The Elimination of Metaphysics Through Logical Analysis of Language
originally: Überwindung der Metaphysik durch logische Analyse der Sprache
Rudolph Carnap, 1931
Translated by D.S. Thorne
I omit chunks of this to avoid copyright violation...
Translated by D.S. Thorne
I omit chunks of this to avoid copyright violation...
- Introduction
- The Meaning of a Word
- Metaphysical Words without Meaning
- The Meaning of a Proposition
- Illusory Propositions of Metaphysics
- Meaninglessness of all Metaphysics
- Metaphysics as the Expression of One’s Feeling for Life
1. Introduction
From the Greek Skeptics to the Empiricists of the 19th century there have been many opponents of metaphysics. The ways in which these scruples have been brought forth have been very different in nature. Some pronounced the teachings of metaphysics to be false, since they contradict, as they said, knowledge based in experience. Others consider them only uncertain, since their questions transgress the boundaries of human knowledge. Many anti-metaphysicians declared the occupation with metaphysical questions to be fruitless: regardless whether or not they can be answered, it is in any event useless to pay attention to them; one should instead devote oneself entirely to the practical matters that come man’s way every day!
Through the development of modern logic it has become possible to give a new and clearer answer to the question of the validity and justification of metaphysics. The investigations of “applied logic” or “the theory of knowledge”—whose task is to clarify the cognitive content of scientific propositions and thereby the meaning of terms, or concepts, appearing in those propositions by means of logical analysis—lead to a positive and a negative result. The positive result is worked out in the area of empirical science; the individual terms of the various branches of science are clarified; their formally logical and cognitively theoretical coherence is shown. In the area of metaphysics (including all philosophies of value and sciences of norms), logical analysis leads to the negative result that the supposed propositions of this discipline are entirely meaningless. In this way metaphysics is radically eliminated, which was not yet possible from earlier anti-metaphysical standpoints. There are indeed among some earlier considerations of this matter similar thoughts, e.g. in those of the nominalist variety; but the critical completion is possible only today, now that logic has been whetted to an instrument of adequate sharpness though the development it has undergone in recent decades.
When we say, that the so-called propositions of metaphysics are meaningless, we mean this in the strictest sense. One sometimes characterizes a proposition or a question as meaningless in a looser sense, when establishing that it is useless (e.g. the question: How great is the average body weight of those people in Vienna whose telephone numbers end with a ‘3’?); or also a proposition which is manifestly false (e.g. “in 1910 the population of Vienna was six”), or a proposition of the sort that is not only empirically but also logically false, or contradictory (e.g. “of two people A and B, each is one year older as the other”). Propositions of this nature are actually meaningful, even if they are fruitless or false; for only meaningful propositions can be divided into the (theoretically) the fruitful and fruitless, true or false at all. On the other hand, a series of words is meaningless in the strictest sense, when it forms no proposition at all within a given, definite language. It happens, that such a series of words looks at first glance like a proposition; in this case we call it an illusory proposition (Scheinsatz). Now, our thesis is that the supposed propositions of metaphysics are revealed as illusory propositions through logical analysis.
A language is made up of vocabulary and syntax, i.e. a stock of words that have a meaning and rules for the formation of propositions; these rules govern how propositions can be formed of words of different types. Accordingly there are two kinds of illusory propositions: either it has a word which one mistakenly takes to have a meaning, or the words have meanings but are put together in a counter-syntactical way, such that they do not convey meaning. We shall see in examples that illusory propositions of both varieties occur in metaphysics. After that we will have to consider our reasons for believing that all of metaphysics is made up of such illusory propositions.
2. The Meaning of a Word
If a word (within a certain language) has a meaning, it is often said that it describes a “concept” (Begriff); if it only appears that a word has a meaning, while in reality it has none, we speak of an “illusory concept” (Scheinbegriff). How can we account for the origin of such a thing? Is not every word introduced into language in order to express something definite, so that it as a definite meaning from its first use on? How can there be meaningless words in traditional language? Originally every word (apart from rare exceptions, an example of which will be given later) has indeed a meaning. In the course of historical development a word often changes its meaning. An now it sometimes also occurs that a word loses its old meaning without a receiving a new one. In this way the illusory concept comes to be.
