A Philosophical Discourse on Language (Ludwig Wittgenstein)

by Marni Riddle

Language and philosophy have an intimate connection to one another; without a philosophical examination of the meanings and structure of language, we cannot easily ascertain the objective truth of the statements we make, nor can we usefully discuss abstract concepts. The philosophy of language seeks to understand the concepts expressed by language and to find a system by which it can effectively and accurately do so. This is more difficult than it appears at first; philosophers are looking for a theory of language which avoids the minute errors of meaning and usage which occur in all discussions of abstract concepts and which tend to lead those discussions into complicated dead-ends.
Since so much of philosophy is currently concerned with the linguistic representation of reality, the bond between the philosophical and the linguistic is growing stronger. Philosophers can only write syntax for the languages they want to use in expressing theory with some knowledge of linguistics; and linguists can use philosophical principles to solve problems of meaning and syntax (Moravcsik 89). This strong link can be exploited to the advantage of both sides.
In recent history philosophers have struggled with the question of precision in language and have sought to construct a system under which meanings can be discussed without danger of falling into circular or metaphysical traps. Two major approaches to this question have arisen in scientific circles of the twentieth century. Logical empiricism, also known as logical positivism, seeks to produce a language which consists of symbols combined precisely in accordance with specific rules; this would eliminate the philosophical convolutions that arise from the use of imprecise and confusingly ordered language (Sengupta 14). Ordinary language theory, on the other hand, suggested that these philosophical problems appear when language is used improperly; the language itself is perfectly acceptable and can be easily applied to the discussion of abstract and philosophical concepts without undue modifications, as long as it is used and interpreted properly (Katz 69). Each of the! ! se movements in linguistic philosophy had its strengths and weaknesses, and its supporters and detractors.

Pure metaphysical speculation which is not based on fact is, to the empiricists, neither relevant nor useful. The only truth, in this philosophy, is that which is mathematically provable or experimentally observable (Katz 18-19). This truth can be divided into two categories: analytic truths are based on inherent meanings and can be observed through the application of reason, if not experiment; synthetic truths are those facts which are obtained from the experience of reality (Quine, in Rosenberg & Travis, 63). How does this apply to linguistic philosophy? Any system of communication must, in order to be meaningful, include some way to represent the truth accurately; any empiricist will tell you that this truth is only valuable and meaningful if it can be considered absolute and provable. In order to be perfectly accurate in representing the truth, language must conform to a certain set of specifications designed to prevent it from wandering into speculation, and under which it becomes possible to ascertain absolutely the truth of any statement. These rules are called formalist semantics (Moravcsik 27). Syntax differs from semantics in that syntax guides the proper formation of elements of a language into statements, whereas semantics consists of the correct association of elements of language with elements of the real world (Moravcsik 28).
In one view, philosophy itself cannot be anything other than logically empirical because the purpose of philosophy is to elucidate and clarify truth, and this clarification consists of the examination of language to see that it conforms to the concrete facts of reality. The philosopher’s task is to analyze language and untangle the convolutions of common language into the simplicity of logical language (Sengupta 23). Sengupta quotes Ayer, one of the mainstays of the logical positivism movement, as saying that the philosopher “is not concerned with the physical properties of things. He is concerned only with the way in which we speak about them” (23). Thus philosophy must be empiricist and formalistic in order to discuss aspects of reality with accuracy and truth.

Origins and concepts
The names most commonly associated with logical positivism come from the philosophical movement known as the Vienna Circle. This group of philosophers included such famous names as Carnap, Schlick, and Godel; their main goal was to establish a context for philosophical thought which relied on observable fact and disregarded the metaphysical (Qadir 1-6) They came together in 1920’s Vienna to create a philosophy that would oppose metaphysics and provide a basis for clarity in science.
In the 1930’s, Rudolf Carnap, one of the chief logical positivists, developed what he called Logische Syntax. The aim of this work was to describe a theory of the logical syntax of language that would allow meaningful sentences to be created without reference to the meanings of specific symbols or words. This theory would be held to the same strict standards as a scientific theory (Cirera 221, 229). Essentially logical syntax formed a mathematical model of language which could be manipulated and proved just as any other mathematical or logical construct. It is a reductionist model, attempting to show that the logic of sentences is based upon the order of the word-symbols in that sentence, and do not require any reference to anything outside that sequence (i.e., reference to the semantical associations of the word symbols) in order to be meaningful (Cirera 292). In other words, a sentence arranged according to formalistic syntax will be meaningful no matter w! ! hat the word-symbols themselves represent.
G.E. Moore, although not an original member of the Vienna circle, was another philosopher whose work furthered the aims of logical positivism; he helped to formulate the movement’s view of the purpose of philosophy. In his view, as Qadir summarizes, “the business of Philosophy is clarification and elucidation of concepts and not the discovery of facts” (8). This concept is one of logical positivism’s unique identifiers, distinguishing it from definitions of philosophy which purport that the task of the philosopher is to provide new knowledge.

