Pathways to PhilosophyKindle eBooks by G Klempner

on this page

Or send us an email

Application form

Pathways programs

Letters to my students

How-to-do-it guide

Essay archive

Ask a philosopher

Pathways e-journal

Features page

Downloads page

Pathways portal

Pathways to Philosophy

Geoffrey Klempner CV
G Klempner

International Society for Philosophers
ISFP site

Home   Woods 1   Woods 2   Woods 3   Woods 4

pathways (essays)

Justin Woods

Naïve Symbolism and the Metaphysics of Meaning

"There are no facts, only interpretations" — Nachlass Friedrich Nietzsche


In this essay, I shall expound the naïve view, or theory, of symbolism, which assumes, or argues, that in the symbolic system of language thoughts are expressed by words, and that words have meaning, thereof.

I shall show that the naïve theory of symbolism is invalid, and offer an alternative view incorporating my own empirical theories of meaning and language. I shall also argue against the naïve view of such a relation as 'expression' in the context of a symbolic theory of language. I shall further assert that meaning cannot be a property of words, or any linguistic symbol.

In naïve symbolism, 'expression' is the relation connecting thoughts and words, allowing us to compile phrases such as, 'thoughts and their expressions', where 'their' implies a possessive quality that thoughts and words share.

Here, the words are possessed by the thoughts; words belong to thoughts, and have a causal relation. It is statements such as these that I will pay philosophical attention to. I shall be considering whether such a relationship is possible, how it comes about, and the logic of such a relation.

I will show that there is no such causal relation, nor logically mutual dependence between thoughts and words. Further, I will argue it cannot be said, with any logical validity, that symbolism in language implies a possessive relation such as, 'thoughts and their words'. Thoughts are not in the possession of words, for this implies an illogical causal relation.

The term 'expression', and all its conjugated forms, is bogus in the naïve position it plays between words and thoughts. There are thoughts, there are words, or symbols. However, the two are neither logically related, nor mutually dependent.

The conclusion of this essay will be the argument, logically and philosophically defended that the game of language we all share in order to communicate and understand each other, is a guessing game. The thoughts of others cannot be known through language, only pointed to. Expressions are merely indicators of meaning. Understanding, in the generally accepted sense, never happens.

Finally, a piece of speculation into the theory of language development.

Much of our ability to understand and use language comes from the tools we attained in our formative years. One such tool was the copy-cat style of learning the empirical expressions.

Here, phrases are recognized and repeated in similar situations. We attribute empirical phrases to our thoughts by way of observing similar social situations, which are accompanied by other empirical phrases. The implications of such a method of attributing such phrases to thoughts would obviously be flawed by its inaccuracy. I put it forward that such a situation is exactly the current state of language.


It is naïve to propose that thoughts can be expressed by words, as though the thoughts of others can be understood, without any logical argument to keep such a proposition valid and unassailable. However, this is the view held by the naïve symbolist.

Naïve symbolism is, in my opinion, the most commonly held view amongst the general public, as is seen by its promotion in teaching and everyday life. It is also a rarely contested and unspoken view held in philosophy, as well as the fundamental tenet of non-behavioural psychology. The importance of such a target of philosophical attention is made the clearer by this lack of contention and dispute.

Naïve symbolism is the idea that thoughts are symbolised by words; words represent thoughts. A comprehension of linguistic symbols logically entails a comprehension of thoughts. This so, due to the posited causal and possessive relations falsely held between thoughts and words.

The consequence of this mental manifestation is that it must be at least possible to understand the thoughts of another person- where understanding is synonymous with knowing, or having the meaning. So, statements are made such as, 'I know what they're thinking', 'I know what you mean', 'I understand you'.

In other words, naïve symbolism assumes what the mind thinks can be manifested physically, and thereby recognised sensibly as a thought; a sensible form can symbolise the thought, a word represents meaning.

If symbolism is the representation of thoughts, then a causal relation is logically implied between the two. This implication is reinforced by the acceptance, in general use, of phrases describing the 'meaning of words', as though words possess meaning.

The logical possibility of this implication will be shown invalid in this essay.


What is meaning?

I contend that meaning is the private thought we attach to expressions. In naïve symbolism, meaning is attributed to expression- words have meaning, as do phrases, sentences, and books.