In what does the meaning of a word consist? What must be established in relation to a word for it to have a meaning? (Whether this establishment is expressly mentioned, as with some words and symbols of modern science, or tacitly agreed to, as it seems to be the case with most words of traditional language, is irrelevant to our purposes.) First the syntax of the word must be established, i.e. the manner of its appearance in the simplest propositional form in which it can occur. We call this propositional form the elementary proposition (Elementarsatz). For example, the elementary propostion for the word “rock” is “x is a rock”. In propositions of this form some designation or other from the category of things, e.g. “this diamond”, “this apple”, stands in the place of “x”. Secondly, for the elementary proposition S of the relevant word, an answer must be given to the following question, which we can formulate in various ways:
From which kinds of propositions is S derivable, and what propositions are derivable from S?
(1) is the correct formulation; the formulation (2) conforms to the manner of logic, (3) conforms to the theory of knowledge, (4) conforms to philosophy (phenomenology). Wittgenstein has noted that that which philosophers mean by (4), is grasped by (2): the meaning of a proposition lies in its criterion of truth. [(1) is the “meta-logical” formulation; an explicit presentation of meta-logic as theory of syntax and of meaning, i.e. of derivative signification, should be given another time.]
With many words, and indeed with the overwhelming majority of scientific words, it is possible to give its meaning by reduction (Zurückführung —“leading back”) to other words (“constitution”, definition). E.g.: “‘Arthropods’ are animals with segmented bodies, segmented limbs and an exoskeleton of chitin.” In this way the above mentioned question answered for the elementary propositional form of the word “arthropod”, namely for the propositional form “the thing x is an arthropod”; it is certain that a proposition of this form should be derivable from premisses of the form “x is an animal”, “x has a segmented body”, “x has segmented limbs”, “x has an exoskeleton of chitin”, and that, vice-versa, each of these propositions are derivable from a proposition. Through these determinations of the derivability (or to put it another way: of the truth criterion, the method of verification, the meaning) of the elementary proposition for “arthropod”, the meaning of the word “arthropod” is established. In this way every word in a language is reduced (zurückgeführt) to other words and at last to words appearing in the so-called “observational propositions” or “protocol sentences”.
Since the meaning of a word is determined through its criterion (to put it another way: through the derivative signification of its elementary protocol, through its truth-conditions, through the method of its verification), it is not possible to add anything else beyond the establishment of its criterion, that one would “mean” by that word. One may give nothing less than the criterion if the word is to retain a clear meaning; but neither can one give more than the criterion, because through this criterion everything else is determined. The meaning is implicitly contained in the criterion; all that remains is to bring it to light explicitly.
Let us assume, for example, that someone forms the new word “babig” and claims that there are things that are “babig” and things that are not. To find out the meaning of this word, we will ask him for its criterion: How can we concretely establish whether a thing is babig or not? First we want to assume that he, of whom we are asking this, owes an answer, and he say that babigness does not have an empirical feature. In this case we will not consider the use of the word permissible. If the one using this word despite this says that there are babig and non-babig things, but it remains to the finite and pathetic understanding of man an eternal secret, which things are babig and which are not, than we must regard this as idle talk. But perhaps he will assure us that he actually does mean something with the word “babig”. From this we only learn the psychological fact that he associates some images and feelings or other with the word. But this does not give the word a meaning. If no criterion for the word is established, the propositions in which the word appears convey nothing—they are merely illusory propositions.
