Abstract
The point of departure is that the traditional linguistic «components» or subsystems like phonology, semantics, lexicon and grammar are based on the linguistic sign (or symbol), and the article discusses the relation between these components: Are they independent and supplementary, or are some systems part of others? And what is their relation to the sign and its two parts, expression and content (meaning)? First, the linguistic sign is presented and discussed, and defined 1) as a general unit with any expression, not limited to speech, 2) non-mentalistically, in a manner compatible with a usage-based approach to language, namely as a social or conventional product that also comprises physical entities like sounds or letters. Then some problems concerning the relation between the sign and the subsystems are pointed out and discussed in the following sections. The main conclusions are: 1) Lexicon and grammar include (e.g. phonological) expressions, phonology deals with the expression system. 2) There is no semantic system – meaning is part of lexicon and grammar. 3) Grammar is not «autonomous», but consists of (complex) signs and includes meaning. 4) Lexicon and grammar are neither expression nor content, but represent the sign or lexicogrammatical level. 5) Languages and texts are not signs and do not consist of expression and content, but of subsystems and parts. 6) Words are the basic part of language, presupposed by grammar.
Keywords
Sign, Expression, Content, Phonology, Semantics, Lexicon, Grammar
1. Introduction
By subsystems I mean what is usually called ‘components’ or ‘modules’, i.e. phonology, morphology, syntax, semantics and the lexicon, which are the generally recognized components of spoken language. In written language we have graphology instead of phonology, but the same other parts. Morphology and syntax are usually referred to as grammar. The terms for the subsystems also denote the study of the systems, parts of linguistics and not language – e.g., phonology denotes both the phonological system and the study of it – but it is the systems themselves which are the object here.
If languages are sign systems, the sign must be the base for the division of language in subsystems. I will therefore start with a short definition of the linguistic sign. Then I will point out some problems concerning the relation between the subsystems and the sign and try to solve each of the problems. What is the relation between these subsystems? Are they all independent and supplementary, or are some systems part of others? And what is their relation to the two parts of the sign, expression and content? Do they consist of expression or of content – or of both? Which “side” or “level” of language do they belong to? How many levels are there?
2. The Linguistic Sign
By the linguistic sign I mean the bilateral sign of Saussure and Langacker, by Langacker called symbol – not the sign concept of everyday language or the “semiotic/semiological” sign, which are related unilateral concepts. The bilateral sign consists of two necessary parts which are connected through convention: a signifying part, usually called expression (or phonological pole), and a signified part called content or meaning (or semantic pole).
According to Saussure
| [31] | Saussure, Ferdinand de (1967 [1916]). Cours de linguistique générale [A Course in General Linguistics]. Paris: Éditions Payot et Rivages. |
[31]
(p. 97 ff), both parts are
mental, respectively a
sound image (
image acoustique) and a
concept.
This also means that Saussure’s sign is restricted to
spoken language, which Saussure
| [31] | Saussure, Ferdinand de (1967 [1916]). Cours de linguistique générale [A Course in General Linguistics]. Paris: Éditions Payot et Rivages. |
[31]
(p. 45) like other structuralists regards as the only form of language. Also for Langacker
| [23] | Langacker, Ronald W. (1987). Foundations of Cognitive Grammar, Volume I: Theoretical Prerequisites. Stanford U.P. |
[23]
(p. 11) the sign is mental as a whole: Symbols
are said to consist of a semantic and a phonological “representation”. However, in Langacker (2013)
| [26] | Langacker, Ronald W. (2013). Essentials of Cognitive Grammar. Oxford: Oxford UP. |
[26]
(p. 15) the «phonological representation» is said to include
gestures and
written characters. Langacker’s sign is thus wider than Saussure’s, and includes also written and gestural signs.
I shall follow Langacker on this point, as it is obvious that we have to do with sign systems also in writing and gestural (or «sign») language. Signs can have any expression as long as there is a definite content as well as a definite expression – also traffic lights or drawings on toilet doors. The most important sign systems are spoken and written language – usually called “verbal language” – which are to a high degree parallel systems, most obviously in alphabetic writing, where also the expressions and the expression systems are parallel. Here, I use verbal signs, mostly written ones, as examples.
