Handbook of epistemology/Speech

From Wikibooks, open books for an open world
Jump to: navigation, search

What is speech?[edit]

In the beginning was the Word,
and the Word was with God,
and the Word was God.
He was with God in the beginning.
Through him all things were made;
without him nothing was made that has been made.
In him was life,
and that life was the light of all mankind.
The light shines in the darkness,
and the darkness has not overcome it.

(John 1, 1-5)

Everything speaks[edit]

If one hears the concept of speech in its broadest sense, everything speaks. Life is a speech. Even being, just being, is a speech. To be, for all bodies, living or not, is to act on other beings. All material beings do it, simply by being material. To be visible, or audible, or to be perceived in any other way, is always to show oneself, and that is always to speak. Living beings perceive other beings and live by reacting to what they have perceived. Their reactions show how they understand the speech expressed by the beings they perceive.

Animal communication[edit]

In a more restricted sense, animals speak only when they emit signals so that they are perceived and influence those who perceive them. They show themselves to the eyes which look at them, or they make their presence felt in another way. It is not necessary to assume that the animal gives its signals voluntarily. A wasp indicates by its yellow and black robe to a possible predator that it is a dangerous prey. Evolution by natural selection is sufficient to explain such animal communication. Wasps have an advantage in showing their dangerousness and predators have an advantage in perceiving it. There is no need for will or conscience. Blind instincts are sufficient to establish this communication.

Influence the imagination and the will[edit]

In an even more restricted sense, animals speak only if they emit signals because they want to influence the imagination and the will of those who perceive them. This assumes that the animals which emit and receive these signals are capable of imagining and wanting, and especially that they have the knowledge of the mind, that is, they know themselves and others as beings who imagine and want.

Speech is the voluntary emission of signals to influence the imagination and the will of those who receive them.

Meaning through imagination[edit]

When one understands a description, one imagines what is described. Words and verbal expressions arouse the imagination as soon as we understand their meaning. One imagines what is described when one simulates its perception, when one activates, in simulation mode, the detection systems which would be awakened if we perceived what is described. When concepts detected by our perception systems are associated with verbal expressions which name them, we can both describe what we perceive, by naming perceived concepts, and imagine what is described by simulating the detection of named concepts (Saussure 1916).

Silent knowledge is the knowledge that precedes speech and results from perception, imagination, emotion and will. It can be translated into words as soon as the detection systems that it uses are named by verbal expressions. Descriptions are then a translation into words of the knowledge of perception and imagination. The laws which relate the description of conditions to the description of consequences are a translation of silent inferences. A reasoning which combine these laws is a translation of a sequence of silent inferences. In this way, silent knowledge can be translated into words, and thus communicated.

Silent knowledge is fundamental to the development of reason, because talking knowledge begins as a translation of silent knowledge. It can then fly on its own because it can speak about speech, but it needs silent knowledge to take off, because words must awaken the imagination to make sense.

A verbal expression names a concept when it is associated with the concept's detection system. Such an association is an interpretation of the expression. The same verbal expression can be interpreted in many ways by being associated with different detection systems.

The understanding of speech can be conceived in a purely passive mode, as if the words were the notes of a score and the understanding, the music performed by an inner mechanical piano. But imagination is not only awakened by words in this passive way. The understanding of speech is also and especially active. We understand because we want to understand, and what we understand, that is, what we imagine, depends on our expectations.

Understanding words means knowing how to use them[edit]

Any way of using speech as speaker or listener is a way of understanding it (Wittgenstein 1953, Turing 1950). Understanding a language is nothing other than its use. The speaker understands what he says when he knows what he is doing by saying it. The listener understands what is said when he knows what to do with it. Understanding words means knowing how to use them. To explain how we understand words we must explain how they prepare us for action.

A description can be communicated on its own. In this simple case, the speaker understands what he says if he knows what he is describing, and the listener understands what is said as soon as he imagines what is being described. But obviously there are many other uses of speech than the communication of descriptions. Knowing how to describe what one perceives and knowing how to imagine what is described are only the most fundamental part of language understanding. It is fundamental because all other uses of language make use of descriptions.

Many words and expressions do not have a directly descriptive function. They do not name concepts or objects, but serve as binding words or as expression modifiers. They are very useful both to enrich our descriptions and to determine the ways to use them. The study of the meanings of verbal expressions is not limited to their descriptive function. It includes all the uses of language, all the ways of using descriptions, because they are all modes of understanding, therefore ways of giving meaning to our words.