I must posit that meaning is no such attribution of expressions. For, in order to know the meaning of a word would logically entail the existence of thought outside thinking, which is impossible. This is so, because meaning is substantially thought. It could not be anything else, if we look at the roles available for terms to play in a theory of language. It could not be objective, or a sensible thing, as these roles are filled by 'expressions'.

Expressions are objective in that their knowledge is shared, and therefore empirical. To say that meaning is not substantially thought would be to deny its subjectivity, since no one would deny the subjectivity of thought.

How we attach thought to expressions, I argue, is by associating them together in memory. That is, when we remember one, we also remember the other. This is a somewhat speculative idea, but I am compelled to expound it.

So, to continue, when we want to say something, we are reminded of the associative expressions of our thought. We choose the best ones and express them in linguistic units: words, gestures, phrases, etc.

Meaning is, substantially, thought. Expressions are public. The former, the product of a mind, the latter, available for all to sense, and thereby know.

If meaning is, as we have seen, subjective, and expressions objective, then language conveys messages that none can ever know. For, language does not convey the meaning of the expressor. Meaning is the thought of the expressee, when they perceive the language.

Language is the system of expressions known objectively. Yet, language is what we use when we mean something and want others to know what it is that we mean. Since we have already seen that meaning is private, we must admit that our expressions will not transfer our meanings. Meaning is the thought that occurs prior to an expression, from the expressor, and after the perception of the expressee.

Meaning and expression, therefore, are mutually independent. Furthermore, words are meaningless.

So, words and phrases are not expressions that stand for what I think when I say them. They do not represent what I mean. This view is subjective; I do not, nor cannot, have an objective view of any word. The meaning is composed of my thoughts about an expression, and therefore subjective.

A consequence of this is the fact that when someone refutes what I have said, they are doing nothing more than refuting their own view of the words that I have used to express my thoughts. When someone disagrees with you, they are technically disagreeing with themselves.

Although I submit that the individual, to express his thoughts, uses words, I do not submit that this is actually done. It is not done. His thoughts are not expressed, in the sense that they take the form of sensible objects. Nor is the thought expressed in the sense that it has a causal relation with an expression. Thoughts are only expressed in the sense that noises and gestures are made at the time of thinking.

An expression is desired, learned words and phrases that have in the past been associated with the event expressed come to mind and a group of them are chosen and then pronounced. The fact that a choice of 'expressions' can be made goes to show that if any relation exists, it does so between words and words. The choices of expression means that any expression p, q, r... can take on the role of expressing a given thought x. This proves, logically, the mutual independence between words and thoughts. If they were mutually dependent, then they could not be interchangeable. Interchangeability between expressions and meaning is possible only if there is no logical dependence between one and the other. In language it is possible to interchange words without altering meaning, because meaning comes not from the words but from the person thinking about the context of the language given.

So how can I propose that words have no meaning yet write this essay, which contains these meaningless words?

The answer is simple: neither I nor anyone else has a choice.

Understanding the thoughts of others is not logically possible in the sense that we understand our own thoughts: we cannot have any access to the thoughts of others as we do our own. Any attempt at understanding what others mean, when they express their thoughts, is merely guesswork.

It is guesswork made more easily by common experience and the context surrounding the expression. This occurs within groups who are taught the same words for the same situations. They are taught as they grow up together with their linguistic teachers- such as schoolteachers and family- that also grew up together.

I do not propose that thoughts cannot be expressed. I propose merely that the expression has no causal, nor possessive, relation with the thought. Expressions do not have meaning.

We understand our own thoughts. This is a tautology. Understanding your own thoughts, is simply thinking. All logical propositions are tautologies. Understanding them is a matter of thinking them. There are no true or false tautologies. Just as there are no true or false desires. I desire x, or I do not.

As to the expression of logical propositions, their meaning does not subsist. Logical thinking does. When I read a logical proposition, I attribute my own meaning to it, as with empirical and conceptual expression. The statement, 'all black cats are cats', has no meaning until I give it one. The only difference between logical expressions and the others is that I think logically in order to attribute meaning to them.

