Back to Home Page: http://www.frasouzu.com/

BEING AND UNIVERSALS IN THE PHILOSOPHY OF QUINE AND ITS RELEVANCE TO AFRICAN PHILOSOPHY

by

 AKPAN ETOROBONG GODWIN

TABLE OF CONTENTS

INTRODUCTION                                                                                     1

CHAPTER ONE:         THE  BACKGROUND OF QUINEAN ONTOLOGY        

1.                 Intellectual Emergence of  Quine                                                3

2.                 Quine  and  Dewey                                                          3

3.                 Carnap’s  Ontology                                                                   5

CHAPTER TWO:        THE PROBLEM OF ONTOLOGICAL CONFUSION       

1.                 Wrong  Characterization of Ontological  Statements         9

2.                 Synthetic-Analytic Cleavage                                             9

3.                 Reductionism                                                                            14

4.                 The Basis of Ontological  Confusion                                17

CHAPTER THREE: QUINEAN SOLUTION TO THE PROBLEM

   OF MEANING AND REFERENCE                          

1.                 The Theory of Description                                                        20     

2.                 Translation  and Meaning                                                 22

3.                 Indeterminacy of Reference and Ontological Relativity.     29

CHAPTER FOUR:       BEING AND UNIVERSALS IN QUINEAN

ONTOLOGY                                                       

1.                 Ontological  Commitment                                                          35

2.                 Problems of Universals                                                    39

3.                 Evaluation                                                                       42

CHAPTER FIVE: QUINEAN BEING AND UNIVERSALS AND

IT RELEVANCE TO AFRICAN PHILOSOPHY.          

1.                 Quine and African Philosophy                                         46

2.                 Conclusion.                                                                    50

Works cited

INTRODUCTION

Schooled or unschooled, man is vested with the responsibility of making intelligible every data of everyday’s experience.  This responsibility is manifested in the natural desire to know, which has pushed men to ask many questions, about the nature of reality.  These questions are at once philosophical and the beginning of the philosophical enterprise.  Prominent among the questions philosophers ask, is that of being.  The question of being has found diverse formulations in the different periods of philosophy.  Associated with the question of the nature of being is that of the ontological status of universals.

Quine is of the opinion that ontological   questions suffer confusion within Empiricism because of confusion of meaning with  naming.  Thus, he feels that the statement of ontological  starting point of ontological  investigation should be the  clearing of this confusion.  His clearing  of the  confusion results   in indeterminacy of  meaning,  of  reference, and relativity  of ontological  reference.  It is   this  relativity  of ontological  reference that conditions his  ontological  commitment, and his  treatment  of the problem  of being and universals.

The  focus of this paper is  to show what Quine says there is and  the consequence of such  for the problem of  universal and the   relevance  of that ontological  system to African philosophy.  In doing  this,  the paper takes into consideration Quine’s  philosophical  background, the problem of ontological   confusion, the  problems of meaning  and reference their solutions and  Quine’s  notion  of being  and universals and  its relevance to African philosophy.

CHAPTER ONE

THE BACKGROUND  OF QUINEAN ONTOLOGY

1.                 The Intellectual Emergence of Quine

Every philosophy has within it the  imprints of the age,  culture,  environment and persons responsible for its existence.  This has conspicuously manifested itself in the philosophy  of Williard V. O. Quine.   This name  refers to an American  philosopher and logician, who was born in 1908 and happened to live and   flourish in the twentieth (20th) century.

Quine  began  studies under the tutelage of Alfred North  Whitehead, in Harvard University.  But  following  the attractions of Logical positivism,  he got converted to it in the 1930s.  Consequently, he left Harvard and became  a student  of  Carnap,  an important figure in the movement,  at about the  same time.

His inclination to pragmatism  drew him closer to the naturalism of  Dewey.  These and many other influences join forces together to mould  the Quinean philosophical orientation.

2.                 Quine and Dewey

In his ‘Ontological Relativity and other Essays’ (1969) Quine professes his bond with Dewey over naturalism.  In this respect,  his views concerning the study of knowledge, mind and  meaning are Deweyian.  This he states as thus:

Philosophically I am bound  to Dewey by the  naturalism that dominated his last three decades.  With Dewey I hold  that knowledge, mind,  and meaning are part of the same world that they  have to  do with, and that they are to be studied in the same empirical  spirit that animates natural  science.  (Quine, 1969:26).