A speaker acts on those who listen to him. To know what he does when he says what he says, he must know what the audience does or can do with it. A speaker must be able to put himself in the place of the listeners and understand what they understand, otherwise he does not really understand himself. Conversely, to know what to do with what is said to them, listeners must understand the speaker's intentions, why he says what he says. They must therefore be able to put themselves in the speaker's place and understand what he does, otherwise they do not really understand what is said to them. The understanding of words is one of the forms of mutual understanding, in which everyone knows others and himself, and knows that he is known by others in the same way as he knows them.


« Do you accept my description of the process of thinking ?
- How do you describe it ?
- As a discourse that the mind carries on with itself about any subject it is considering. You must take this explanation as coming from an ignoramus ; but I have a notion that, when the mind is thinking, it is simply talking to itself, asking questions and answering them, and saying Yes or No. When it reaches a decision — which may come slowly or in a sudden rush — when doubt is over and the two voices affirm the same thing, then we call that its judgment. So I should describe thinking as discourse, and judgment as a statement pronounced, not aloud to someone else, but silently to oneself.
- I agree. »

(Plato, Theaetetus, 189e-190a)

Thought is the imagination of speech.

Solitary thought is a kind of imaginary social life. The human comedy is internalized by imagination, as if all society could speak on the interior scene.

Why do we think? What is the use of talking to oneself imaginatively? Why want to influence with words one's own imagination and one's own will?

Speech gives knowledge and will a social existence, independent of the individuals who know and want. Thanks to the word, everyone can benefit from the knowledge acquired by all the others, make others benefit from the knowledge he or she has acquired, and participate in projects carried by a common will. Silent imagination is enough to develop a knowledge, but it produces only private knowledge, which can only be passed on to someone capable of putting himself in our place. Talking imagination, that is to say thought, makes it possible to appropriate and develop a collective knowledge, potentially universal.

As agents build themselves by imagination and will, and as they can influence the imaginations and wills of other agents by speech, they can each act on how others build themselves. By the word they build, train and develop themselves all together. As the word can be shared, it is a common resource for building oneself. We can all receive the same benefits of the word to train us.

We build ourselves by word and thought, but we can also say that it is the word that builds us.

Theoretical frameworks and the priority of the a priori[edit]

Theoretical frameworks are the talking equivalent of silent conceptual frameworks. A theoretical framework is determined with a system of names, intended to name concepts, and rules of inference. A theoretical framework is interpreted when the named concepts are identified with detection systems. The system of detection systems thus determined is the conceptual framework associated with the theoretical framework thus interpreted. It gives meaning to theoretical expressions, which can then be used to formulate descriptions. The silent inference rules of the conceptual framework are the interpretation of the inference rules of the theoretical framework.

Conceptual frameworks are a first degree of generality and abstraction. They are general because they can be applied to many particular realities. They are abstract because they are distinct from the concrete beings to which they are applied. The theoretical frameworks are a second degree of generality and abstraction. They are doubly general because they can be interpreted by many different conceptual frameworks, which are themselves general frameworks. They are doubly abstract because they are distinct both from concrete beings and from conceptual frameworks applied to concrete beings.

When the named concepts are identified with systems of perception, they have an empirical meaning. The truth of the statements is then determined by the truth of the descriptions, and therefore by the truth of perception. Descriptions are true as soon as the perceptions they translate are (Locke 1690).

Perception is not the only criterion of truth, because theoretical frameworks impose the truth of their principles (Leibniz 1705, Kant 1787). These are accepted by definition of their terms. Their truth is supposed to be known as soon as the meaning of words is understood. A theoretical framework thus gives abstract meanings to names of concept, independent of our perceptions and of the results of our experiments. For example, the truth of the principle of transitivity, 'if x is greater than y and y is greater than z then x is greater than z' is accepted by definition of the relation 'is greater than'. This relation can be interpreted in many empirical ways, but the principle of transitivity can never be contradicted by our perceptions. If it leads to an erroneous anticipation, it will be said that the observed relation has been misnamed, that it is not an empirical meaning that can be given to the expression 'is greater than'.

