This chapter discusses natural language ontology, focusing on the nature of sentences. Two contrasting views about such elements are considered. First, the naturalistic view takes sentences to be elements of the physical world. There are two variants. One, which regards sentences as utterances, has few if any current advocates. The other, which views sentences as mental/biological things, is currently dominant, defining the position of Noam Chomsky. Second, there is the nonnaturalistic, Platonist view advocated intensively by Jerrold J. Katz, which takes sentences to be abstract objects. This view is consistent with the fact that sentences are timeless, locationless entities entering into no causal relations. Only the naturalistic view is inconsistent with actual linguistics, where sentences are uniformly treated in set-theoretical terms.
I argue that what determines whether a science is ‘formal’ or ‘empirical’ is not the ontological status of its objects of study, but, rather, its methodology. Since all sciences aim at generalizations, and generalizations concern types, if types are abstract (non-spatiotemporal) objects, then all sciences are concerned to discover the nature of certain abstract objects. What distinguishes empirical from formal sciences is how they study such things. If the types of a science have observable instances (‘tokens’), then the nature of the types may be determined empirically. If they types have either abstract tokens, or no tokens at all, their nature must be determined by non-empirical methods involving intuition, reasoning and proof. I conclude that the status of (theoretical) linguistics depends on the methodologies of syntax, semantics, phonology, morphology and orthography (and any other subdiscipline that is concerned with the study of the structure of language).
The ‘Biolinguistics’ program seeks to establish specific neuroanatomical models corresponding to the representations and operations characterizing the species-specific language faculty in human beings. Yet after decades of research, no neural structures corresponding to specific linguistic structures, rules, constraints or principles have ever been identified. A key to biolinguistics’ failure is, I suggest, its long-term adherence to two dubious assumptions: (i) a kind of literalism in envisaging the relationship between neural anatomy and linguistic representations, reflecting a seriously misconstrual of Marr’s (1982) tripartite division of cognition, and (ii) a view of such representations as objects fundamentally different from other components of human cognitive capacity. (ii) rests on the premise that phrase markers are the optimal formal representation of natural language sentences, despite major empirical difficulties that syntactic accounts based hierarchical phrase structure face in handling a wide variety of grammatical patterns, including non-canonical coordinations and ellipsis constructions. In contrast, proof-theoretic approaches such as type-logical grammar do not face these difficulties, and their foundational assumptions link language to the higher-order cognitive functions supporting deductive reasoning. This conclusion suggests a promising alternative to the current, essentially result-free ‘Biolinguistic’ paradigm.
It may appear counterintuitive to suggest a connection between language evolution and linguistic realism. Only biological objects evolve but linguistic realism holds that natural languages are abstract objects. However, given the fact that currently no approach to language evolution can account satisfactorily for all aspects of language, I suggest that reconsidering the ontological status of natural languages might lead to novel approaches to language evolution puzzles. Most contemporary work on language evolution assumes without argument that natural languages are either biological entities or produced by biological organs (human brains), and focuses on brain evolution, language acquisition, and communication systems of other primates. Yet, so far such approaches have been unable to account for some aspects of grammar. Furthermore, to date little is known about the bio-physiological implementation of natural languages. I suggest that the debate could profit from paying closer attention to the ontological status of language and the exact relationship between language and biology. Finally, I discuss the kinds of evidence used in linguistic research and demonstrate that, contra to widespread belief, the linguistic Platonist is neither relying on inferior evidence nor ruling out evidence that is clearly relevant to linguistic research.
The essay is divided into four Parts A to D: Part A (Sections 1 to 3), Topic and background; B. Grammatical description (Sections 4 to 6); C. Grammars and theories of language: motivating axiomatization (Sections 7 to 9); and D. Grammars as axiomatic theories (Sections 10 and 11). The essay characterizes grammatical description, both informal and formal, from a realist point of view as the description of abstract objects, not to be confused with the concrete data to which the description must be ultimately related. The importance of theories of natural human languages for grammars is emphasized, also in view of comparative grammar writing, and is demonstrated by the detailed analysis of a grammatical statement taken from an informal grammar. There is a discussion of adequacy problems that arise in current frameworks for formal grammars due to an absence of theories of language from such frameworks. A format for axiomatic grammars is outlined by which an axiomatic grammar ‘presupposes’ a theory of language, in a technical sense. The view of grammars is non-reductionist; concepts of theory integration are characterized that allow us to integrate grammars with linguistic and non-linguistic theories. The conception of linguistics itself is non-reductionist, too, through applying a concept of inter-discipline that relates linguistics to other disciplines.
