Content uploaded by Eric Werner
Author content
All content in this area was uploaded by Eric Werner on Oct 05, 2018
Content may be subject to copyright.
Language, Mind and Society:
A Critique of Previous Theories∗
Eric Werner †
Oxford Advanced Research Foundation
eric.werner@oarf.org
https://www.ericwerner.com
Abstract
Work on the theory of language has tended to go in three directions: Ref-
erentialism, representationalism and social theories of mind and language
-including symbolic interactionism and social functionalism. We give a
critical review of the conceptual foundations and limitations of each. We
show that none are adequate as a complete formal theory that integrates
language, mind and society. The work presents the critical, conceptual
and foundational motivation for a full theory of communication and co-
operation for social agents.
∗Please cite as: Werner, E. "Language, Mind and Society: A Critique of Previous The-
ories", Preprint, DOI: *Insert DOI listed in header of this article*, 2018.
†©Werner 2018. All rights reserved.
1
Eric Werner, Language, Mind and Society: A Critique of Previous Theories
Contents
1 Introduction 3
1.1 Referentialism ............................. 3
1.2 Representationalism ......................... 3
1.3 Social interactionism and functionalism .............. 3
1.3.1 Mead and symbolic interactionism ............. 3
1.3.2 Social functionalism ..................... 4
2 Referential theories of communication 4
3 Advantages of referential theories of meaning 4
3.1 Rigorous ................................ 4
3.2 Compositional ............................. 4
3.3 Coordinated with syntax ....................... 5
3.4 Related to the world ......................... 5
3.5 Relate language and truth ...................... 5
4 Critique of representational and social functionalist theories of
meaning 5
4.1 Problems with representational theories of meaning ........ 5
4.1.1 Not formal .......................... 5
4.1.2 Not compositional ...................... 5
4.2 Problems with functionalist theories of meaning .......... 6
4.3 Functionalism and representationalism ignore the world ..... 6
4.4 The referentialists appear to be the winner ............ 6
5 Critique of referential theories of meaning 6
5.1 Restricted scope: Referentialists fail to cover a continent ..... 7
5.2 No mind no meaning ......................... 7
5.3 No social action ............................ 7
5.4 No explanation of social coordination ................ 7
5.5 No theory of understanding ..................... 8
5.6 No theory of language acquisition .................. 8
5.7 No explanation of speech pathologies ................ 8
5.8 No account of emotions as having evaluative effect ........ 8
6 Conclusion: A global unified theory of language 9
References 9
Cite as: Werner, E. "Language, Mind and Society: A Critique of Previous Theories", Preprint, DOI:
*Insert DOI listed in header of this article*,2018.
2
Eric Werner, Language, Mind and Society: A Critique of Previous Theories
1 Introduction
Work on the theory of language has tended to go in three directions: Refer-
entialism, representationalism and the symbolic interactionism including social
functionalism. We look at each group in turn. The work presents the criti-
cal, conceptual and foundational motivation for a full theory of communication
and cooperation for social agents[24,25]. For an informal introduction to the
concepts underlying my theory of communication and cooperation see1.
To avoid confusion critique will not cover Shannon’s information theory[17].
Shannon’s theory is a purely syntactic theory and not a theory of meaning.
For a more detailed critique of the misuse and misapplication of Shannon’s
information theory see Werner[29].
1.1 Referentialism
One group, the referentialists, has emphasized the relationship between language
and world. Here the central problems are to explain how language refers to the
world, how language says things about the world, and how language communi-
cates information about the world. Members of this group include Tarski[18],
Russell, Early Wittgenstein[32], Montague[13], Davidson, and Barwise[1].
1.2 Representationalism
A second direction, the representationalists, is represented by those who have
emphasized and tried to give an account of the relationship between language
and mind. They realized that to give a theory of meaning for a language one
must also give a theory of understanding-to explain how language is understood.
Other problems are to account for how a language is learned, how speech is
possible, what psychological structures are necessary for the capacity to learn,
speak and understand a language. The central hypothesis is that a theory
of language must involve variables that range over cognitive, mental states.
Members of this group include Winnograd[31], Fodor[7] and Chomsky[3,4].
1.3 Social interactionism and functionalism
The third direction has focussed on the relationship between language and so-
ciety. This direction has two subgroups: the symbolic interactionists and the
social functionalists.