The consequences of this belief are exciting to note.  Quine like Dewey reduces meaning to being the property of behaviour, language to being its mode of behaviour and meaning to  being  understood only as what is expressed in behaviour or all dispositions to behaviour known or unknown.  Quine also rejects with Dewey, the view of uncritical semantics which is the myth of a museum  (The myth holds that meanings are the entities meant while languages are labels).

The gravest consequence of this affinity with Deweyian naturalism  is  his denial, with Dewey of the existence of matter of  fact in ontological  issues.  It must be  noted that Quine is not  here saying that there is no reality.  It is rather that what is known is of  conceptualization, which  is quite torrential in comparison with its meager input.  Thus,  it becomes absurd to seek  a reference,  which is founded  on a one-to-one correspondence between  the scheme and its input when we know that the output is far greater than the  input.    He,  therefore,  directs attention  to background language as a ground for reference.  The ideal which  guides the creation  of such language is convenience.

The  first position stands in opposition  to Carnap’s view (Quine’s teacher), which holds a contrary view, as proposed by radical   reductionism .  But the last sentence agrees with Carnap’s views on  conceptual   schemes.

3.                 Carnap’s Ontology

Although Quine  acknowledges that no one has influenced this philosophical  thoughts more than Carnap (Quine, cited in Feigl, 1972:597), it is also revealing  to note that no  one has suffered the consequence of Quine’s  critical  philosophical  preoccupation more than Carnap. 

Carnap held  unto  the views of  reductionism  and allowed it to influence his general  philosophical concerns.  Using it as the  basis for his ontology,  he writes as follows:

If someone  wishes to speak  in his language about a new kind of entities he has to introduce a system of new ways of speaking,   subject to new rules:  we shall call this  procedure the construction  of a linguistic  framework for the new entities in question.  And now we must distinguish two kind of  question of existence: First,  questions of the existence of certain  entities of the new kinds,  within  the framework,’ we call them  internal  questions and second  questions concerning the ‘existence or reality of the system of entities as a whole, called external questions, Carnap, (cited in Feigl, 1972:586).

The above quotation hits two  important points within  the context of our discourse.  One is the  discovery of Carnap’s  conception of scientific schemes as dependent on language.  According to Carnap, one can   speak  of new entities so long as he devices a new system   of language for doing so.  Consequently, questions of ontology are to an extent questions of schemes or internal questions.   This  gave a good lead to  Quine’s theory of Ontological  Relativity,  which uses  background  as its reference point.

Secondly, it reveals Quine’s main  disagreement with  Carnap as an individual and Logical  Positivism as a movement.  Quine argues that Carnap’s distinction  between  internal  and external questions of  existence is founded on the distinction between analytic and synthetic statements, (Quine cited in Feigl, 1972:601).  This distinction he rejects, in his Two Dogmas of  Empiricism (1971), as an  unempirical  dogma of empiricists, a metaphysical  article  of faith (Quine, cited in Feigl, 1972:89).

For Quine, the only  real questions are the internal questions and nothing more.  Quine observes that Carnap’s  adherence to the contrary  view promotes illusion, rather than knowledge.  The basic illusion is reductionism, which is  the belief that there is one-to-one correspondence between  statements and the external world.

Thus,  to overcome such illusions as expressed by Carnap, Quine offers the  following clarifications  regarding ontological  issues:  “Ontological questions, likewise questions of logical  and mathematical  principles (and every scientific hypothesis) are questions not of fact but of choosing a convenient conceptual  scheme or framework for science”  (Quine, cited in Feige, 1972:601).

To hold  the contrary  view, therefore,  is to confuse  truth  with an illusive desire.  Quine argues this thesis very extensively in his Two Dogmas of Empiricism.  The consequence is the rejection of absolute reference.  The rejection  gives rise to relative reference which carves an avenue for ontological  relativity.   It reduces ontological  controversy to that about  language or schemes  (i.e the quest for correspondence with background language).  The Quinean belief is  based on  the argument that the output is greater than the input.  The consequence is the untenability  of reduction  and by  implication synthetic-analytic cleavage which presumes the truth of the former (reductionism).  Thus,  the  same linguistic consideration of ontological issue,  which influenced Carnap,  led Quine to reject the dogma, which Carnap professes with it.  The secret  is Quine’s ability to clear the illusion that beclouds Carnap’s thoughts;  misconception of meaning and reference, leading to ontological  confusions.