However, both Saussure’s sign and Langacker’s symbol are «mentalistic» and place signs and language in the
individual minds and not in
speech communities – although that is exactly where Saussure places signs and language elsewhere, e.g. by saying that signs are “social by nature”
| [31] | Saussure, Ferdinand de (1967 [1916]). Cours de linguistique générale [A Course in General Linguistics]. Paris: Éditions Payot et Rivages. |
[31]
(p. 34). This precludes that signs and languages can consist of
physical objects like the sounds or letters that occur in usage and texts. In my view, this is not compatible with a
usage-based view of language, where usage and language are supposed to be of
the same kind, as Langacker
| [25] | Langacker, Ronald W. (2010). How not to disagree. The emergence of structure from usage. In Kasper Boye and Elisabeth Engberg-Pedersen (eds.): Language Usage and Language Structure, 107–143. Berlin: De Gruyter Mouton. |
[25]
(p. 109) points out: «…structure [...] is not independent of usage or radically different in nature. Rather, structure emerges from usage, is immanent in usage, and is influenced by usage on an ongoing basis». As usage and texts comprise physical units like sounds or letters, these units must also exist in the language, as general
types instead of individual
tokens (exemplars). Else language and usage would be radically different.
In other words, usage – the regular or conventional use of language – consists of tokens of the types in the language, else it is ungrammatical. Phonology does not deal with sound images, but with sound types like the famous «thick /l/» of parts of Norway and Sweden. It is not the image or mental representation of the sound type that is part of these dialects, but the sound type (an apico-postalveolar flap) itself. But to be able to pronounce it, one must have a “representation” or mental image of the sound type (and expressions where it occurs, i.e. words with “thick /l/”), built from previous encounters with the sound (and the words). To use a language, one must know it and be competent in using it. But that is knowledge of and competence in language, not language: A representation presupposes something that is represented, e.g. speech sounds or other linguistics units, and knowledge of language presupposes language. Mentalism confuses these two objects, or sees only one of them.
3. Some Problems with the Subsystems
It should be obvious that the linguistic components must have something to do with expression and content and therefore with signs, since phonology can be said to study the expression of spoken signs while semantics studies the content of the signs. But expression and content seem relevant also in the other subsystems. As to the lexicon, it is generally recognized that words have both an expression (a pronounciation or a spelling) and a meaning (which is described in dictionaries). Does the same go for grammatical units? And what about language as a whole and texts, the product of the sender’s usage: Do they consist of expression and content as well? There are five problems involved here:
1) There seems to be agreement that the sound system is an independent system. At the same time, sounds are relevant also in lexicon and grammar, since grammatical units consist of either morphemes or word forms and therefore have a pronunciation. Should we distinguish between two aspects of speech sounds, both as an independent system and as a part of lexicon and grammar?
2) Is there a corresponding semantic system, a parallel to the sound system? Does meaning constitute an independent system or is it just a part of lexicology and grammar?
3) What is the relation between semantics and grammar? Few issues have been discussed as much as the place of meaning (if any) in grammar, especially syntax: Are syntax and semantics two independent systems, such that syntax is “autonomous”, independent of meaning, or do grammatical units include meaning?
4) Saussure’s definition of the sign is usually understood to mean that also language as a whole has an “expression side” and a “content side” (or “level”). If so, it seems clear that phonology belongs to the expression side and semantics to the content side. But on what side should we place lexicon and grammar? Are words and grammatical units expression units or content units – or none of them?
5) And do languages and texts really consist of expression and content?
In the following I discuss each problem in this order.