What makes logical expressions certain is their defense against contradiction. However, certainty in logic is achieved through the agreement that one thing has certain given properties, and classifications. This consistency ensures certainty. Once agreement is shared, contradiction is a mere breach of such an agreement.

For example, if I agree that Socrates is classed as a man, and if I agree that under the class of mortal things is the class of men, then I have already agreed that Socrates is classed as a mortal thing. My contradiction of this could only come about by my not being immediately aware of that which I already know. This is entirely possible, as it takes some time for memory association to remind me of the contents of each class, and then the ability to know which things share common properties. Therefore, given sufficient time, no logical proposition would ever be contradicted.

So, meaning is mutually independent from logical propositions. I attribute meaning to the expressions. Their validity is merely their agreement with you that certain things share certain properties.

Understanding the metaphysics of meaning, I now know necessarily that my thoughts are not being known by the other when I attempt, and ultimately fail, to express them. Nonetheless, I have grown up expressing my thoughts and have decided that this fact does not affect my life with other people any more than their lives are affected: we all live in a similar world, after all. Accuracy in the expression of thoughts is not necessary in order to live. We all get along in the world, without too much disaster, even with the many linguistic mistakes we make.

So, if you want someone to know what you know, if you want to let someone know what you think, then you will never be satisfied: it will never be done. Instead, you must rely on the feeling of satisfaction, which irrationally tells you that the recipient of your expressions understands you.

The problem with the metaphysics of meaning is not what is so obvious, i.e. that a thought is not its manifestation. This is very much taken for granted, but it is well held in the world that the thought 'egg' and an actual egg are clearly not identical. No. The problem with symbols and what they symbolize is how is it possible that they symbolize at all. The solution to the problem is that symbolism does not occur. It does not occur, because the meaning cannot be represented, since this means a causal relation.

That symbolism is a legitimate function in language is based upon this proposition of causal relations between thoughts and words, meanings and expressions. Such propositions are held by the naïve theory of meaning, and will be discussed in the next chapter.

A proposition without your meaning cannot be a proposition. All propositions are so, if and only if, they are proposed out of a meaning, i.e. someone's thought.

The conclusion is that words do not have meaning. Meaning has a metaphysical source, the mind, and is not tied in with any logical symbolism.

Meaning is metaphysical in the sense that it exists in our minds, beyond our reach, beyond our ability to express it. Its direction may be pointed to, or it may be shown, in the Wittgensteinian sense, but it can never be expressed.


Now, there needs to be more said of causality and the possessiveness we attribute to words as having of thoughts.

What, in other words, are the conditions under which a particular thought is related to a particular word, whereby that word is said to 'belong' to its thought?

Answer: The conditions must be causal. That is, X could not have happened without Y.

If each thought has its own expression, or set of expressions, then we can say that those expressions belong to that thought. This implies that words are related to thoughts causally; one belongs to the other, one causes the other. Namely, expression could not have happened without thought.

The naïve view holds that when I speak, the words are expressions of my thoughts. This also implies a causal relation between thoughts and words. As the thoughts arrive, I am able to manifest them in speech by using those expressions that best suit my thoughts.

Bertrand Russell "The Analysis of Mind"

In Russell's early theory of meaning, he stated, 'The meaning of a word is an entity, the entity for which the expression stands.'

I disagree. The expression is used as a result of one of my thoughts; thoughts which will never be the same. Therefore, there is never one particular meaning for an expression, in so far as the meaning is that which is thought. Nevertheless, I do have the thought first and then my memory gives me the expression closest to it by memory association. So, therefore, there are two meanings: what I thought, or meant, and what the expression means.

The meaning of a word, by Russell's theory, if an entity, is two entities- logically distinct, mutable, transient and interchangeable.

I alone, however, attribute meaning to a word that I use. Other people also attribute their own meanings to that same word, in order to understand it. There can be no single, objective meaning of a word, because who has that meaning? If there were such an immutable entity, then no one would ever know what it was, since it would have to be expressed by someone who hadn't attributed their own meaning to its expression.

Therefore, it cannot be said to be known that 'the meaning of an expression is an entity.'

Russell states as part of his theory of descriptions that the meaning of a word or phrase is to be found in the taking of the proposition as a whole. I disagree with this theory, also.