CHAPTER TWO

THE PROBLEM OF ONTOLOGICAL   CONFUSION

1.                 Wrong Characterization  of Ontological  Statements.

Radical reductionism  which is  the view that every  meaningful statement is that which is translatable into a  statement  (true or false) about  immediate experience is hooked  up with  the verification theory of meaning.   “This  theory  states that the meaning of a statement  is the method of empirically confirming or infirming it (Quine, cited in Feigh, 1972:90).  Here, there would have to be a limiting  case,  which is confirmed  come what may.  The statement in this case is analytic.  What this reveals  is a dual characterization of statements, viz analytic or synthetic and a belief that a meaningful statement must be a statement  with  empirical  reference.  What is evident in the second part of this confused characterization of statements  as Quine would  put it, is the confusion of  meaning with reference.  

Quine believes, that the clarification  of these concepts and their correct  employment in ontological  discourse is a route to having  a sound ontological  theory.

2.                 Synthetic – Analytic Cleavage

The  synthetic  - analytic cleavage as shown above is related to the verification theory of meaning, which  in turn is related to reductionism.  In this connection, the analysis of one logically leads to the analysis of the other.  Following the example of Quine, which  appears to make the  matter simple  and clear this  paper begins with the analysis  of synthetic – analytic dogma, because, it allows for careful attention  to analyticity.  It synthetic aspect is well  represented in the treatment of reductionism.

The Quinean point of departure on the investigation of the distinction  between  analytic and synthetic statements hinges on the investigation of  the meaning of  analyticity.  Reporting the Kantian  view on the matter, he states, that an analytic statement is one, the truth of which depends on  meaning and independent of facts,  whereas a synthetic statement, is one,  the truth of which is dependent  on facts  (Quine, cited in Feigl, 1972:81).  Quine,  leaves off the analysis of synthetic statement to  face its analytic counterpart, because he believes that the former has its place in radical  reductionism,  which shall be examined afterwards.  He begins by asking for  the meaning of analyticity.  In asking  this question, he threatens the foundation of the synthetic-analytic cleavage.  According to him “both dogmas .. are ill  founded”  (Quine, cited in Feigl, 1972:81).  Thus, he  calls to question the various  basis on which the  first dogma is founded.  They  include; meaning,  definitions,  interchangeability  and semantical rule.

In the analysis of meaning to discover how much  it can  support the  notion of analyticity, Quine warns that meaning  should not be  identified with naming  because terms can name  the same thing but differ in  meaning (Quine,  cited in Feigl, 1972:81).  Quine urges that the naming  view of meaning should be abandoned for its alternative, which holds that “the primary business of the theory of meaning is simply the synonymy of linguistic forms and the analyticity of statements” (Quine, cited in Feigl,  1972:82).  But this only  resurrects the  problem  of analyticity.

In want of progress,  Quine  resorts to the  use of  popular examples; They are:

No unmarried man is  married  -   (1)

No bachelor is married              -     (2), (Quine, cited in Feigl, 1972:82).

Quine observes that the analytic statement  of the first class remains true under all possible  reinterpretations of its components other than the logical  particles (Quine, cited in Feigl, 1972:82).  Yet the analyticity of those of the second class is unclear.

According  to him, there has been  a belief that statements of the second class can turn  to those of the first class by putting  synonyms for  synonyms.  But  he argues, that the notion  of synonymy is in need of clarification as  analyticity itself.

Quine  also observes, that some  persons  have argued, that the  analytic statement  of the second class can turn  to those of the  first by  definition.  But  Quine  argues, that definition reports selected instance of synonymy and such arises from usage (Quine, cited in Feigl, 1972:82).  “So,  just what the  interconnections may be, which  are necessary  and sufficient in order that two linguistic forms be properly  describable as synonymous, is  far from clear” (Quine, cited in Feigl, 1972:83).  The notion of interchangeability depends again on cognitive synonymy (recognizable at first sight).  But such according to Quine, depends on the knowledge of analyticity of statements.  It is only an analytic statement that can offer such cognizance.  This argument is back to the  notion of analyticity.  Quine observes this in  the following statements:

Analyticity at first seemed most naturally  definable by  appeal to a realm of meanings.  On refinement, the appeal  to meanings gave way to appeal to synonymy or definition. But definition turned out to be best understood only by  dint of a prior appeal  to analyticity itself (Quine, cited in Feigl, 1972:87).