The paradox of Condorcet (1785), in political science, illustrates the priority of an a priori principle:

One can assume that electoral results give strength to the various candidates and consider measuring this force. Assume that in an election where each voter must rank three candidates A, B and C in order of preference, the three ABC, BCA and CAB orders have each been chosen by one-third of the electorate. It seems that the force of A is greater than that of B, since two-thirds of the electorate prefers A to B. Similarly B is stronger than C, and C is stronger than A. The principle of transitivity is therefore contradicted by experience. But it is not refuted, it has only been poorly applied, because such an electoral system does not make it possible to measure the strength of the candidates. It will not be said that the theoretical framework (the measure of the magnitude of the forces) is false, but only that it is not adapted to the perceived reality.

We may say of principles that they give abstract meanings to their terms, or that they abstractly determine their meaning, because knowledge of principles is part of the understanding of language. To understand each other, the speaker and the listener must know the same principles because reasoning is part of preparation for action, because one often needs to reason to know what to do with speech.

To develop a theoretical knowledge about knowledge, we must define a theoretical framework which enables us to reason about the development of knowledge. This is precisely the main ambition of this book.

Freedom of interpretation[edit]

Words and verbal expressions can be interpreted in many ways and thus receive many empirical or abstract meanings. To determine an empirical interpretation, it is enough to associate to the named concepts perceptual systems which enable us to detect them. In order to determine an abstract interpretation, it is enough to associate to the named concepts systems of principles which enable us to reason with them.

In general, an expression can have many empirical meanings, and new ones can always be invented. But when principles are true by definition they impose constraints on interpretation, limits on the empirical meanings we can give to our expressions. Empirical interpretations are not independent of abstract interpretations. Truth by definition is generally a priority. If an expression is used in a way which contradicts a principle, it will be said that the interpretation is not correct, or that it is not one of the interpretations permitted by the principles. In this way, we can be sure that our principles are always true, because an interpretation that would make them false is a priori excluded.

According to the interpretation given to it, the same statement can be at the same time an abstract truth and an empirical truth. Such an ambiguity can be very useful, because by developing an abstract knowledge, one at the same time obtains an empirical knowledge. We do not even need to change the wording. Obviously for such magic to occur, the theoretical frameworks and their interpretation must be adapted to the observed reality.

When we learn a language, we learn at the same time new expressions, new ways of perceiving, which give empirical meanings to these expressions, and new principles with which we can reason. We develop thus our empirical knowledge and our abstract knowledge at the same time. The two are intertwined in an inextricable way because in general one and the same expression combines both empirical and abstract meanings.

To know if a statement is true, one must first specify its meaning. The same speech may be sometimes true, sometimes false, it depends on its interpretation. Most of our controversies come whether from misunderstandings, because we give different meanings to the same expression, or lack of precision, because we leave in the vague the truth conditions of what we say. We do not explain the principles which decide abstract truth or we do not specify the systems of perception which decide empirical truth.

The diversity of interpretations can make communication of knowledge very difficult. The speaker must respect a principle of clarity: to provide clarification so that his speech can be interpreted correctly. The listener must respect a principle of charity: always to interpret a discourse in the way that is most favorable to it, as far as possible. It is always possible to dispel misunderstandings and to reach consensus, because we can all do the same reasonings and perceive the same world.


A silent inference can be translated into talking knowledge in the form of a conditional law: if the conditions then the consequence. The inference from the conditions to the consequence can then be put in the form of a logical reasoning: if the conditions then the consequence, now the conditions are true, hence the consequence is true. 'hence' separates the conclusion from the premises from which it logically results. In this way, sequences of silent inferences can also be put in the form of a logical reasoning: if A then B, and if B then C, now A, hence B, hence C.

A reasoning (a demonstration, an argument, a logical or mathematical proof ...) is a series of affirmations intended to prove its final one, its conclusion (Aristotle, Organon). It begins with premises, which may be truths known elsewhere, or hypotheses, or principles. Any affirmation may be chosen as premise, but the force of a reasoning, its value as proof of its conclusion, depends on the truth of its premises. A reasoning establishes that its conclusion is true provided its premises are true. If these are false, the reasoning proves nothing.