1.3.1 Mead and symbolic interactionism
One outstanding member of the first subgroup, G. H. Mead, recognized that
the very possibility of a society depends on language[12]. One of the most
1Werner, E. ’A Brief Conceptual History of Communication Theory for Distributed
Social Agents in Social-Economic Ecologies.’, Preprint, October 5, 2018. DOI:
10.13140/RG.2.2.22151.24487
Cite as: Werner, E. "Language, Mind and Society: A Critique of Previous Theories", Preprint, DOI:
*Insert DOI listed in header of this article*,2018.
3
Eric Werner, Language, Mind and Society: A Critique of Previous Theories
fundamental problems in social theory is to give an account of how language or
linguistic communication coordinates social action. The implicit presupposition
is that any theory of language must be capable of serving social theory; it
must give an account of how language functions to make social action possible.
Mead and his followers (Habermas[8] who adapted Searle[15,16] who adapted
Wittgenstein[34,33]) thus focus on the relevance of linguistic theory to social
theory.
1.3.2 Social functionalism
The other subgroup, the social functionalists, emphasize a somewhat reverse,
direction namely, the relevance of social function to linguistic theory. They
attempt to give a functionalist theory of meaning. The meaning of an expression
is given by its social function in a well defined subspace of a social economic
system. In the later Wittgenstein’s terminology the meaning of an expression
is given by its use in a language game[34,33].
2 Referential theories of communication
Referential theories of meaning have been preferred by some because they, it
is argued, are formalizable. By extending the model theoretic methods de-
veloped by Tarski, it is claimed, one can give a precise mathematical account
of the semantics of natural languages[18,19]. This is the approach taken by
Montague[13], Davidson, Barwise[1] and many others.
3 Advantages of referential theories of meaning
Such theories of linguistic meaning have several paramount advantages:
3.1 Rigorous
First, they are rigorous; they specify and define linguistic concepts exactly. This
makes the theory testable. The theories are precise enough as to be refutable.
The theories make specific predictions about semantic content that can be tested
against the linguistic observations.
3.2 Compositional
Second, these semantic theories are compositional. The meanings of complex
expressions are built up out of the meanings of their parts. In situation semantics
the meaning of the parts may be influenced by their linguistic, physical and
social context, but overall, compositionality is maintained.
Cite as: Werner, E. "Language, Mind and Society: A Critique of Previous Theories", Preprint, DOI:
*Insert DOI listed in header of this article*,2018.
4
Eric Werner, Language, Mind and Society: A Critique of Previous Theories
3.3 Coordinated with syntax
Third, formal referential theories are combinational or generative. The semantic
rules for building up the meanings of complex linguistic expressions are nicely
correlated with the generative syntactic rules. Indeed, the semantics is induc-
tively defined in such a way that for each generative syntactic rule there is
a corresponding semantic rule that defines the meaning of the new complex
syntactic expression in terms of its parts. Thus, combinationality requires com-
positionality. Combinationality allows us to define the meaning for an infinite
set of expressions of the language. And thus it allows the specification of our
true meanings.
3.4 Related to the world
Fourth, these theories relate language to the world. They start with the hypoth-
esis that language talks about situations in the world. They give an account of
how language talks about the world and, potentially, how language communi-
cates information.
3.5 Relate language and truth
Finally, they give us a theory of truth and validity. They give us an account
of the conditions of truth of complex sentences. Furthermore, they account for
the fact that there are valid relations of logical implication between sentences
of the language.
4 Critique of representational and social func-
tionalist theories of meaning
When compared with the accomplishments of the referentialists those of the
representationalists and social functionalists are rather meager. It is even ques-
tionable whether they are semantic hypothesis at all.
4.1 Problems with representational theories of meaning
4.1.1 Not formal
They are certainly not formal or rigorous theories. Instead they offer vague
slogans: The meaning of an expression is the idea, representation or other mental
state associated with the speaker hearer. Or, the meaning of the expression is
given by its use in a language game.
4.1.2 Not compositional
Whatever the slogan, because the semantic theories are not rigorous or detailed,
they are not and cannot be compositional. Thus, they offer no account of how
Cite as: Werner, E. "Language, Mind and Society: A Critique of Previous Theories", Preprint, DOI:
*Insert DOI listed in header of this article*,2018.
5
Eric Werner, Language, Mind and Society: A Critique of Previous Theories
the meaning of complex expressions is a function of the meanings of the parts.
Thereby, they are not generative or combinational either. They offer no coherent
account that correlates the formal syntax with the semantics.
