On Referring by P.F. Strawson
An overview of Strawson’s critique of Russell’s Theory of Descriptions
Introduction
The aim of Strawson’s influential article ‘On Referring’ is to criticize Russell’s theory of meaning as given in his work ‘On Denoting’. He does this through pointing out cases in which Russell’s theory falls short in explaining uses of language. He then offers his own theory as a solution and as a ‘complete’ theory of meaning.
Overview of Russell’s Theory
Read this blog for an understanding of Russell’s theory of meaning, specifically his theory of descriptions — Bertrand Russell — On Denoting
Critique of Russell’s Theory
Strawson questions the grammatical/logical language distinction proposed by Russell. Strawson claims that reference is not an abstract relation between an object and an expression or a sentence. He disagrees with Russell on the need to analyze sentences on the basis of their logical properties. He denied the existence of ‘logically proper names’. Rather, he emphasizes on the need to focus on the conversational use of language.
As per Strawson, Russell, through his theory, implied that for there to be a meaningful sentence, there must be a possibility for the formation of a genuine subject/predicate sentence. The subject should be a logically proper name that designates something.
So, the logical analysis of a meaningful sentence should look like:
Sentence: “The King of France is wise.”
- There exists a King of France
- There is only one King of France
- There exists nothing that is a King of France and is not wise
Since the first point is false, the rest of them are automatically rendered false. (The falsity of one conjunct entails the falsity of the whole conjunction)
Strawson believes that Russell’s theory is flawed because of his failure in understanding the distinctions between 1) sentence/expression, 2) use of sentence/use of expression and 3) utterance of sentence/utterance of expression.
Strawson’s Concept of Meaning
It is because of his flawed conception of ‘meaning’ as a function or a use of a sentence/expression that Russell’s theory failed, claimed Strawson. He makes clear this distinction between mentioning and meaning to highlight the flaws in Russell’s theory.
The Distinctions
Sentence vs Use of Sentence vs Utterance of Sentence
A sentence is an abstract linguistic entity that has a grammatical structure and represents a proposition. As it is a syntactic construct, it can be studied independently of its particular use/occurrence in communication. A sentence has a meaning. But the question of truth/falsity does not arise for it. For example: The dog is wagging its tail.
The use of a sentence refers to its employment within a certain linguistic context. This sentence is being used to convey a meaning, expresses a proposition or facilitate communication. The use of a sentence (or a part of it) is for referring, and thus the question of truth/falsity arises. For example: I am using the sentence “The dog is wagging its tail” to assert a statement about a particular dog.
The utterance of a sentence is that particular instance when the sentence is spoken or written by someone in a particular context. It refers to the action of expressing a sentence in a given situation. For example: I am speaking in a conversation, or writing the sentence “The dog is wagging its tail.”
So, there is never ‘a sentence’ about a particular thing or situation, but only ‘a use’ of it to communicate about the particular thing or situation.
Sentence Token vs Sentence Type
A sentence type refers to a category that contains a shared grammatical structure, meaning and use of sentences. The sentences of the same type have a common pattern as per which their formation/interpretation is determined. For example: All vegan have a dairy-free diet.
Whereas, a sentence token refers to a particular instance or occurrence (within a certain context) of a sentence. In other words, it is the manifestation of a sentence type. For example: A certain instance of writing/saying the sentence ‘All vegans have a dairy-free diet.’
Strawson believes that while it is the sentence type that is meaningful, its the token (or its use) that is true or false. Russell fails to draw this distinction because of his mistake of uniquely existential commitment.
Expression vs Sentence
Expressions refer to any meaningful unit of language (could be a word, phrase, complete sentence). They are regarded as the building blocks of language and differ in complexity and length. Sentences, as mentioned above, are linguistic entities that serve to convey meaning or express complete thoughts. For example: ‘the dog’ and ‘wagging its tail’ are expressions within the sentence “The dog is wagging its tail.”
Strawson claims that expressions do not refer at all. Instead, it is the act of use of the expression by a person that refers.
Therefore, the question of truth or reference mainly arise in the context of the use of sentence tokens. The above-mentioned distinctions help in a clearer understanding of language and its relation to truth, meaning and reference.
Entailment vs Presupposition
Consider the following propositions:
- The King of France is wise
- There is a King of France
As per Russell’s theory, (1) entails (2), whereas as per Strawson’s theory, (1) presupposes (2). The difference between the two is that Russell’s view (that 1 entails 2) means that (1) cannot be true unless (2) is true, and if (2) is false, (1) is false. Whereas Strawson’s view, (that 1 presupposes 2) means that (1) cannot be true or false unless (2) is true, and if (2) is false, (1) is neither true nor false.
Truth-Value Gap
This is how (following from the above distinction) Strawson introduces the concept of truth-value gaps, as per which there are some meaningful statements have tokens that are neither true nor false. A sentence about the King of France made in the 16th century would have been either true or false, but in the 21st century (in which there exists no King of France), any such sentence would be neither true nor false.
Two Tasks of Language: Referring and Attributing
Strawson distinguishes between the two roles of language — referring/identifying something and attributing something. He believes that mentioning/ referring/ identifying is the characteristic of the ‘use’ of an expression and is performed by the subject part of the sentence/expression. Attributing, on the other hand, is the function of the predicate (which has been largely ignored by most philosophers of language). These two tasks (referring by subject and attributing by predicate) are not independent of each other. When the former (referring) fails, the latter (attributing) cannot take place. Here arises the truth-value gap, as has been discussed earlier in this blog.
Strawson holds that there is no relation between knowing the meaning of a sentence/expression and knowing a particular use of it. A sentence can be used to make a true/false assertion if and only if the person using it is talking about something that exists. But if the person is talking about a non-existent thing, the sentence is neither true nor false, as even the question of its truth/falsity does not arise. Thus, Russell’s Theory of Descriptions gives an incorrect explanation of reference failures.
Conclusion
It is because Russell confuses between sentence type and sentence token (or between sentences and their use), and between referring and attributing, that he gives an incorrect account of the use of language. Strawson contradicts Russell by discarding the importance of considering the logical properties of sentences. Instead, he believes that in ordinary language, referring is an act that the speaker performs, without any notable consideration for logic.
There has been some criticism of Strawson’s arguments for placing too much emphasis on the descriptive content of names and for neglecting the primary function of names as tools for reference. Nevertheless, it remains an important contribution to the debate surrounding meaning and language.