Secondly, if we grant the case that the criterion for a new word, say “bebig”, is established, and moreover that the proposition: “this thing is bebig” is always and only true when the thing has four corners. (It is not relevant for our thinking whether this criterion has been given us explicitly or whether we determine it by observing in which contexts the word is used in affirmation and in which it is used in negation.) We will say here: the word “bebig” has the same meaning as the word “quadrangular”. And we will regard it as impermissible, when those useing the word ay to us that they “mean” something different by this word than “quadrangular”; in fact, every quadrangular thing is beig and vice=versa, but that is not because quadrangularity is the visible expression of bebigness, which is actually an occult property, itself not perceptible. We will respond, that after, or even before, the criterion has been established, “bebig” means “quadrangular”; and one longer has the freedom to use the word to “mean” something different, be it this or that.
The result of our deliberations can be succinctly summarized. Let “a” be any word and “S(a)” the elementary proposition in which it occurs. The sufficient and necessary condition that “a” has a meaning can be given in the following formulations, which basically express the same thing:
3. Metaphysical Words Without Meaning
It will now be shown that many words in metaphysics do not fulfill the conditions just given, and so that they are without meaning.
Let us take as example the metaphysical terminus “principle” (indeed as principle of being not as principle of knowledge or fundamental proposition [Grundsatz]). Various metaphysicians give an answer to the question concerning what the highest “principle of the world” (or “of things”, “or being” [des Seins], “of existents” [des Seienden]) is, e.g.: water, number, form, motion, life, spirit, idea, the subconscious, the deed, the good and many ore of this sort. In order to find the meaning that the word “principle” has in this metaphysical question, we must ask the metaphysician, under what conditions is a proposition of the form “x is the principle of y” true and under which it is false. In other words: we ask after the features or the definition of the word “principle”. The metaphysician answers roughly like this: “x is the principle of y” means “y proceeds from x”, “the being of y depends on the being of x”, “y exists through x”, and so forth. But these words are ambiguous and indefinite. Often they have a clear meaning, e.g. we say of a thing of an event y, it “proceeds” from x, if we observe that things or events of type y often or always follow upon things or events of type x (causal relation in the sense of a regular sequence). But the metaphysician tells us that he does not mean this empirically determinable relation, because indeed otherwise his metaphysical theses would become simple propositions of experience of the same kind as those of physics. The word “proceed” should not in this case have the meaning of a relation of temporal series or conditions that it usually has. But no criterion is given for any other meaning. Consequently the alleged “metaphysical” meaning, that the word is supposed to have here as opposed to its empirical meaning does not exist at all. When we think about the original meaning of the word “principium” (and of the corresponding Greek word “ἀρχή”), that we meet with the same process of development. The original meaning “beginning” (“Anfang”) is explicitly taken from the word; it is no longer to mean the temporally first, but rather the first in another, specifically metaphysical regard. But the criteria for this “metaphysical regard” are not given. Thus in both cases the earlier meaning taken from the word without giving it a new one; the word remains behind as an empty husk. From an earlier, meaningful period various images still adhere to it; they link with the images and feelings by way of the coherence with which the word is now used. But the word does not thereby have a meaning. And it further remains meaningless, as long as a way of verification is not given.
Another example is the word “God”. Apart from the variations in use within each of these sections, with this word we must distinguish linguistic usage in three separate contexts or historical periods that temporally flow into one another. In the mythological use of language the word has a clear meaning. Sometimes corporeal beings are signified with this word (or its parallels in other languages), throned upon Olympus, in heaven or in the underworld, endowed with power, wisdom, goodness and fortune in more or less perfect measure. Sometimes the word also signifies spiritual-mental beings, that although they have not humanlike body, are yet somehow seen in the things and events of the visible world and therefore are empirically verifiable. By contrast, in the metaphysical use of language “God” signifies something above the empirical. The meaning of a corporeal or a spiritual being contained in a corporeal thing is explicitly removed from the word. And because it is given no new meaning it becomes meaningless. However it often appears that one also gives the word “God” a meaning in metaphysics. But the definitions that are given prove to be illusory definitions (Scheindefinitionen) on closer observation; either they reduce (“führen…zurück”) to logically impermissible combinations of words (to which we shall return later) or they reduce to other metaphysical words (e.g. “original ground”, “the Absolute”, “the Unconditioned” ,”the Independent”, “the Self-Existent”, etc.), but in no case do they lead to the truth conditions of their elementary propositions. With this word not even the first requirement of logic is fulfilled, namely the demand for a specification of syntax, i.e. of the form of its occurrence in the elementary proposition. The elementary proposition here would have to have the form “x is a God”; but the metaphysician either refuses this form entirely, without giving an alternative, or if he accepts it he does not give the syntactical category of the variable x. (Categories are e.g.: bodies, properties of bodies, relationships between bodies, numbers, etc.)