Chomsky, a representationalist and one of the first to develop a generative
syntax of a fragment of English, never succeeded in constructing a formal com-
positional, combinational semantic theory. Chomsky’s semantic theory does not
give an account of the most elementary semantic facts[4,6,5]. For example,
how the meaning of the sentence ’John loves Mary or Beth’ differs from the
meaning of the sentence ’John loves Mary and Beth.’
4.2 Problems with functionalist theories of meaning
The functionalists such as Mead[12], Wittgenstein[34,33], Habermas[8,9], and
Searle[15,16] are even in a worse state. They are at a state of theory construction
that is at best classificatory. They have not even defined a rigorous syntactic
theory, let alone a formal semantic theory. They have nothing even close to a
compositional and combinational semantic or pragmatic theory.
Because Searle gives not formal semantics of speech acts his account is at
best a syntactic classification with informal descriptions and presuppositions
that the reader will understand the meaning. Other than a classification, he
never precisely defines what the force of speech act is. In contrast Werner[23]
presents a formal semantic and pragmatic theory of speech acts including a
formal theory of speech act force.
4.3 Functionalism and representationalism ignore the world
Both the representationalists and the functionalists tend to ignore the relation
of language and world. Because of this, they have a difficult time accounting
for the referential aspects of language. They can give no formal account of how
a language communicates information about the world. Furthermore, they offer
no theory of truth and validity. Finally, their vagueness, their lack of detail
and rigor have the devastating consequence that these semantic theories are not
really testable. One can argue their pros and cons ad infinitum.
4.4 The referentialists appear to be the winner
Faced with these well known facts, one can justifiably conclude that the refer-
ential theory is the most rigorous and workable theory of meaning we have. It
gives the best formal account of a significant proportion of the semantic data.
5 Critique of referential theories of meaning
Despite the overwhelming explanatory power of the referential theory, and even
though my own formal training has been in the halls of logic and referential
semantics, I cannot bring myself to accept the referential theory as a complete
or any way final account of what a language is and does.
Cite as: Werner, E. "Language, Mind and Society: A Critique of Previous Theories", Preprint, DOI:
*Insert DOI listed in header of this article*,2018.
6
Eric Werner, Language, Mind and Society: A Critique of Previous Theories
My reasons include the following:
5.1 Restricted scope: Referentialists fail to cover a conti-
nent
On the surface, the most embarrassing feature of the referential theory is that
it is restricted in scope. Usually, only assertions, sentences that are true or
false, are analyzed. For example, requests, statements of intention, commands
and threats tend to be shoved aside with the rhetorical ploy-that they will be
treated at some time in the future.
We will see that it is no accident that all these types of sentences or messages
belong to a common class. They are all used to regulate social interactions. Let
us refer to them as strategic sentences or strategic messages.
Referentialists, thereby ignore and avoid a whole continent of linguistic phe-
nomena that are absolutely essential for the most important function of language
which is to govern and coordinate human social activity. Below the surface, be-
cause the referential theory restricts itself to the relation of language and world,
it ignores mental states. Thus it offers no account of the relation of language
and mind.
This has some consequences:
5.2 No mind no meaning
In particular, to analyze strategic messages, I will argue that we need a semantic
theory of language that includes mental states as essential variables. Specifically,
we need a theory of intentionality and a theory of language that uses and fits
well with such a theory of intentions as well as other mental states. For example,
we will argue that the force or point of a command is that it effect the intentions
of the hearer in such a way that he will attempt to carry out the content of the
command.
5.3 No social action
The referential theory also gives no account of how linguistic communication
functions in society. There is no explanation of the phenomena that language is
used to govern human social activity. Not only the analysis of strategic sentences
may difficult by the lack of an adequate psychology of the agent-speaker, but
so is the explanation of how language functions in social action. For, we will
see that any deep explanation of linguistic functioning in society also requires a
theory of intentions and a theory of the relation of intentions and language use.
5.4 No explanation of social coordination
Hence, in avoiding the relation of language and mind, referentialists give up
any hope of explaining how language functions in the coordination of social
action. This has the unfortunate consequence that their linguistic theory is, one
Cite as: Werner, E. "Language, Mind and Society: A Critique of Previous Theories", Preprint, DOI:
*Insert DOI listed in header of this article*,2018.
7
Eric Werner, Language, Mind and Society: A Critique of Previous Theories
might say, unsociable. There is no possibility of providing the foundations of a
theory of linguistic communication in society. The obvious important connection
between social processes, social organization and linguistic communication is not
accounted for both at a semantic level, in that strategic sentences are not given
an adequate semantics, as well as at the level of theory construction in that the
referential linguistic theory is not compatible with the foundations required for
social theory.