The metaphysical reply
Syed Ataur Rahim, a professor at the University of Karachi, composed for his doctoral thesis a reply of metaphysics to the attacks posed by logical positivism. He asserts that the logical positivists misdefine metaphysics in their attempt to disarm it; metaphysics is not meaningless or unrelated to facts, but rather is “an epistemologico-ontological inquiry” (186). Rational metaphysical inquiry is necessary for the construction of rational thoughts and for the furthering of scientific inquiry; unobservable ideas, which would be considered irrelevant by logical positivists, must be entertained before the formation of rational explanations about reality can take place (187). Therefore, a language based only on the concepts of logical positivism would be bereft of the contributions of metaphysics, and would lead only to a sterile field of scientific thought.

Logical positivism and linguistics
During its heyday, the movement contributed several theories to the study of language. Its distinction between cognitive and non-cognitive meanings separated the functions of language into informational and emotional categories; only the first category, however, was considered meaningful by the empiricists (Qadir 22). Logical problems arise when the second category is treated in the same way as the first; it must be remembered that a statement with qualities of emotion or appeal is not subject to distinctions of truth or falsity, and is therefore meaningless (Qadir 22). Carnap suggested that both of these two categories could be found in any one sentence; but only the cognitive portion of the sentence was significant; all emotive value was irrelevant (Feigl 6). Additionally, a line could be drawn between the significant and the non-significant uses of language; questions of religion and ethics were considered insignificant, since they did not refer primari! ! ly to aspects of the material world, and only language which could be dealt with utilizing methods of empirical proof was considered significant.
The ideal language sought by the empiricists has several identifiable properties. It was intended primarily to add conventions to language at points where the guidelines of semantics and syntax were loose enough to allow metaphysical speculation or, equivalently, nonsense (Katz 21). For instance, Katz cites Carnap, one of the founders of the Vienna Circle, as suggesting the inadequacy of normal grammar because it allows sentences such as “Caesar is a prime number” to be considered grammatically correct (22). This ideal is based on mathematical and logical models; it represents the structure of a natural language but supposedly eliminates the vaguenesses and possibilities for misconception to which natural languages are prone.
In essence, logical empiricists have several main postulates. Anything that is not based on rational fact is not eligible for philosophical contemplation. Philosophy’s main concern is with science and scientific language. The only way to make sure that language remains factual is to devise an artificial language that can only be used to create statements which refer to analytically or synthetically true information (Sengupta 24-25).

Possible flaws
Formalism does have its faults, however. It presumes that all meaningful concepts can be expressed in terms of synthetically or analytically provable language, which may not necessarily be true; such a view eliminates the possibility of statements which have validity although we have yet to find methods of proving them (Sengupta 40). It is difficult to translate assertions from the ideal logical language into meaningful common language, since the former assertions are devoid of the added non-cognitive meanings which we commonly use to clarify cognitive meanings; and formalistic syntax cannot accommodate certain concepts of significance that are associated with normal language, nor can it express ideas above a certain level of complexity (Sengupta 41-43), since increased complexity often incorporates philosophical concepts which have a partially metaphysical nature.
In some ways the absolute refusal of empiricism to accommodate the idea of the existence of concepts other than the concrete may have contributed to its fall from philosophical popularity. This narrowed view cut off the logical positivists from consideration of all the possibilities of language. As Duke undergraduate and philosophy aficionado Allan Stevens noted, “Imagine the scientist who demands logical proof for every idea he acquires. There are two courses of action open to him. He can either spend all of his time trying to rationally verify minor reasonable truths, or he can just disregard the ideas that it is inconvenient to prove. Either example would extremely limit his total body of knowledge” (Mentation 1:1).
Although this movement was considered by some to be more concerned with the reform of science than of language, one of logical positivism’s major foci was the reform of language to make it more precise and so to make it a better tool for the description of reality. This goal was never realized before the movement decreased in strength, although modern linguistic theory owes a great deal to empiricist concepts