The theological use of language regarding the word "God" is situated between the mythological and the metaphysical use of language. There is not separate meaning here, but rather one that fluctuates between the other to manners of usage. Many theologians have a clearly empirical (that is, in our manner of signification, “mythological”) concept of God. There are not illusory propositions in this case; but the drawback for the theologian is that in this usage the propositions of theology are empirical propositions and so are subordinated to the judgment of empirical science. With other theologians the metaphysical use of language is meant. Bu still others the use of language is unclear; they sometimes follow this use of language and sometime the other, or they make their way with expressions that are not concrete but shimmer in both directions.
Just as with the examples we have seen, “principle” and “God”, most other specifically metaphysical termini also without meaning., e.g. “Idea”, “the Absolute”, “the Unconditioned”, the “Infinite”, “the Being of the Existent”, “the Not-Existent”, “Ding an sich”, “Absolute Spirit”, “Objective Spirit”, “Being”, “Being-in-Itself”, “Being-in-and-of-Itself”, “Emanation”, “Manifestation”, “Ausgliederung[?]”, “the Ego”, “the Non-Ego”, and so forth. These expressions do not behave any differently than the word “babig” in the example contrived earlier. The metaphysician tells us that the empirical conditions of truth cannot be given, but when he then says that such a word has a meaning nevertheless, we then know that only associated images and feelings are meant by which, however, the word is not given a meaning. The metaphysical, alleged propositions that such words bear have no meaning and say nothing; they are merely illusory propositions. We shall consider later how we might account for their emergence in history.
[Section 4 in summary: Martin Heidegger, the gloves are off...]
From the Greek Skeptics to the Empiricists of the 19th century there have been many opponents of metaphysics. The ways in which these scruples have been brought forth have been very different in nature. Some pronounced the teachings of metaphysics to be false, since they contradict, as they said, knowledge based in experience. Others consider them only uncertain, since their questions transgress the boundaries of human knowledge. Many anti-metaphysicians declared the occupation with metaphysical questions to be fruitless: regardless whether or not they can be answered, it is in any event useless to pay attention to them; one should instead devote oneself entirely to the practical matters that come man’s way every day!
Through the development of modern logic it has become possible to give a new and clearer answer to the question of the validity and justification of metaphysics. The investigations of “applied logic” or “the theory of knowledge”—whose task is to clarify the cognitive content of scientific propositions and thereby the meaning of terms, or concepts, appearing in those propositions by means of logical analysis—lead to a positive and a negative result. The positive result is worked out in the area of empirical science; the individual terms of the various branches of science are clarified; their formally logical and cognitively theoretical coherence is shown. In the area of metaphysics (including all philosophies of value and sciences of norms), logical analysis leads to the negative result that the supposed propositions of this discipline are entirely meaningless. In this way metaphysics is radically eliminated, which was not yet possible from earlier anti-metaphysical standpoints. There are indeed among some earlier considerations of this matter similar thoughts, e.g. in those of the nominalist variety; but the critical completion is possible only today, now that logic has been whetted to an instrument of adequate sharpness though the development it has undergone in recent decades.