5.5 No theory of understanding
In ignoring the mind, the referential theory offers us no deep theory of under-
standing. According to some referentialists, to understand the sentence is to
know its truth conditions. There is no account of understanding a sentence if
it has no truth conditions. Furthermore, no explanation is given as to what is
involved in knowing the truth conditions.
5.6 No theory of language acquisition
The problem of language acquisition, how are language is learned, is difficult
to formulate without implicit reference to the mind and its function in relation
to language. And, again, the referentialists tend to ignore this issue. Language
translation is another problem difficult to treat in the referential framework.
For it too seems to involve psychological representations.
5.7 No explanation of speech pathologies
Another problem is that in the referential framework gives no explanation of
the possibility of language and speech pathologies. There is no connection be-
tween language and mental functioning. No theoretical framework in which the
question in the relation of language pathology and mental functioning can even
be formulated. Mental states and mental functioning is ultimately realized in
a new physiological base. Because of this, language pathologies resulting from
disturbances in the realization of functional mental states are not explainable
or even describable in a referential theory.
Conversations and the principles that guide human dialogue also have a
difficult time trying to find a place in the referential home. At the heart of the
matter, is the fact that human goals and interests and principles of rationality
related to interests seem to govern the formation of the linguistic intentions
of the conversions. And yet, interests of values, rationality, and intentions are
too close to psychological sphere to be adequately represented in a referential
theory. Truth and reference are just one aspect of a conversation.
5.8 No account of emotions as having evaluative effect
What may be a minor point to those whose paradigm is the physical sciences
but most important for the psychologist, economist and social scientist, is that
Cite as: Werner, E. "Language, Mind and Society: A Critique of Previous Theories", Preprint, DOI:
*Insert DOI listed in header of this article*,2018.
8
Eric Werner, Language, Mind and Society: A Critique of Previous Theories
a whole other class of messages is ignored by the referentialists as well - the
evaluative sentences. I mean those that express emotion, value and utility. For
example, to analyze "loves" of "likes" simply as a relation between objects misses
its central aspect, namely, the emotion and evaluation involved. By ignoring
the deeper meaning of evaluative messages, social utility theory and its relation
to language is thereby not adequately formalizable in the referential framework.
Neither is the effect of language on the emotional evaluative states formalizable.
The referentialists might try to include the utility functioning of the agent
into the satisfaction conditions of evaluative assertions, but as soon as he does
so he has implicitly brought psychological states into the picture.
6 Conclusion: A global unified theory of language
Although the different groups have been in continuous conflict with those of
different persuasions, I believe that each of these directions has recognized an
important aspect about the nature of language and communication: Language
talks about the world. Language and mind are intimately related. Linguistic
communication is the foundation of social action and organization. And to
fully understand the nature of language one cannot ignore its function in the
social-economic ecology.
The purpose of this article is to motivate a formal theory of language that
attempts to be global in the sense that it attempts to account for all of the
above dimensions of language, in particular the relation of language and world,
language and mind, and language in society. Such a theory must integrate
the distinction made by Morris[14] some 80 years ago between semantics and
pragmatics. My theory of communication attempts to achieve this integration
of language, mind and society[24,25,26,27,28,22,29,30] For an informal
introduction to the concepts underlying my theory of communication and coop-
eration see2.
References
[1] J. Barwise and J. Perry. SITUATIONS AND ATTITUDES. Bradford Books, MIT Press,
1983.
[2] L. Boltzmann. Gesamtausgabe: Ludwig Boltzmann. Akademische Druck und Ver-
lagsanstalt, Braunschweig, 1981.
[3] N. Chomsky. Syntactic structures. MIT Press, 1957.
[4] N. Chomsky. Generative grammar. Studies in English Linguistics and Literature, 1988.
[5] N. Chomsky. Aspects of the Theory of Syntax, volume 11. MIT press, 2014.
[6] N. Chomsky and D. W. Lightfoot. Syntactic structures. Walter de Gruyter, 2002.
[7] J. A. Fodor. The language of thought. Language & thought series. Crowell, New York,
1975.