My aim in this chapter is to extend the Realist account of the foundations of linguistics offered by Postal, Katz and others. I first argue against the idea that naive Platonism can capture the necessary requirements on what I call a ‘mixed realist’ view of linguistics, which takes aspects of Platonism, Nominalism and Mentalism into consideration. I then advocate three desiderata for an appropriate ‘mixed realist’ account of linguistic ontology and foundations, namely (1) linguistic creativity and infinity, (2) linguistics as a theory of types (and not tokens) and (3) independence but structural respect between language and the linguistic competence thereof. My own brand of mixed realism, what I call ante rem realism, is defended along the lines of an ante rem or non-eliminative structuralism, the likes of which has been offered for mathematics by Resnik (1997) and Shapiro (1997). In other words, grammars describe a mind-independent (but not necessarily unconnected) linguistic reality in terms of linguistic patterns or structures also known as natural languages. I further amend this picture to allow for the possibility of a naturalistic account of language acquisition and evolution by arguing against a particular view of the type-token distinction.
In the paradigm of Linguistic Realism, phonology deals with abstract objects. Consequently, phonological units cannot be derived from phonetics as the material side of speech production and reception. I suggest a theoretical approach that conceives of phonology as autonomous from phonetics; hence Autonomous Declarative Phonology. The central questions of this kind of phonology are: What are the phonological elements of a specific language system (in particular: German)? How can these phonological elements be combined in the language under analysis? I give a definition of what phonology is and show how individual phonological units can be motivated. In order to give a model-theoretic reconstruction of phonological sequences, I make reference to the concept of syllable, employing and re-interpreting ideas from generative phonology and other approaches such as hierarchical syllable structure, CV-phonology, and sonority.
The present paper examines foundational issues of a realist word-formation theory. A realist linguistic theory, as it is understood here, takes linguistic units and the linguistic systems that determine them to be abstract entities. With respect to such a word-formation theory, the following two questions are discussed:
What are the word-formation facts to be described and explained or predicted?
What linguistic objects are those word-formation facts about?
Presupposing the axiomatically formalized Pattern-and-Restriction Theory (PR), it is proposed that the word-formation facts to be described and explained or predicted are true statements of word-formation relations in the linguistic system under consideration, and that those facts are about abstract lexical units in the sense of the realist framework of Integrational Linguistics (IL). On the example of a word-formation pattern in some spoken Modern German system it is shown how deductive-nomological (DN) explanations or predictions of word-formation facts can be logically derived from theorems of the PR theory and theorems of a grammar and a dictionary of the linguistic system.
The paper presents a cognitive conception of propositions as semantic contents of (some) declarative sentences. The conception expands solution spaces for previously intractable empirical problems in natural-language semantics and pragmatics, while also explaining how an agent who is unable to cognize propositions can know or believe them, and how sophisticated agents acquire the concept and believe things about them by monitoring their own cognitions. Finally, an account is given of what it is for a sentence to mean that p in a language that doesn’t require having thoughts about p or L. Nevertheless, semantics isn’t psychology; agents with different psychologies can speak semantically identical languages, while those with the same purely internal states (embedded in similar immediate environments) can speak different languages. Cognitive semantics can be realist and naturalistic without being a branch of psychology.