When we say, that the so-called propositions of metaphysics are meaningless, we mean this in the strictest sense. One sometimes characterizes a proposition or a question as meaningless in a looser sense, when establishing that it is useless (e.g. the question: How great is the average body weight of those people in Vienna whose telephone numbers end with a ‘3’?); or also a proposition which is manifestly false (e.g. “in 1910 the population of Vienna was six”), or a proposition of the sort that is not only empirically but also logically false, or contradictory (e.g. “of two people A and B, each is one year older as the other”). Propositions of this nature are actually meaningful, even if they are fruitless or false; for only meaningful propositions can be divided into the (theoretically) the fruitful and fruitless, true or false at all. On the other hand, a series of words is meaningless in the strictest sense, when it forms no proposition at all within a given, definite language. It happens, that such a series of words looks at first glance like a proposition; in this case we call it an illusory proposition (Scheinsatz). Now, our thesis is that the supposed propositions of metaphysics are revealed as illusory propositions through logical analysis.
A language is made up of vocabulary and syntax, i.e. a stock of words that have a meaning and rules for the formation of propositions; these rules govern how propositions can be formed of words of different types. Accordingly there are two kinds of illusory propositions: either it has a word which one mistakenly takes to have a meaning, or the words have meanings but are put together in a counter-syntactical way, such that they do not convey meaning. We shall see in examples that illusory propositions of both varieties occur in metaphysics. After that we will have to consider our reasons for believing that all of metaphysics is made up of such illusory propositions.
2. The Meaning of a Word
If a word (within a certain language) has a meaning, it is often said that it describes a “concept” (Begriff); if it only appears that a word has a meaning, while in reality it has none, we speak of an “illusory concept” (Scheinbegriff). How can we account for the origin of such a thing? Is not every word introduced into language in order to express something definite, so that it as a definite meaning from its first use on? How can there be meaningless words in traditional language? Originally every word (apart from rare exceptions, an example of which will be given later) has indeed a meaning. In the course of historical development a word often changes its meaning. An now it sometimes also occurs that a word loses its old meaning without a receiving a new one. In this way the illusory concept comes to be.
In what does the meaning of a word consist? What must be established in relation to a word for it to have a meaning? (Whether this establishment is expressly mentioned, as with some words and symbols of modern science, or tacitly agreed to, as it seems to be the case with most words of traditional language, is irrelevant to our purposes.) First the syntax of the word must be established, i.e. the manner of its appearance in the simplest propositional form in which it can occur. We call this propositional form the elementary proposition (Elementarsatz). For example, the elementary propostion for the word “rock” is “x is a rock”. In propositions of this form some designation or other from the category of things, e.g. “this diamond”, “this apple”, stands in the place of “x”. Secondly, for the elementary proposition S of the relevant word, an answer must be given to the following question, which we can formulate in various ways:
From which kinds of propositions is S derivable, and what propositions are derivable from S?
- Under which conditions is S to be true, and under which is it to be false?
- How is S to be verified?
- What meaning does S have?
(1) is the correct formulation; the formulation (2) conforms to the manner of logic, (3) conforms to the theory of knowledge, (4) conforms to philosophy (phenomenology). Wittgenstein has noted that that which philosophers mean by (4), is grasped by (2): the meaning of a proposition lies in its criterion of truth. [(1) is the “meta-logical” formulation; an explicit presentation of meta-logic as theory of syntax and of meaning, i.e. of derivative signification, should be given another time.]
With many words, and indeed with the overwhelming majority of scientific words, it is possible to give its meaning by reduction (Zurückführung —“leading back”) to other words (“constitution”, definition). E.g.: “‘Arthropods’ are animals with segmented bodies, segmented limbs and an exoskeleton of chitin.” In this way the above mentioned question answered for the elementary propositional form of the word “arthropod”, namely for the propositional form “the thing x is an arthropod”; it is certain that a proposition of this form should be derivable from premisses of the form “x is an animal”, “x has a segmented body”, “x has segmented limbs”, “x has an exoskeleton of chitin”, and that, vice-versa, each of these propositions are derivable from a proposition. Through these determinations of the derivability (or to put it another way: of the truth criterion, the method of verification, the meaning) of the elementary proposition for “arthropod”, the meaning of the word “arthropod” is established. In this way every word in a language is reduced (zurückgeführt) to other words and at last to words appearing in the so-called “observational propositions” or “protocol sentences”.