4. Expressions and Expression Systems
Most sign systems, such as speech, alphabetical writing and gestural («sign») language, have «double articulation»
| [28] | Martinet, André (1965). La linguistique synchronique [Synchronic Linguistics]. Paris: PUF. |
[28]
(p. 2) or «duality of patterning»
(p. 90), which means that the expression normally is «articulated» or
complex, consisting of meaningless
expression units (occasionally one, as in the English article allomorph
a) selected from an
expression system: in spoken language
syllables and
speech sounds forming a
sound system, in writing
letters forming an
alphabet. An example from Langacker
| [23] | Langacker, Ronald W. (1987). Foundations of Cognitive Grammar, Volume I: Theoretical Prerequisites. Stanford U.P. |
[23]
(p. 298) is the spoken word form written
picnics, consisting
phonologically of the
syllables /pik/ and /niks/, which can be further analyzed in phonemes, but
grammatically of the
stem /piknik-/ and the
inflectional suffix /-s/, which are
signs with a
meaning. Syllables, sounds and letters have no meaning by themselves, but can combine according to certain rules to form
expressions with meanings, e.g. /pik-nik/ or /p-i-t/. Thereby they also distinguish (the expression of) different signs from each other, e.g. /pit/ from /bit/ or /pin/ (distinctive function).
Double articulation means that in both spoken language and alphabetical written language, we must distinguish between the expression system – the sound system or the alphabet – that governs the expression in all signs, and individual sign expressions, i.e. pronunciations or spellings, which consist of selected units from the expression system. For example, we should distinguish between English a as a general expression unit (phoneme or letter) and as a specific expression with a meaning, namely the indefinite article. In most words, a is just a sound or letter; in the article it’s something more – a meaningful sound or letter, functioning as a complete sign expression by itself.
The phoneme is the subject of phonology, which studies all sign expressions, determining which sounds occur in the language and their distribution in syllables and words (e.g. that /h/ only occurs word-initially in English). The article is the subject of lexicology and morphology, which describe specific words and word classes such as articles, both expression and content, and is also relevant in syntax as an important element in noun phrases. Variation of expression in inflection (allomorphy), e.g. syncopated forms of Norwegian adjectives like naken (naked), definite or plural nakn-e (conditioned by the following vowel), is often called “morpho-phonology”, but must be part of morphology since it concerns the inflection of certain adjectives (and nouns) and not the sound system. So there is phonology in morphology too – and in syntax and lexicology. All these subsystems deal with signs, which necessarily have an expression, in spoken language a pronunciation.
5. The Place of Meaning
Also the content can often be analyzed in content units or semantic factors, i.e. minimal elements of meaning. For example, the meaning of man can be divided in the factors ‘person’, ‘adult’ (compare boy) and ‘male’ (compare woman), and the meaning of the indefinite plural suffix -er in Norwegian gutt-er (boys) can analyzed in the meanings ‘plural’ and ‘presumed unknown to the receiver’ (unlike in English, the form is opposed to the definite plural gutt-ene, the boys). Is there a finite system of such minimal content units that in various combinations constitute the content of all minimal signs (morphemes) – a parallel to the sound system or the alphabet?
Many linguists have taken for granted that content and expression are parallel, and have assumed a semantic system corresponding to the sound system, usually without further description. One example is Hockett
| [18] | Hockett, Charles F. (1958). A Course in Modern Linguistics. New York: MacMillan. |
[18]
(p. 138). Another is Wierzbicka
| [36] | Wierzbicka, Anna (1998). The semantics of English causative constructions in a universal-typological perspective. In Michael Tomasello (ed.): The New Psychology of Language. Cognitive and Functional Approaches to Language Structure, 113–153. Mahwah, New Jersey: Lawrence Erlbaum Associates. |
[36]
, who posits an innate and universal «natural semantic metalanguage» with around 60 «conceptual primitives» such as ‘negation’ and ‘good’. According to Aitchison
| [1] | Aitchison, Jean (2003). Words in the Mind. An Introduction to the Mental Lexicon. Oxford: Blackwell. |
[1]
(p. 80–83), a general semantic system has not been demonstrated. At best, there are
limited systems of semantic factors within «semantic fields», small groups of semantically related words such as hyponyms and antonyms, kinship terms or personal pronouns, but no
general system that comprises
all words.