Section 1 is an extended commentary on Edward Sapir’s formulation nearly a century ago of what he considered the two most fundamental properties of human language, first that each one is a formally complete system of reference to experience and second that each one is formally distinct from every other. Section 2 considers some aspects of the development of these formulations, noting that they have been considered separately and not integrated as fully fleshed out systems of reference, as Sapir envisioned. Section 3 examines more closely what such an integration looks like in a case involving simple arithmetic. Section 4 begins with a brief review of the accomplishments of Greco-Roman logic and more recent developments in the theory of logic, leading to a consideration of what may be needed to fulfill Sapir’s program. Section 5 summarizes some of my own recent research on extending first-order logic by replacing the unordered set of individuals with a specific ordering of a set of sets of individuals that is isomorphic to an ordering of sets of sets of numbers that contain no pairs of divisible numbers, which was investigated by Richard Dedekind shortly before the turn of the twentieth century.
Undoubtedly, F. de Saussure is among the founders of modern linguistics and ‘semiology’/semiotics – though, strangely enough, his success mainly resulted from a book he did not write in the strict sense. Since the publication of the Cours, edited by his disciples in 1916, his notion of ‘arbitrariness’ has become one of the main canonical catchwords of linguistics that are hardly ever called into question. After a reconstruction of the Saussurean concept of the linguistic sign and its arbitrariness the attempt is made to work out its shortcomings and partly correct them. It is argued that instead of arbitrariness motivation must be taken as the central semiotic feature of linguistic signs which, after their formation, may be subject to a continuous historical process of increasing arbitrarization, possibly leading to complete arbitrariness in the end. In the final section the question of whether and to what extent the Geneva linguist may be considered a realist is tackled.
Undoubtedly, F. de Saussure is among the founders of modern linguistics and ‘semiology’/semiotics – though, strangely enough, his success mainly resulted from a book he did not write in the strict sense. Since the publication of the Cours, edited by his disciples in 1916, his notion of ‘arbitrariness’ has become one of the main canonical catchwords of linguistics that are hardly ever called into question. After a reconstruction of the Saussurean concept of the linguistic sign and its arbitrariness the attempt is made to work out its shortcomings and partly correct them. It is argued that instead of arbitrariness motivation must be taken as the central semiotic feature of linguistic signs which, after their formation, may be subject to a continuous historical process of increasing arbitrarization, possibly leading to complete arbitrariness in the end. In the final section the question of whether and to what extent the Geneva linguist may be considered a realist is tackled.
This chapter discusses natural language ontology, focusing on the nature of sentences. Two contrasting views about such elements are considered. First, the naturalistic view takes sentences to be elements of the physical world. There are two variants. One, which regards sentences as utterances, has few if any current advocates. The other, which views sentences as mental/biological things, is currently dominant, defining the position of Noam Chomsky. Second, there is the nonnaturalistic, Platonist view advocated intensively by Jerrold J. Katz, which takes sentences to be abstract objects. This view is consistent with the fact that sentences are timeless, locationless entities entering into no causal relations. Only the naturalistic view is inconsistent with actual linguistics, where sentences are uniformly treated in set-theoretical terms.
I argue that what determines whether a science is ‘formal’ or ‘empirical’ is not the ontological status of its objects of study, but, rather, its methodology. Since all sciences aim at generalizations, and generalizations concern types, if types are abstract (non-spatiotemporal) objects, then all sciences are concerned to discover the nature of certain abstract objects. What distinguishes empirical from formal sciences is how they study such things. If the types of a science have observable instances (‘tokens’), then the nature of the types may be determined empirically. If they types have either abstract tokens, or no tokens at all, their nature must be determined by non-empirical methods involving intuition, reasoning and proof. I conclude that the status of (theoretical) linguistics depends on the methodologies of syntax, semantics, phonology, morphology and orthography (and any other subdiscipline that is concerned with the study of the structure of language).
The ‘Biolinguistics’ program seeks to establish specific neuroanatomical models corresponding to the representations and operations characterizing the species-specific language faculty in human beings. Yet after decades of research, no neural structures corresponding to specific linguistic structures, rules, constraints or principles have ever been identified. A key to biolinguistics’ failure is, I suggest, its long-term adherence to two dubious assumptions: (i) a kind of literalism in envisaging the relationship between neural anatomy and linguistic representations, reflecting a seriously misconstrual of Marr’s (1982) tripartite division of cognition, and (ii) a view of such representations as objects fundamentally different from other components of human cognitive capacity. (ii) rests on the premise that phrase markers are the optimal formal representation of natural language sentences, despite major empirical difficulties that syntactic accounts based hierarchical phrase structure face in handling a wide variety of grammatical patterns, including non-canonical coordinations and ellipsis constructions. In contrast, proof-theoretic approaches such as type-logical grammar do not face these difficulties, and their foundational assumptions link language to the higher-order cognitive functions supporting deductive reasoning. This conclusion suggests a promising alternative to the current, essentially result-free ‘Biolinguistic’ paradigm.