Logical empiricism did not go unchallenged by the philosophical community. One of the main counter-movements was called ordinary language theory; this theory suggested that everyday language without any special, more formal semantics could be used to discuss philosophical thought; it just had to be used correctly. Errors in usage, not deficiencies in the structure of language, led to philosophical misconceptualization (Katz 69). Ludwig Wittgenstein, originally a member of the positivists’ Vienna Circle, had a major influence on the formation of this theory.
Although at first he agreed with the principles of logical empiricism, Wittgenstein came to believe that it was too scientific for the topics it attempted to address and could not accommodate those valid parts of philosophical thought which were characterized by ambiguity (Katz 70). The artificial language created by the empiricists was overly scientific and tried to assign absolute meaning to non-absolute terms. Some terms necessarily possessed a degree of vagueness and could not be bound by the strictures of formalism, or else some valuable meanings would be lost (Katz 70-71). The search for understanding would not benefit so much from an insistence on absolute and perfect precision of meaning, created within the framework of an artificial logical language, as it would from the correct usage of the already existing language and the avoidance of errors propagated from its misuse (Katz 75).
Wittgenstein and ordinary language
Although Wittgenstein originally supported the philosophy of the Vienna Circle, he later reconsidered his views. His metamorphosing views embody the conflict between logical empiricism and ordinary language because over the course of his own scholarly career he supported the positions of both sides, at some point switching his consideration from one to the other. His Tractatus Logico-Philosophicus is a standard reading in logical positivism; his later Philosophical Investigations refutes and criticizes the thought described in the Tractatus. The main thrust of the Investigations’ opposition to the ideas described in Table 1 is that the logical, fixed form of the world and the objects in it, described by the empiricists in the formulation of their theories, is in itself a metaphysical construction. Empiricism relies on the existence of a fixed form of the world because only in this fixed state can it be assured that the real! ! world is composed of unassailable facts to which language can refer (Malcolm). In his later philosophical work, Wittgenstein began to question the unassailability of this reality; Malcolm describes his new view as a realizatio n “…that the formation of concepts, of the boundaries of what is thinkable, will be influenced by what is contingent – by facts of nature, including human nature” (19). Therefore the logical suppositions of positivism were themselves metaphysical concepts, as constructs of an ideal world.
Wittgenstein considered the positivistic view, that underneath all apparently complex statements and concepts lie essentially and absolutely simple elements of reality, to be illusory (Malcolm 53). This simplicity was necessary to the aforementioned construction of a fixed and logical universe to which formalistic language could be applied. But reality has not been shown to reduce to these simple elements and thus the assumption that it does is unjustifiable (Malcolm 52).
Wittgenstein also altered his conception of the proposition as correspondent to reality. In the positivistic view a thought is a representation of a certain specific reality (Malcolm 100). Propositions are verbal expressions of specific thoughts; in order for the propositions to be logically empirical, the thoughts must likewise be reducible to simple and fixed elements, which Wittgenstein decided was unnecessary. Any method by which thoughts and propositions were connected with aspects of reality could be interpreted in various ways by various people, and thus thoughts on the same real object could differ from one another according to the path through which the object was intellectually interpreted (Malcolm 100). Therefore, propositions (correspondent to thoughts) which purportedly referred to the same aspect of reality might not be equivalent, and so the empiricist principle that a statement should have a singular correspondence with reality would not be maintain! ! ed.

POSITIONS TAKEN BY THE TRACTATUS WHICH WERE LATER REJECTED IN THE INVESTIGATIONS.There is a fixed form of the world independent of whatever is the case.
The fixed form of the world is constituted of things that are simple in an absolute sense.
The simple objects are the substratum of thought and language.
Thoughts, composed of “psychical constituents,” underlie the sentences of language.
A thought is intrinsically a picture of a particular state of affairs.
A proposition or thought cannot have a vague sense.
Whether a proposition has sense cannot depend on whether another proposition is true.
To understand the sense of a proposition it is sufficient to know the meaning of its constituent parts.
The sense of a proposition cannot be explained.
There is a general form of all propositions.
Each proposition is a picture of one and only one state of affairs.
When a sentence is combined with a method of projection the resulting proposition is necessarily unambiguous.
What one means by a sentence is specified by an inner process of logical analysis.
The pictorial nature of most of our everyday propositions is hidden.
Every sentence with sense expresses a thought which can be compared with reality.
Adapted from Nothing is Hidden, Malcolm 1986.

Ordinary language theory constituted a major shift in view away from empiricism. Instead of being considered a poor vehicle for philosophical thought, language came to be viewed as perfect. Sengupta quotes Gellner’s justifications for ordinary language: “(i) it is its own standard (there can be no other) and so there can be no ‘improvements’; (ii) ordinary language has stood the test of time; (iii) innovations are possible but extra-philosophical, and also (iv) innovations are called for in technical subjects not in ordinary language” (40).
The ordinary language philosophers believed that empiricists made the assumption that statements about reality had to be logically based on events observable in reality; it was considered more productive to examine the workings of language in order to find out more about unobservable reality. Additionally, the formalist attempt to indicate truth by underscoring the absolute meanings of terms has not been fruitful (Sengupta 40).