If so, we have no parallel to the sound system or the alphabet on the content side, only a parallel to the pronunciation or spelling of individual signs, and semantics is not on par with phonology. Meaning does not constitute a separate system, but must be a part of the meaningful components of language, namely lexicon and grammar. As to the expression, it’s not quite as straightforward: There, we have to assume both specific expressions in lexicon and grammar – pronunciations or spellings – and a general phonological or graphological system that determines possible pronunciations or spellings in the actual language.
Thus, the two parts of the sign are not parallel in spoken language and alphabetical written language: The expression is governed by a system, the content is not. Our ancestors managed to «articulate» the expression, but not the content, likely because we can only produce a limited number of sounds, making it necessary to combine sounds to complex expressions to achieve a sufficient number of different sign expressions. Something similar is not necessary, probably not even possible, on the content side, since there is an unlimited number of concepts, often not clearly delimited.
6. Grammar and Meaning
One consequence of regarding meaning as an independent system is that meaning and grammar, especially syntax, by many linguists are regarded as two independent systems, e.g. in generative grammar
| [37] | Åfarli, Tor A. & Laila Sakshaug (2006). Grammatikk: Syntaks og morfologi med norsk i sentrum [Grammar: Syntax and Morphology with Norwegian in Center]. Oslo: Det Norske Samlaget. |
[37]
(p. 15). Also Jackson
| [21] | Jackson, Howard (1995). Grammar and Meaning. A Semantic Approach to English Grammar. London: Longman. |
[21]
(p. 4) assumes «...a semantic system, concerned with the meaning relations between elements of a construction, and a grammatical system, concerned with the grammatical relations between elements of a construction». However, «...syntax and semantics interweave in the structure of language» and «...a ‘grammatical’ description really has to take account of both aspects”
| [21] | Jackson, Howard (1995). Grammar and Meaning. A Semantic Approach to English Grammar. London: Longman. |
[21]
(p. 165–166). Then how can they be different systems? With such a general theory, it’s not surprising that we often get contradictory statements and practices (more examples below). A contradiction between theory and practice usually indicates that the theory is wrong.
The separation of semantics and grammar or more specifically syntax means that grammar is seen as
meaningless, and excludes the possibility that grammatical relations such as the subject and object functions are meaningful. Faarlund
| [8] | Faarlund, Jan Terje (2001). The notion of oblique subject and its status in the history of Icelandic. In Jan Terje Faarlund (ed.): Grammatical Relations in Change, 99–135. Amsterdam: John Benjamins. |
[8]
(p. 103) explicitly denies that grammatical categories have meaning: «Considering subjects as a morphosyntactic category, I will disregard semantic criteria from the outset». So what is the purpose of syntax, then?
According to Sundman
| [34] | Sundman, Marketta (1987). Subjektval och diates i svenskan [Subject choice and Diathesis in Swedish]. Åbo: Åbo Akademis Förlag. |
[34]
(p. 5), syntax is «...a rule system that relates the meaning (content) of a sentence to its form (expression)», and the sentence constituents she places (p. 8) «between form and meaning». Morphemes, however, she refers to (p. 17) as «meaningful elements». And since sentences
consist of morphemes, they must also be meaningful. Faarlund
| [9] | Faarlund, Jan Terje (2005). Revolusjon i lingvistikken. Noam Chomskys språkteori [Revolution in Linguistics. Noam Chomsky’s Theory of Language]. Oslo: Det Norske Samlaget. |
[9]
(p. 41) describes syntax as «a mediator between expression and content», and Teleman et al.
| [35] | Teleman, Ulf, Staffan Hellberg & Erik Andersson (1999). Svenska Akademiens Grammatik [The Grammar of the Swedish Academy], vol. 1. Stockholm: Svenska Akademien. |
[35]
(p. 41) describe grammar as «a bridge between the expression side and the content side of language», and place (p. 31) vocabulary, grammar and text study
between phonology/graphology and semantics, suggesting that the first three parts do not
include the latter two but are
supplementary to them. However, the authors inform (p. 34) that meaning is described «…in connection with the specific grammatical categories and structures, whereas semantics is not given a unified account». A sensible and common procedure, but hardly in agreement with the account on page 41 and 31. The same goes for the statement (p. 34) that grammatical categories like word classes and sentence constituents «…normally have a semantic motivation, i.e. they have a general meaning that characterizes the category». If grammatical units
have meaning, meaning must be
part of grammar.