It may appear counterintuitive to suggest a connection between language evolution and linguistic realism. Only biological objects evolve but linguistic realism holds that natural languages are abstract objects. However, given the fact that currently no approach to language evolution can account satisfactorily for all aspects of language, I suggest that reconsidering the ontological status of natural languages might lead to novel approaches to language evolution puzzles. Most contemporary work on language evolution assumes without argument that natural languages are either biological entities or produced by biological organs (human brains), and focuses on brain evolution, language acquisition, and communication systems of other primates. Yet, so far such approaches have been unable to account for some aspects of grammar. Furthermore, to date little is known about the bio-physiological implementation of natural languages. I suggest that the debate could profit from paying closer attention to the ontological status of language and the exact relationship between language and biology. Finally, I discuss the kinds of evidence used in linguistic research and demonstrate that, contra to widespread belief, the linguistic Platonist is neither relying on inferior evidence nor ruling out evidence that is clearly relevant to linguistic research.
The essay is divided into four Parts A to D: Part A (Sections 1 to 3), Topic and background; B. Grammatical description (Sections 4 to 6); C. Grammars and theories of language: motivating axiomatization (Sections 7 to 9); and D. Grammars as axiomatic theories (Sections 10 and 11). The essay characterizes grammatical description, both informal and formal, from a realist point of view as the description of abstract objects, not to be confused with the concrete data to which the description must be ultimately related. The importance of theories of natural human languages for grammars is emphasized, also in view of comparative grammar writing, and is demonstrated by the detailed analysis of a grammatical statement taken from an informal grammar. There is a discussion of adequacy problems that arise in current frameworks for formal grammars due to an absence of theories of language from such frameworks. A format for axiomatic grammars is outlined by which an axiomatic grammar ‘presupposes’ a theory of language, in a technical sense. The view of grammars is non-reductionist; concepts of theory integration are characterized that allow us to integrate grammars with linguistic and non-linguistic theories. The conception of linguistics itself is non-reductionist, too, through applying a concept of inter-discipline that relates linguistics to other disciplines.
My aim in this chapter is to extend the Realist account of the foundations of linguistics offered by Postal, Katz and others. I first argue against the idea that naive Platonism can capture the necessary requirements on what I call a ‘mixed realist’ view of linguistics, which takes aspects of Platonism, Nominalism and Mentalism into consideration. I then advocate three desiderata for an appropriate ‘mixed realist’ account of linguistic ontology and foundations, namely (1) linguistic creativity and infinity, (2) linguistics as a theory of types (and not tokens) and (3) independence but structural respect between language and the linguistic competence thereof. My own brand of mixed realism, what I call ante rem realism, is defended along the lines of an ante rem or non-eliminative structuralism, the likes of which has been offered for mathematics by Resnik (1997) and Shapiro (1997). In other words, grammars describe a mind-independent (but not necessarily unconnected) linguistic reality in terms of linguistic patterns or structures also known as natural languages. I further amend this picture to allow for the possibility of a naturalistic account of language acquisition and evolution by arguing against a particular view of the type-token distinction.
In the paradigm of Linguistic Realism, phonology deals with abstract objects. Consequently, phonological units cannot be derived from phonetics as the material side of speech production and reception. I suggest a theoretical approach that conceives of phonology as autonomous from phonetics; hence Autonomous Declarative Phonology. The central questions of this kind of phonology are: What are the phonological elements of a specific language system (in particular: German)? How can these phonological elements be combined in the language under analysis? I give a definition of what phonology is and show how individual phonological units can be motivated. In order to give a model-theoretic reconstruction of phonological sequences, I make reference to the concept of syllable, employing and re-interpreting ideas from generative phonology and other approaches such as hierarchical syllable structure, CV-phonology, and sonority.