One may  ask,  what is analyticity?  Some have argued that the question can be answered by recourse  to artificial language.  But Quine argues, that recursion to semantical rules and artificial   language is incapable of solution to the problem.  According to him:

The gravity of the problem is not  perceptibly less for the artificial languages than for  natural  ones.  The problem of making sense of the idiom ‘S is analytic for L’, with  variable ‘S’ and ‘L’ retains its stubbornness even if  we limit the range of the variable ‘L’ to  artificial   language (Quine, cited in Feigl, 1972:87).

Quine argues, that the problem  with such expression  is that “we understand what expressions the rule attribute analyticity to, but we do not understand what the rule attributes to those  expressions”  (Quine, cited in Feigl, 1972:88).  In  other words we are unaware  of what ‘analytic’ or  ‘analytic for’ means.  Furthermore, it is difficult to say what a semantical rule is except by identification as heading in writing.  Consequently  semantical rule stands in need of explanation as analyticity, itself.

After the  foregoing exposition of the difficulties inherent in any adherence to the synthetic-anlytic cleavage, Quine states an equally  faulty but most likely  characterization of ontological statements, as follows:

It is  obvious that truth in general  depends on both  language and extra linguistic fact.   The  statement ‘Brutus killed Caesar’ would be false if  the world had been   different in certain  ways,  but it would also be false if the  word ‘killed’ happened rather to have the  sense of ‘begat’.  Thus,  one is tempted to suppose in general  that the  truth of a statement  is  somehow analyzable into a linguistic component and a  factual component.  Given  this  supposition it next seems reasonable  that in some  statements the  factual component should  be null, and these are analytic statements (Quine, cited in Feigl, 1972:89).

Quine does not  take side with this supposition.  For  to do so, would   be to encourage an illusion  as well as give rise to  radical  reductionism.  This  explains why Quine decided to call the supposition a ‘temptation’. A clearer version of his interpretation is expounded in Reductionism, below.

3.                 Reductionism

Quine  becomes really  strict over the issues of  the  charaterization of ontological statements by positing that:

It is nonsense and the root  of much  nonsense to speak of a linguistic component and a factual  component in the truth of any individual statement.  Taken collectively science has its double dependence  upon language and experience; but this duality is  not significantly traceable into the  statement of science taken  by one (Quine,  cited in Feigl, 1972:92).

This position was not  introduced by Quine.  It had already been held  by  Duhem.

Accordingly, for him, it is the whole  science that faces the  tribunal of experience, not individually but as a corporate body.  Thus, the demand for a one-to-one correspondence between our statement  and the external  world is absurd.  Quine has drawn this conclusion to include the whole of his philosophical system.  The act  is based on his philosophy  that a clear conception of the theory  of meaning and reference holds the key to a sound ontology.  The reaction was also apt  against the dogma of reductionism, which states that “every meaningful  statement is held to be translatable into a statement  about  immediate experience” (Quine, cited Feigh 1972:90).  The beliefs of the proponents of this doctrine is that to each  statement or each  synthetic statement,  there is associated a unique range of possible sensory events, such that the occurrence of any of them  would add to the likelihood of truth of the statement,  and that  there is  associated also  another unique range of possible  sensory  events whose occurrence  would  detract from the likelihood  (Quine, cited in Feigl, 1972:91).  But this view according to Quine, is untrue and can only  be held by one who, until  now, is  unaware of the truth, that our  output  is more than our input.  Thus, one-to-one correspondence of reference, with regards to the external  world  is not possible.  Yet our ontology  can be  questioned relatively  to background schemes.

Thus, the search  for absolute reference which is based on the belief that our  statements are reports of immediate experience is misleading.  Quine makes this point  in the following  observations:

the totality  of our  so called knowledge or beliefs, from the  most casual matters of geography  and history to the  profoundest  laws of atomic physics or even of pure mathematics and logic, is a man-made fabric which impinges on experience only along the edges (Quine, cited in Feigl, 1972:92).

In this  position,  Quine  destroys any possible basis for either of the dogmas and creates a new conception of our science.  Science is a tool in the hands of men for  structuring of their experience and the prediction  of future possible  occurrences.  It is absurd, therefore, to query  such a scheme for absolute correspondence with  experience.    Quine observes that the ground for the temptation to such querying is reductionism which is itself  based on the confusion  of meaning with naming (reference).  The  consequence of the above for ontology is the inability to deny ontological  statements about non being, and it results in the imputation of being,  where we could have been content to acknowledge nothing (Quine, cited in Feigl, 1972:545).  Thus, radical  reductionism  is the basis of ontological confusion.