I also suggest that Russell's entire theory of descriptions is fundamentally flawed because of this thesis.

Wittgenstein states in the Philosophical Investigations that 'the meaning of a word consists in its use.'

It is neither. The meaning of a word is not objective. It is purely subjective. It is to be found in neither objective, general holism nor particular use.

The meaning of anything is only to be found in the mind of the speaker and not in the symbolism used. The symbols of language do not have meanings as part of their properties, as cars have of wheels, as apples have of taste. Symbols are associated with thoughts in a parallel relation, not a causal one. In this sense, symbols do not symbolize; nor are expressions expressed.

Symbols are not mutually dependent upon their meanings. The meaning is the thought, and it is logically independent from any expression.

In theory, this mutual independence allows for meanings, or thoughts, to be interchangeable with expressions, even antonyms. In practice, this often happens.

One scenario where this interchange occurs is in the telling of a lie. Here is the obvious example of a word, or expression, having an opposing, contradictory or wrong meaning attributed to it. This would not be possible if words/expressions were logically dependent upon thoughts. Nor would it happen if there was a causal relation between the two.


It is naïve to assume that you have understood someone from what they have said to you- to know what they mean, in other words. Consider the following problems.

When I sit down to breakfast and my wife asks me if I should like some tea, I hear the expression, 'Would you like some tea, dear?'

Firstly, I can anticipate this question's meaning, before it is expressed, because I have a habit of taking tea in this routine manner. The meaning is understood prior to its expression.

So, what about the first time you were asked? I hear you say. The question, at its first inception, could have begun, 'Would you...', and I could have guessed the rest. It is a question, since it begins with 'would you'; the tone is that kind of tone, which is used to sound inviting; I haven't had tea yet; Tea is usually asked of one in the mornings, under these circumstances, as to be inviting.

This guesswork is the technique employed in establishing the meaning of expressions.

Secondly, the general and frequent use of a stock of expressions enables one to quickly guess at that which the other intends us to understand. If I ask, 'How was your trip on the bus?' I can be assured that either answer will probably come from the stock of answers one acquires, most of which share the same meaning. For example, 'Good thanks', and 'Not bad at all' are equal in meaning under the above circumstances, but symbolically, make respective positive and negative propositions. This different meaning was not sought after, and so was not understood.

How do I know that two expressions share the same meaning? I don't. I am the one who decides on the meaning of an expression. Meaning is subjective, remember!

Thirdly, often one encounters moments when a particular word could be replaced by a number of others, without changing the meaning or validity of the statement. I do not refer to the synonyms here. For it is not the objectivity of the word that allows us to hold any meaning, or validity. It is the context in which the expression is given.

For instance, if a telegraph pole falls on top of me, whatever I shout will be understood as a cry for help, whether I say 'Help!', or 'Bananas!' Although, one would not generally consider that a synonym for 'help' is 'bananas'.


What follows is an alternative account from the Expressive Theory of the nature of thought, and also a defense against the argument that my theory implies or entails a 'private language', in the Wittgensteinian sense.

But, firstly I would like to make a point against the argument that thoughts are not private.

The Privacy of Thought

If thoughts are not private, then they are public, and thinking other's thoughts is at least possible by definition. However, asking for the meaning of an expression, when the author is standing in front of you and capable of communicating in the conventional sense, certainly implies the privacy of thought.

Asking for the meaning of x is expressing the desire to think that which is thought by whomever expressed x. If it is not this desire, then the desire must be for an objective meaning. However, it cannot be to ask for the objective meaning of x, for this entails objective thought. Only thought can lead to the expression of a word. If the meaning is objective, then the thought must be objective. If the thought is objective, then the thinker is objective. An objective thinker is a logical impossibility, since the very term 'objective' is used to distinguish those things which are 'external to the mind' from those which are not. (The Concise Oxford Dictionary, Ninth Edition)

A further argument, supporting that one cannot think the thoughts of others, is the following: If one could think the thoughts of others, one would not need a means for communication (this role being taken by common/public thought). There would, then, be no possibility for a language. The fact that there are languages entails the impossibility of thinking other's thoughts.