Since the meaning of a word is determined through its criterion (to put it another way: through the derivative signification of its elementary protocol, through its truth-conditions, through the method of its verification), it is not possible to add anything else beyond the establishment of its criterion, that one would “mean” by that word. One may give nothing less than the criterion if the word is to retain a clear meaning; but neither can one give more than the criterion, because through this criterion everything else is determined. The meaning is implicitly contained in the criterion; all that remains is to bring it to light explicitly.
Let us assume, for example, that someone forms the new word “babig” and claims that there are things that are “babig” and things that are not. To find out the meaning of this word, we will ask him for its criterion: How can we concretely establish whether a thing is babig or not? First we want to assume that he, of whom we are asking this, owes an answer, and he say that babigness does not have an empirical feature. In this case we will not consider the use of the word permissible. If the one using this word despite this says that there are babig and non-babig things, but it remains to the finite and pathetic understanding of man an eternal secret, which things are babig and which are not, than we must regard this as idle talk. But perhaps he will assure us that he actually does mean something with the word “babig”. From this we only learn the psychological fact that he associates some images and feelings or other with the word. But this does not give the word a meaning. If no criterion for the word is established, the propositions in which the word appears convey nothing—they are merely illusory propositions.
Secondly, if we grant the case that the criterion for a new word, say “bebig”, is established, and moreover that the proposition: “this thing is bebig” is always and only true when the thing has four corners. (It is not relevant for our thinking whether this criterion has been given us explicitly or whether we determine it by observing in which contexts the word is used in affirmation and in which it is used in negation.) We will say here: the word “bebig” has the same meaning as the word “quadrangular”. And we will regard it as impermissible, when those useing the word ay to us that they “mean” something different by this word than “quadrangular”; in fact, every quadrangular thing is beig and vice=versa, but that is not because quadrangularity is the visible expression of bebigness, which is actually an occult property, itself not perceptible. We will respond, that after, or even before, the criterion has been established, “bebig” means “quadrangular”; and one longer has the freedom to use the word to “mean” something different, be it this or that.
The result of our deliberations can be succinctly summarized. Let “a” be any word and “S(a)” the elementary proposition in which it occurs. The sufficient and necessary condition that “a” has a meaning can be given in the following formulations, which basically express the same thing:
- The empirical features for “a” are known.
- It is fixed from which protocol sentences “S(a)” can be derived.
- The truth conditions for “S(a)” are fixed.
- The manner of verification of “S(a)” is known.
3. Metaphysical Words Without Meaning
It will now be shown that many words in metaphysics do not fulfill the conditions just given, and so that they are without meaning.
Let us take as example the metaphysical terminus “principle” (indeed as principle of being not as principle of knowledge or fundamental proposition [Grundsatz]). Various metaphysicians give an answer to the question concerning what the highest “principle of the world” (or “of things”, “or being” [des Seins], “of existents” [des Seienden]) is, e.g.: water, number, form, motion, life, spirit, idea, the subconscious, the deed, the good and many ore of this sort. In order to find the meaning that the word “principle” has in this metaphysical question, we must ask the metaphysician, under what conditions is a proposition of the form “x is the principle of y” true and under which it is false. In other words: we ask after the features or the definition of the word “principle”. The metaphysician answers roughly like this: “x is the principle of y” means “y proceeds from x”, “the being of y depends on the being of x”, “y exists through x”, and so forth. But these words are ambiguous and indefinite. Often they have a clear meaning, e.g. we say of a thing of an event y, it “proceeds” from x, if we observe that things or events of type y often or always follow upon things or events of type x (causal relation in the sense of a regular sequence). But the metaphysician tells us that he does not mean this empirically determinable relation, because indeed otherwise his metaphysical theses would become simple propositions of experience of the same kind as those of physics. The word “proceed” should not in this case have the meaning of a relation of temporal series or conditions that it usually has. But no criterion is given for any other meaning. Consequently the alleged “metaphysical” meaning, that the word is supposed to have here as opposed to its empirical meaning does not exist at all. When we think about the original meaning of the word “principium” (and of the corresponding Greek word “ἀρχή”), that we meet with the same process of development. The original meaning “beginning” (“Anfang”) is explicitly taken from the word; it is no longer to mean the temporally first, but rather the first in another, specifically metaphysical regard. But the criteria for this “metaphysical regard” are not given. Thus in both cases the earlier meaning taken from the word without giving it a new one; the word remains behind as an empty husk. From an earlier, meaningful period various images still adhere to it; they link with the images and feelings by way of the coherence with which the word is now used. But the word does not thereby have a meaning. And it further remains meaningless, as long as a way of verification is not given.