4.                 The Basis of  Ontological Confusion

The basis of ontological confusion  is radical  reductionism.  Any other confusion  is built  on it.  The belief that every  meaningful statement is reducible to statements about  immediate experience  has serious ontological  implications.  There is first,  the equation  of meaning  with  naming.  Thus, for a statement to be meaningful, it  has to  foot an ontological  bill.  It will have  to name some  entities.  Though Quine,  has repudiated this   point, his analysis of the issue  has up till now not surfaced in this project.  The second implication is the  imputation of non being  where we would  have been  content to acknowledge nothing.  A typical  example of this is what Quine  calls the  Plato’s Beard  (Quine, cited in Feigl, 1972: 545).  According  to the Beard,  nonbeing must in some sense be, otherwise,  what is it that there is not.  The notion of “it that there is not”, here denotes that the mention of nonbeing refers to some named entity otherwise that statement would have been meaningless.  Thus, one consequence has root  in the other.    The beard Quine  opines, has ‘for ages dulled  the edge of Ockham’s razor (Ockham’s razor states that entities need not be  multiplied beyond necessity).  According to the view,  it is nonsense to try to  deny the existence of nonbeing (nothing)  at all.  If it is non-existent, it means we cannot be talking about  anything when  we mention it.  Thus any  attempt to deny it tantamount to nonsense.  Yet when  we talk about it,  something of it is understood, which  means that it has being  in some  sense,  otherwise our statement will not be meaningful.  Note that meaning here is   used as if it were naming.  Just like the denial  of a known  and a real being  tantamounts to self contradiction the denial  of nonbeing results in nonsense.  Thus, to avoid this quandary and to maintain  coherence with the system, nonbeing  is   imputed with being.  Such that nonbeing is said to be.

The difficulty experienced, therefore,  in any attempt to deny affirmative statements  about   nonbeing, is  inability to formulates the denial without incoherence and confusions.

Quine observes that  any affirmation  of the being  of nonbeing has no sound basis.  The only possible  reference is to the fact, that nonbeing exist either as ideas in the minds of  men or as some unactualised  possibles.  But  Quine  argues, that the denial  of nonbeing  is not the denial  of the idea in mentality, but the  view,  that there  is no  such thing within  space and time.  Even the  notion of mentality  as a real existence is unacceptable to Quine.

The  consequence of the commitment  to this  type of  Ontology  is the needles multiplication of entities in the world  (the dulling  of the edge of Ockham’s  razor).  This confusion  is rooted in  reductionism which in turn  is rooted in the  view of meaning in terms of   objective reference (naming).  Quine  views that to move a step further in the  formulation of sound  ontology, one has to clear  the beard and the confusion in the conception  of meaning as naming.  This feat according  to Quine, has been  achieved by Russell in his Theory  of description.

CHAPTER THREE

QUINEAN  SOLUTION TO THE PROBLEMS OF MEANING AND REFERENCE

1.                 The Theory  of Description

The analysis  of the Quinean  system   of thought reveals that a clear understanding of the theory of meaning and reference holds the key to  sound  ontological  commitment.  This explains his effort in employing  Russell’s theory  of Description as an instrument for demonstrating the possibility of  speaking without  any  particular ontological  commitment, especially, to  universals.

The Russellian  theory  of description is of the  opinion, that there is  a sense in which we can speak  meaningfully about  objects without countenancing an ontology.  This  we can do through contextual definition.  This definition Quine  argues was first developed  by Bentham,  who called it paraphrases and  has since remained  dormant until  it flowered lately  in Russell’s theory of description.  Russell’s theory  affords a rigorous and an important way of how expressions can be made to parade as names and then be  explained away as a mere manner  of speaking, by  explicit paraphrase of the context into  an innocent notation.

The  main  aim of  the theory as used here by Quine, is to counter reductionism  and thus  demonstrate that terms need not name  entities to be meaningful.  It is also aimed to show how statement about nonbeing can be negated without contradiction or confusion.    So the  theory  uses descriptive names which are themselves complex names.    For instance,  the king of France, the author of  Waverly  and the round square copula in Berkeley  college.  They are actually  to be  fragments of sentence.  Then descriptive phrases are to be   followed by  bound  variables or variables of quantification, which  are: every, nothing, something.

