kick the bucket. These expressions can all be used in the same context without changing the truth-conditions of a sentence. They are however no absolute synonyms because the distribution is not the same for all of those lexemes. There are apparently social taboos, which prevent them from being absolute synonyms. Imagine the judge in court speaking of nibble off instead of die. Even the membership to a particular social group can be deduced by the use of certain words instead of others.

Another example of cognitive synonymy is:

6a

This ice cream tastes good.

Mmmh.

The content of these utterances perceived in the same way. The difference is in the way in which they are expressed. The first sentence in this example gives us the information in semantic means. The message is delivered through the meaning of words. This is called propositional mode. The second utterance conveys its meaning through an expression and is therefore called expressional mode.

II.4 Plesionyms

Plesionyms, or near synonyms, are words, that are almost synonyms. They are distinguished from cognitive synonyms by the fact, that the connotations of the pairs are different and therefore they yield different truth conditions in a given context.

7a

XY was freedom fighter.

XY was a terrorist.

These sentences have different truth conditions, though they are referring to the same subject. The difference lies in the pragmatic or interpersonal connotations of the user.

7b

It was misty last Friday or, more exactly, it was foggy.

It was foggy last Friday or, more exactly, it was misty.

In this example it is shown, that it is nearly impossible to distinguish between the two truth conditions. Both sentences are referring to the same thing, but they seem to exclude each other, though it is impossible to tell what the difference in the truth condition may be. This hints to the assumption that plesionyms rely much more on personal impressions than on actual states of reality. Though there is a difference between the lexemes foggy and misty, this difference exists only in connotation and the speakers expressed attitude. Other examples point to that observation.

7c

We weren’t clobbered, we were battered.

We weren’t battered, we were clobbered.

7d

I was not ravished, I was delighted.

I was not delighted, I was ravished.

IV. Summary

I tried to show with this term paper, that the linguistic phenomenon of synonymy is a very vague field. The problem we encounter in this field is the development of language which makes the existence of absolute synonyms doubtful. Few significant statements can be made due to the lack of an unambiguous testing ground and the fact that language is arbitrary in its use. Complete synonymy is intuitively a fact, but when trying to find a scientific method to generalize a rule for it, we see ourselves confronted with the problems of searching traits that could make up a pair of synonyms. The same counts for cognitive synonyms as well. The boundaries between two lexemes are blurred and to find even an intuitive distinction between synonymy and zero synonymy is harder than one would imagine. With plesionyms we have the problem of interpersonal differences when speech is used. Strictly, plesionyms can’t be synonyms, because they could refer to different truth conditions, thus they can’t have the same meaning. If the problem of synonymy is to be solved, the problems of meanings would have to be solved first. To establish a connection between lexemes such as synonymy, one would truly have to know all the nuances that one words implements in all uses. The question here is not only if the same reference-object is termed or if the lexemes are interchangeable, but a much more subtle one: How are words perceived and why do individuals construct analytical truths that build up these nuances of differences in words. In other words, we would have to know the actual usage of words in all their details. But this is highly impractical because words win their nuances through use and are not used because of their nuances.

IV. Literature cited

Cruse, Alan. (2000). Meaning in language: An introduction to semantics and pragmatics. Oxford: University Press

Cruse, Alan. (1986). Lexical Semantics. Cambridge: University Press

Lyons, John. (1981). Language and Linguistics an Introduction. Cambridge: University Press