The present paper examines foundational issues of a realist word-formation theory. A realist linguistic theory, as it is understood here, takes linguistic units and the linguistic systems that determine them to be abstract entities. With respect to such a word-formation theory, the following two questions are discussed:
What are the word-formation facts to be described and explained or predicted?
What linguistic objects are those word-formation facts about?
Presupposing the axiomatically formalized Pattern-and-Restriction Theory (PR), it is proposed that the word-formation facts to be described and explained or predicted are true statements of word-formation relations in the linguistic system under consideration, and that those facts are about abstract lexical units in the sense of the realist framework of Integrational Linguistics (IL). On the example of a word-formation pattern in some spoken Modern German system it is shown how deductive-nomological (DN) explanations or predictions of word-formation facts can be logically derived from theorems of the PR theory and theorems of a grammar and a dictionary of the linguistic system.
The paper presents a cognitive conception of propositions as semantic contents of (some) declarative sentences. The conception expands solution spaces for previously intractable empirical problems in natural-language semantics and pragmatics, while also explaining how an agent who is unable to cognize propositions can know or believe them, and how sophisticated agents acquire the concept and believe things about them by monitoring their own cognitions. Finally, an account is given of what it is for a sentence to mean that p in a language that doesn’t require having thoughts about p or L. Nevertheless, semantics isn’t psychology; agents with different psychologies can speak semantically identical languages, while those with the same purely internal states (embedded in similar immediate environments) can speak different languages. Cognitive semantics can be realist and naturalistic without being a branch of psychology.
Section 1 is an extended commentary on Edward Sapir’s formulation nearly a century ago of what he considered the two most fundamental properties of human language, first that each one is a formally complete system of reference to experience and second that each one is formally distinct from every other. Section 2 considers some aspects of the development of these formulations, noting that they have been considered separately and not integrated as fully fleshed out systems of reference, as Sapir envisioned. Section 3 examines more closely what such an integration looks like in a case involving simple arithmetic. Section 4 begins with a brief review of the accomplishments of Greco-Roman logic and more recent developments in the theory of logic, leading to a consideration of what may be needed to fulfill Sapir’s program. Section 5 summarizes some of my own recent research on extending first-order logic by replacing the unordered set of individuals with a specific ordering of a set of sets of individuals that is isomorphic to an ordering of sets of sets of numbers that contain no pairs of divisible numbers, which was investigated by Richard Dedekind shortly before the turn of the twentieth century.
Undoubtedly, F. de Saussure is among the founders of modern linguistics and ‘semiology’/semiotics – though, strangely enough, his success mainly resulted from a book he did not write in the strict sense. Since the publication of the Cours, edited by his disciples in 1916, his notion of ‘arbitrariness’ has become one of the main canonical catchwords of linguistics that are hardly ever called into question. After a reconstruction of the Saussurean concept of the linguistic sign and its arbitrariness the attempt is made to work out its shortcomings and partly correct them. It is argued that instead of arbitrariness motivation must be taken as the central semiotic feature of linguistic signs which, after their formation, may be subject to a continuous historical process of increasing arbitrarization, possibly leading to complete arbitrariness in the end. In the final section the question of whether and to what extent the Geneva linguist may be considered a realist is tackled.
Undoubtedly, F. de Saussure is among the founders of modern linguistics and ‘semiology’/semiotics – though, strangely enough, his success mainly resulted from a book he did not write in the strict sense. Since the publication of the Cours, edited by his disciples in 1916, his notion of ‘arbitrariness’ has become one of the main canonical catchwords of linguistics that are hardly ever called into question. After a reconstruction of the Saussurean concept of the linguistic sign and its arbitrariness the attempt is made to work out its shortcomings and partly correct them. It is argued that instead of arbitrariness motivation must be taken as the central semiotic feature of linguistic signs which, after their formation, may be subject to a continuous historical process of increasing arbitrarization, possibly leading to complete arbitrariness in the end. In the final section the question of whether and to what extent the Geneva linguist may be considered a realist is tackled.