Now taking  the author of Waverley  was a  poet as an instance, Russell proceeds thus:  Something  wrote Waverley and was a poet and nothing else wrote  Waverley (the  second part is added because of the article “the”). The  alternation of this gives; either each thing failed to write  Waverley or two or  more things  wrote Waverley.  This  Quine believes negates that first assertion   without being  guilty of contradiction or incoherence,  and without  naming any particular entity,  it is  meaningful.  The bound variable  in this case are free and are not tight down to naming particular entities.  But they  can  represent anything that falls within  their range. 

“To subsume a one word term under Russell’s theory  the word  has to be translated,  first, into  description” (Quine, cited in Feigl 1972:548).  Thus,  the word ‘Etorobong is’, can be translated into “the  youngman that studied at Unical in 2002”.  After that,  the  same  process as above,  can be applied.  If the  word is  obscure, Russell prescribes the  “ex-hypothesis” – being  Etoroobong.   Hence,  it could be translated into  is – Etorobong or the thing  that Etorobongizes.  In this  way Quine thinks  that  Russell’s theory of description has surmounted the Plato’s  Beard, which states that nonbeing  must in some sense be otherwise what is it, that there is  not (Quine, cited in Feigl, 1972:545).  Quine also feels  that the confusion  of meaning  with naming  has been  overcome.  A statement  can be  meaningful without  purporting  to name entities.

Quine believes that his new disposition  makes ontological commitment  free and selective.  We commit ourselves to the ontology of a particular entity when we affirm its being but we do not when we negate it.  So it is all left to us to decide to allow our bound variables range over a certain  realm  of entities.

Though the above  analysis tells us how to avoid  the beard and the confusion  of meaning with reference,  it tells us nothing  about  meaning and reference themselves.  What is meaning?  What is  reference?   How are they to be conceived?

2.                 Meaning and Translation 

The  Quinean  approach to the study  of meaning  is naturalistic.  It is  all hooked up with the naturalism of Dewey.

Here, meaning is to be  studied empirically.  This is  informed by  the understanding  of meaning as a property of behaviour.  Quine observes that Dewey was explicit on this point: “meaning… is not a psychic existence; it is primarily a property of behaviour; (Dewey, cited in Quine, 1969:27).  The naturalistic conception of meaning stands in opposition to Wittgenstein’s copy theory  which  is a good  version of uncritical  semantics.  In this light  naturalism posits that it is the  very fact about meaning  and  not meant entities that must be  construed in term of overt behaviour.

The mode of behaviour   of which meaning is a property is language. “Language is a social art which we acquired on the evidence solely of other people’s overt bahaviour under publicly recognizable circumstance.  Language is specifically a mode of interaction of at least two beings, a speaker and a hearer… It is, therefore, a relationship” (Quine, 1969:27). Therefore, meaning as a   property of behaviour, is a property of language. Meanings are first and foremost meanings of language.

Consequently, to know the meaning of an expression is to take into consideration the overt behaviour of the speaker.  This involves, hearing the phonetic part, being conscious of the stimulus and being aware that the speaker is responding to that particular stimulus and not another.  In this way, the meaning of the expression would be revealed.  Thus,  Quine advices that,  “even in complex and obscure parts of language learning, the  learner has no data to work  with but the  overt behaviour of the other speaker” (Quine  1969:28).

The belief implicit in this demonstration is absolute behaviourism.   This view that all of people’s mental life is expressible in overt behaviour leaves much to be desire. Yet even if this were to be granted, Quine still observed that there is a problem with his naturalistic conception of meaning.  This observation, he opine as follows: “when with Dewey we turn thus toward a naturalistic view of language and a behavioural view of meaning, what we give up is not just the museum figure of speech.  We give up an assurance of determinacy.   Thus,  even though  language  of which  meaning is a property  is socially  learnt,  it is  still  difficult to say when  a particular behaviour had  one meaning and not another, (Quine 1969:28).

Using the notion of likeness in meaning to expose the difficulty involved in determining the meaning of an expression, Quine states the following:

When… we recognize with Dewey that “meaning… is primarily a property of behaviour”, we recognize that there are no meanings, nor likeness nor distinctions of meaning, beyond what are implicit in people’s dispositions to overt behaviour.  For  naturalism the question whether two expressions are alike or unlike in meaning has no determinate answer, known or unknown, except in so far as the answer is settled  in principles by people’s speech dispositions,  known or unknown,  (Quine 1969:29).

In the final analysis, one discovers that this indeterminacy appears to be hopeless.  The solution given to it is somewhat arbitrary.  The   possibility of communication is lost and understanding becomes impossible.