Since one wishes to think the thoughts of others and cannot, one can never know whether one has thought similar thoughts as others. Certain indicators point to one's accuracy at guessing which thoughts are closest to those associated with each expression. The accuracy of understanding is measured by the degree of feelings of satisfaction felt after the interpretation of expression, which may have been modified by the author to assist one.

The Nature of Thought

It will be shown from a philosophical analysis of the nature of thought that meaning is a property of thought.

I take my meaning of 'thought' from the only way possible: my understanding. Since I must attempt to translate my meaning into language, I am bound to discuss my meaning of 'thought' in such a way that appears I am contradicting myself. However, I must write, and I am satisfied that my writing on 'thought' will stimulate thought in the reader that may possibly lead to an understanding of 'thought' in the way I understand it. Though neither of us would ever know, it seems that satisfaction is sympathetic, and we induce that the gestures and indicators of satisfaction in the recipient reflect the understanding of the expression; although, this is speculation, and I am not required to validate such a position.

Definitively, 'thought', as a noun, is the subject matter of thinking. As a verb, it is the past participle of the infinitive 'to think'. Empirically, by my observance of its common use, 'thought' is categorically related to memories, emotions, and the Kantian judgements (ostensive thought immediate to linguistic sense-data). 'Thought' is also often synonymous with 'idea' (my understanding of 'idea' could take the place of 'thought' in this paper, but 'ideas' can have other uses that 'thoughts' do not).

One thinks of the class of thoughts, memories, emotions and judgements. If thoughts are the class of memories, emotions and judgements only, then there can be no talk of the subject matter of deliberation, or problem-solving. So, thoughts share the same internal category as with emotions, memories, and judgements. This gives us a name for the subject matter of our various mental efforts: that is, 'thoughts'.

I infer from this position that thinking is the conscious mental effort upon memories, emotions, judgements, and thoughts, and that further thoughts, memories, etc., can in turn lead from any of these.

Of the properties of thoughts there is that of 'understanding'. If one thinks of x, where x is any expression, the thoughts that result lead to a variety of states of understanding. Some thoughts, for instance, are not as well understood as others. It is, then, possible to have thoughts that are 'not clear'; no-one would doubt this. We can pay such thoughts more thinking time, and often with a little extra knowledge, we can improve the degree of our understanding of a thought.

Thinking is a continuous chain of thought in a conscious mind. However, that which is thought depends on what there is, empirically or rationally, to provoke its existence. I must make it very clear that by 'provoke' I do not imply a causal relation. I have used the word 'provoke', as it is associated with 'stimulate' to mean that which 'calls forth', or 'stimulates', thought in general- any thought, and certainly no thought in particular, which is the naïve symbolist view.

Among the things that 'provoke' a chain of thought are the linguistic symbols of an attempted expression of thought (i.e. words, phrases, etc.). Not understanding such symbols is identical with not knowing what the whole expression 'means'. Meaning is not identical with understanding, but the state of a thought about an expression that has been accompanied by a satisfactory level of the removal of doubt about said expression.

In other words, from the instant that each expression is first known (Kant's 'judgement') a degree of understanding is attributed to that expression, by the recipient. Thinking is stimulated, and thoughts provoked. At a given point in thinking, understanding develops; where the relief of doubt is apparent the expression is mistakenly said to have 'meaning' (meaning is really the thoughts that occur at the time of the expression). Meaning, therefore, is that property of thought provoked by expression that has attained a level of understanding identified with the removal of doubt. The remaining of doubt leads to questions such as, 'What does that mean?', or 'I don't know what you mean'.

We have agreed that thought is private, and that meaning is private. Meaning is private because, as has been shown above, it is a property of thought. If others can share one's meaning, then the meaning is public at the price of others thinking one's thoughts.

However, by this association I do not suggest that thought itself is identical to meaning, but that the thoughts that are provoked by the linguistic symbols (words, phrases, etc.) are identical with the proposed meaning of those linguistic symbols.