Another example is the word “God”. Apart from the variations in use within each of these sections, with this word we must distinguish linguistic usage in three separate contexts or historical periods that temporally flow into one another. In the mythological use of language the word has a clear meaning. Sometimes corporeal beings are signified with this word (or its parallels in other languages), throned upon Olympus, in heaven or in the underworld, endowed with power, wisdom, goodness and fortune in more or less perfect measure. Sometimes the word also signifies spiritual-mental beings, that although they have not humanlike body, are yet somehow seen in the things and events of the visible world and therefore are empirically verifiable. By contrast, in the metaphysical use of language “God” signifies something above the empirical. The meaning of a corporeal or a spiritual being contained in a corporeal thing is explicitly removed from the word. And because it is given no new meaning it becomes meaningless. However it often appears that one also gives the word “God” a meaning in metaphysics. But the definitions that are given prove to be illusory definitions (Scheindefinitionen) on closer observation; either they reduce (“führen…zurück”) to logically impermissible combinations of words (to which we shall return later) or they reduce to other metaphysical words (e.g. “original ground”, “the Absolute”, “the Unconditioned” ,”the Independent”, “the Self-Existent”, etc.), but in no case do they lead to the truth conditions of their elementary propositions. With this word not even the first requirement of logic is fulfilled, namely the demand for a specification of syntax, i.e. of the form of its occurrence in the elementary proposition. The elementary proposition here would have to have the form “x is a God”; but the metaphysician either refuses this form entirely, without giving an alternative, or if he accepts it he does not give the syntactical category of the variable x. (Categories are e.g.: bodies, properties of bodies, relationships between bodies, numbers, etc.)
The theological use of language regarding the word "God" is situated between the mythological and the metaphysical use of language. There is not separate meaning here, but rather one that fluctuates between the other to manners of usage. Many theologians have a clearly empirical (that is, in our manner of signification, “mythological”) concept of God. There are not illusory propositions in this case; but the drawback for the theologian is that in this usage the propositions of theology are empirical propositions and so are subordinated to the judgment of empirical science. With other theologians the metaphysical use of language is meant. Bu still others the use of language is unclear; they sometimes follow this use of language and sometime the other, or they make their way with expressions that are not concrete but shimmer in both directions.
Just as with the examples we have seen, “principle” and “God”, most other specifically metaphysical termini also without meaning., e.g. “Idea”, “the Absolute”, “the Unconditioned”, the “Infinite”, “the Being of the Existent”, “the Not-Existent”, “Ding an sich”, “Absolute Spirit”, “Objective Spirit”, “Being”, “Being-in-Itself”, “Being-in-and-of-Itself”, “Emanation”, “Manifestation”, “Ausgliederung[?]”, “the Ego”, “the Non-Ego”, and so forth. These expressions do not behave any differently than the word “babig” in the example contrived earlier. The metaphysician tells us that the empirical conditions of truth cannot be given, but when he then says that such a word has a meaning nevertheless, we then know that only associated images and feelings are meant by which, however, the word is not given a meaning. The metaphysical, alleged propositions that such words bear have no meaning and say nothing; they are merely illusory propositions. We shall consider later how we might account for their emergence in history.
[Section 4 in summary: Martin Heidegger, the gloves are off...]