2Werner, E. ’A Brief Conceptual History of Communication Theory for Distributed
Social Agents in Social-Economic Ecologies.’, Preprint, October 5, 2018. DOI:
10.13140/RG.2.2.22151.24487
Cite as: Werner, E. "Language, Mind and Society: A Critique of Previous Theories", Preprint, DOI:
*Insert DOI listed in header of this article*,2018.
9
Eric Werner, Language, Mind and Society: A Critique of Previous Theories
[8] J. Habermas. Theorie des kommunikativen Handelns. Frankfurt am Main: Suhrkamp„
1981.
[9] J. Habermas. The theory of communicative action. Beacon Press, Boston, 1984.
[10] S. Kripke. Semantic analysis of modal logic i, normal propositional calculi. Zeitschr. fr.
Math. Logic und Grundlagen d. Math., 9:67–96, 1963.
[11] D. MacKay. Information, Mechanism and Meaning. MIT Press„ Cambridge, MA, 1969.
[12] G. H. Mead. Mind, Self, and Society. University of Chicago Press, 1934.
[13] R. Montague. Formal philosophy; selected papers of Richard Montague. Yale University
Press, New Haven, 1974.
[14] C. Morris. Foundations of the theory of signs. In Writings on the Theory of Signs.
Mouton, The Hague, 1938.
[15] J. R. Searle. Speech acts: An essay in the philosophy of language. Cambridge University
Press, London, 1969.
[16] J. R. Searle. Expression and meaning : studies in the theory of speech acts. Cambridge
University Press, Cambridge, Eng. ; New York, 1979.
[17] W. W. Shannon, C. The Mathematical Theory of Communication. University of Illinois
Press, Urbana, 1949.
[18] A. Tarski. Contributions to the theory of models, i’. Indagationes Mathematicae,
16(572?581), 1954.
[19] A. Tarski. Logic, Semantics, Metamathematics: Papers from 1923 to 1938. Clarendon
Press, Oxford, 1956.
[20] J. von Neumann. Mathematical Foundations of Quantum Mechanics. Princeton Univ.
Press, 1996.
[21] J. Von Neumann and O. Morgenstern. THE THEORY OF GAMES AND ECONOMIC
BEHAVIOR. Princeton University Press, Princeton, NJ„ 1947.
[22] E. . Werner. The ontogeny of the social self. towards a formal computational theory. In
K. Dautenhahn, editor, Human Cognition and Social Agent Technology, pages 263–300.
John Benjamins, 1999.
[23] E. Werner. A formal computational semantics and pragmatics of speech acts. In
COLING-88, Proceedings of the 12th International Conference on Computational Lin-
guistics, pages 744–749, Budapest, Hungary, 1988.
[24] E. Werner. Toward a theory of communication and cooperation for multiagent planning.
In M. Vardi, editor, Theoretical Aspects of Reasoning About Knowledge: Proceedings of
the Second Conference, pages 129–143. Morgan Kaufman Publishers, 1988.
[25] E. Werner. Cooperating agents: A unified theory of communication and social struc-
ture. In L. G. Huhns and M., editors, Distributed Artificial Intelligence, Vol. 2. Morgan
Kaufmann and Pitman Press, 1989.
[26] E. Werner. Distributed cooperation algorithms. In J. Dernazeau, Y.; Mueller, editor,
DECENTRALIZED AI. Elsevier Science Publishers B.V. (North Holland), 1990.
[27] E. Werner. A unified view of information, intention and ability. In J. Demazeau, Y.; Muel-
ller, editor, Decentralized AI, volume 2. Elsevier Science Publishers, 1991.
[28] E. Werner. Logical foundations of distributed artificial intelligence. In G. O’Hare and
N. Jennings, editors, Foundations of Distributed Artificial Intelligence. Wiley, 1996.
[29] E. Werner. Meaning in a quantum universe. Science, 329(5992):629–630, 2010.
[30] E. Werner. A category theory of communication theory. arXiv:1505.07712 [cs.IT],
http://arxiv.org/abs/1505.07712, 2015.
[31] T. Winograd. Understanding natural language. Academic Press, New York, 1972.
[32] L. Wittegnstein. Tractatus Logico-Philosophicus. Routledge & Kegan Paul Ltd„ London,
1961.
[33] L. Wittgenstein. The Blue and Brown Books. Blackwell, Oxford, 1958.
[34] L. Wittgenstein. Philosophical Investigations. Blackwell, Oxford, 1958.
Cite as: Werner, E. "Language, Mind and Society: A Critique of Previous Theories", Preprint, DOI:
*Insert DOI listed in header of this article*,2018.
10