Neither do I identify meaning with understanding. Understanding is the state of mind reached when we feel a degree of certainty that a word's correct mental association (thought) has been discovered. It is possible to have a Kantian judgement, but not to have understood it. Understanding involves the provocation of memories and mental associations with the sense-data of the word. From memories of past uses of the word, a common denominator becomes apparent which is considered as a general meaning for the word in general use. However, other associations are considered as part of the meaning of the word depending on the context of the word, and situation in which it is used; such as 'apple', with 'apple pie'. This searching the memory for associations takes time and begins with an uncomfortable mental feeling, which we call 'confusion', or 'dissatisfaction'. Since the feelings are associated with language, we change the descriptions from confusion and dissatisfaction, to 'misconceive', 'misinterpret', and 'misunderstanding'. Understanding ends with the relief of these feelings. Meaning is the name given to this level of understanding, which some then mistakenly attribute to expressions.

A person's meaning of a word cannot be known, but can be guessed from the context, or situation. However, since guessing is our only key to reaching the meaning of another person's expressions anyway, this is not so serious. To make our guesses more accurate, as we must do when learning a new word, we need to find associative words whose meanings are already known to us. For instance, a teacher will answer a student's question of what a word means by supplying the student with another word, which the teacher assumes the student has a meaning of:

Student: 'What does courageous mean?'

Teacher: 'Courageous means brave.

But how do we know the meaning of 'brave', and especially if no other expression is associated with 'brave' that we, yet, know of? We must construct all of our meanings from 'fundamental meanings' which could only be either discovered empirically or are a priori. There must be a 'common ancestor' in thought, from which all 'thinking life' have evolved. However, this is speculation far more daring than I wish to enter upon, and its resolution is not necessary for the validity of my arguments.

Argument Against Private Language Implication/Entailment

Uttering words in your head is nothing more than remembering them; remembering their shape or noise. It is possible to extract meaning from such an utterance, whether it is uttered 'in your head' or aloud. It is also possible to utter words aloud without knowing the meaning of the words. It is, therefore, not enough that they be uttered, internally or externally, for meaning to be established. Some mental activity is required to be applied to the expressions in order that they can be understood, and thus have meaning. Meaning is the result of these efforts to make sense of- understand- the expressions.

If a private language is the thinking involved in establishing the meaning of an expression or other thoughts, then private language is obviously possible. However, if you agree that language is the system of symbols and their rules used to express and interpret thoughts, then there is no sense in a private language, since no-one would admit that they do this when they 'talk' or 'think' to themselves. Thinking, or judging, are not languages; we do not use language to express and interpret what we are trying to say to ourselves. We do not 'think' with language. The only exception is 'thinking of', or remembering, language; as we do when recalling conversations, and when reading. There are no mental sentences other than the memory of a sentence. Thought has no grammar, no syntax, no expressions let alone sentences. Therefore, there is no mental language, whether private or public.

So, in this case it is no matter if a Private Language is implied, or entailed, by my account of thought, since it is either just another name for thinking, or it cannot be a language; just as 'welding' cannot be an elephant.

Logical Entailment and Implication

A sentence logically entails another by the one being part of the definition of the other. A sentence implies another when this is possible but not necessarily part of the definition. For instance, 'My cat got run over yesterday', entails that I had a cat, but only implies that the cat is now dead.

However, this relation of entailment is not possible with the meaning of a sentence. The meaning of a sentence consists in the thoughts that precede its construction. Entailment would suggest that from a person's sentence, a corresponding thought could be logically deduced, since what is meant by the sentence includes a further meaning that the person using the sentence must make.

So, there can be no argument that my theory logically entails or implies anything. Sentences are entailed, not thoughts. Entailment, as with meaning, is in the mind of the beholder.


Thus, I have shown that the relation between thoughts and linguistic symbolism is neither mutually dependent, nor causal. I have shown that there are no such logical relations between words/expressions and thoughts/meaning. I have suggested the empirical theory that the relation is memory associative, or parallel.

I have shown that the metaphysics of meaning is substantially distinctive from that of the symbolism of language.

Meaning is the content of an individual's thought if it is to be anything at all. Words do not have meaning, people do. Minds have meaning. Therefore, meaning is subjective, and words are objective. The implication of this argument is that understanding never occurs, in the sense of knowing the meaning that preceded the expression. This means that the attempt at understanding is guesswork. We do not know the meaning of any expression, we guess what it could be from the indicators, such as memories and context.

In conclusion, the meaning of a word is not possible. Meaning and symbolism have a parallel relationship — they never meet.