At the present time neither logical positivism nor ordinary language theory are considered useful for the modern study of language. They both possess the flaws described above in their respective sections, and due to their inadequacy another theory had to be devised. Approaches to this theory have been attempted by several noted linguistic scholars, including Jerrold Katz at Harvard and Peter Achinstein at Johns Hopkins.
The theory of language that Katz developed was designed to explain the theory of linguistic structure while remaining true to the factual basis of natural language. Essentially, the explanation posited is a model of communication which suggests that a speaker is following rules with a definite structure when he creates or understands novel sentences; these rules allow him to ascertain meanings compositionally, deducing the meaning of a sentence from the meanings of its parts (Katz 176-177). In this way his theory compromised between empiricism and ordinary language; it asserts that thoughts and ideas are unobservable but that this is no more a meaningless piece of metaphysics than the scientific assertion of unobservable and theoretical particles. The scientific method allows the scientifically unobservable to be considered valid; just because there is no similar method for establishing the empiricism of ideas doesn’t mean that they are any more metaphysical (Katz 18! ! 1).
Peter Achinstein examined a contemporary positivist approach which, although it did not attempt a reconciliation with ordinary language, still modified the original version to adapt it to modern thought. In his view, the anti-positivist was more concerned with the main features of concepts and ignored certain parts that might not apply in all cases. The positivist, on the other hand, wants to give all concepts an absolute and complete set of attributes that are empirically provable and uniform (Achinstein 279). The view which lies between these suggests the necessity of a general set of conditions to define concepts; however, instead of eliminating those which do not fit, this structure can be used to understand more about the concepts which are slightly outside the given structure (Achinstein 289-290).
As we draw closer to a time in which concepts such as love and the soul can be expressed in biological terms, connected intimately with brain tissue and the workings of the body, the scientific language of the logical positivists appears more and more applicable to previously unscientific terms. It is not necessary, however, to consider immaterial concepts meaningless; it is possible that they simply have not yet crossed the boundary into the scientific world. For instance, as we come closer to describing feelings in terms of neurotransmitter levels and neuronal firings, we also come closer to giving them meaning on the empirical level. Who can say that this concept which is presently immaterial (and therefore empirically insignificant) will never come under the auspices of positivistic science? Thus it appears that the empiricists, in insisting that the metaphysical was meaningless, may have spoken too soon; over time the metaphysical may metamorphose into the physical! ! as we learn more and more. And in its place may arise deeper layers of the metaphysical; will we ever eliminate the unprovable?

Works Cited
Achinstein, Peter. “Approaches to the Philosophy of Science.” The Legacy of Logical Positivism: Studies in the Philosophy of Science. Eds. Peter Achinstein and Stephen F. Barker. Baltimore: Johns Hopkins P, 1969.
Cirera, Ramon. Carnap and the Vienna Circle: Empiricism and Logical Syntax. Trans Dick Edelstein. Amsterdam: Rodopi BV, 1994.
Feigl, Herbert. “The Origin and Spirit of Logical Positivism.” The Legacy of Logical Positivism: Studies in the Philosophy of Science. Eds. Peter Achinstein and Stephen F. Barker. Baltimore: Johns Hopkins P, 1969.
Katz, Jerrold J. The Philosophy of Language. New York: Harper & Row, 1966.
Malcolm, Norman. Nothing is Hidden: Wittgenstein’s Criticism of his Early Thought. Oxford: Basil Blackwell, 1986.
Moravcsik, J.M.E. Understanding Language: A Study of Theories in Linguistics and in Philosophy. The Hague: Mouton, 1975.
Qadir, C.A. Logical Positivism. Lahore: Ripon P, 1965.
Quine, Willard van Orman. “Two Dogmas of Empiricism.” Readings in the Philosophy of Language. Eds. Jay F. Rosenberg & Charles Travis. Englewood Cliffs: Prentice Hall, 1971.
Rahim, Syed Ataur. Logical Positivism and Metaphysics. Hussainabad/Karachi: Rahim P, 1990.
Sengupta, Kalyan Kumar. Language and Philosophy. Bombay: Allied P, 1969.
Stevens, Allan. On Belief. [Online] Available http://www.duke.edu/~kellogg/Classes/Mentation/Issue1/stevens.htm

By Marni Riddle

December 16, 1997.
Exploring the Mind
Duke University
Durham, North Carolina
Fall 1997