For a reasoning to be logical, each assertion which follows the premises, including the conclusion, must be naturally recognizable as a logical consequence of the assertions preceding it (Gentzen 1934, Fitch 1952). The simplest logical consequences are generally recognized intuitively. When one learns to speak, one learns to use logical operators: implication (if A then B), conjunction (A and B), negation (not A), disjunction (A or B), equivalence (A if and only if B)... They are present in all languages ​​with very varied formulations. As soon as one understands their meaning, one recognizes at the same time the truth of simple logical rules. For example, B is a logical consequence of 'A and if A then B'. We recognize the truth of this principle as soon as we understand the meaning of the implication and conjunction operators. The truth of the logical principles can be accepted by definition of the logical operators.

As an operator can be identified with the functional relation which it defines, the logical operators can be considered as concepts, which enable us to specify the logical relations between affirmations. Logical principles determine theoretical frameworks which give abstract meanings to logical concepts. These theoretical frameworks can be applied to all our reasonings in order to verify their logical correction.

Logical principles state only the simplest forms of logical reasoning. But a complex logical reasoning can always be put in the form of a succession of simple reasonings, in which the conclusion evidently results from the premises which precede it.

The logical principles always make us pass from the true to the true. When affirmations are true, their logical consequences can not be false. More precisely, whatever interpretation we give to affirmations, if these affirmations are true, according to the supposed interpretation, then the logical consequences are also true, according to the same interpretation. The relation of logical consequence does not depend on the interpretation of what we affirm, it depends only on the meaning of the logical operators.

When we prove a conclusion by logical reasoning, the premises determine sufficient conditions of truth. Whatever interpretation is chosen, if the premises are true, then the conclusion is true. Reasoning serves not only to prove, but also to explain the conditions of truth. To understand a theorem, one must know its proof, because it gives truth conditions which specify how one must interpret it.

Mathematical truth[edit]

A theoretical framework can be identified with the system of principles which determine the correct ways of reasoning with the concepts they use. Formally, a system of principles is a set of axioms and definitions. It defines a theory, which can be conceived as the set of all assertions provable from the principles, the theorems of the theory. A theory gives an abstract meaning to the names with which it forms its assertions, because its principles can be accepted as true by definition of their terms. Sometimes axioms are said to be disguised definitions, because they serve to give meaning to their terms, and thus to define them.

A theory can be used in a way that resembles perception, like a system of detectors. To know if a statement is true or false, it is enough to prove it or to prove its negation. In this way the names of concepts are associated with theoretical detectors that determine whether the concepts are true of the named individuals. Theoretical detectors detect what they must detect by finding logical proofs, based on accepted principles. If they do not find any, their detection has failed. « The eyes of the soul, whereby it sees and observes things, are none other than proofs. » (Spinoza, Ethics, Book V, prop. 23, Scolie) Just as the eyes of the body enable us to see visible beings, logical proofs based on principles enable us to know abstract beings. The perception of abstract beings consists in reasoning from the principles which define them.

To develop a purely abstract knowledge, it is sufficient to choose principles, to admit their truth by definition of their terms, and to use them to prove theorems (Russell 1903). From the point of view of abstract knowledge, any system of principles, as soon as it is not contradictory, can be considered as true by definition. But of course we do not choose our principles arbitrarily. They decide our reasonings, the theories we develop and the meanings we give to our words. They must therefore be chosen very carefully.

Abstract knowledge can be conceived in two ways. According to Platonism (the realism of Ideas) it bears on an abstract reality, immutable and independent of observable reality. According to nominalism, it is only a particular form of knowledge about knowledge, because it consists essentially in studying theories, that there is nothing else to know except principles and theorems .

Abstract theoretical knowledge is the talking equivalent of the silent imagination of fictions. For theoretical beings to exist and be known, it is enough to make a theory of them, to give oneself principles and to reason from them. Theoretical beings exist as objects of theory, simply because it is true that we speak of them. Theoretical beings are completely determined by our definitions and by the theories in which we have defined them.

Just as the imagination of fictions makes us discover our capacities of knowledge by silent inference, the invention of abstract theories makes us discover our capacities of knowledge by reasoning.

Purely abstract theories are sometimes of prodigious utility to know the concrete reality. One can be surprised at the unreasonable effectiveness of mathematics in the natural sciences (Wigner 1960).

Talking ethical knowledge resembles an abstract theory. It is stated with principles which are admitted as true by definition of an ideal.

Following chapter >>