Meaningful or
having meaning is something else than
relating or
mediating meaning and expression, because in the last case there’s something
between the two.
According to this theory, expression and content are not
directly connected, at least in grammar: Between them lies an entirely abstract system – syntax or grammar – that does not include either of them. Compare Gil
| [11] | Gil, David (2000). Syntactic categories, cross-linguistic variation and universal grammar. In Petra M. Vogel and Bernhard Comrie (eds.): Approaches to the Typology of Word Classes, 173–216. Berlin: Mouton de Gruyter. |
[11]
(p. 176), who says that there is «...a crucial difference between human language and most other semiotic systems, such as, for example, traffic lights [...], where
red means ‘stop’ and
green means ‘go’.» Namely that “...the relationship between sounds and meaning is not direct [...]. Rather, the relationship is mediated by various intervening entities: the linguistic forms [...] which constitute the basic building blocks of linguistic analysis”, i.e. lexical and grammatical units. From this follows “the autonomy of syntax”.
Such abstract units certainly don’t occur in usage, and in a usage-based account not in the language that is used either. All grammatical units, from morphemes to sentences, obviously have an expression, e.g. sounds or letters, else we could not observe and understand them. This goes for the system (language) as well as for the use of it. Using a system shows what the system consists of. And if morphemes, unlike phonemes, have meaning, then also word forms have meaning, because they consist of morphemes. And then also syntagms and sentences must have meaning, because they consist of word forms. Otherwise, linguistic communication would be inexplicable.
The view of grammar (and lexicon) as an intermediary between expression and content doesn’t explain where meaning comes from when language is used. How do we know what cat means when we hear or see the word, e.g. in a sentence? Meaning must come from the expression (c-a-t) plus the conventions that connect expression and content into signs of various types. So expression and content are directly connected, just like in the traffic lights, and between them there is nothing. An abstract «bridge» between expression and content we can safely discard with Ockham’s razor. Grammar doesn’t connect expression and content because that is already done in the minimal units of grammar, the morphemes. Instead, grammar combines signs of various types, from morphemes to clauses, and grammatical units consist of both expression and content.
That words (lexemes) have meaning, is assumed by all dictionaries, and inflected forms have grammatical meanings – content factors like ‘plural’, ‘past tense’ etc. – in addition to the lexical meaning. It is also commonly assumed that sentences have meaning, e.g. that interrogative sentences by themselves express a question (but can be
used «rhetorically» to express other speech acts). Then how can they be situated
between content and expression? Grammar
consists of word forms, either by themselves (morphology) or combined (syntax), and includes the meaning (and expression) of these forms, in addition to relational or functional meanings between constituents, like ‘subject’ and ‘object’ or ‘head’ and ‘modifier’. For example,
syntactic constructions are assumed to have a meaning that consists of the semantic relation between the constituents
(p. 78–79).
As to syntactic functions like subject and object, the sentences It (e.g. a dog) can see you and You can see it mean different things, although they consist of the same word forms: In the first it refers to the agent and you to the patient (or goal), in the last it’s the other way round. The reason must be that the syntactic and semantic relation between the word forms is different: In the first sentence it is the subject and you the object (which we can see from their positions), in the last it’s the other way round. And in the active voice subjects refer to agents while objects refer to patients. So, in active sentences the subject and the object must have the meanings of ‘agent’ and ‘patient’, respectively, and to understand such sentences it is necessary to realize this, i.e. to analyze the sentence syntactically, at least intuitively.
Therefore, it is meaningless to ask whether something is grammatical
or semantic, since grammar
includes meaning and things consequently can be
both – e.g the meaning of an inflectional ending, a word class, a syntactic function or a sentence type. And therefore, grammatical terminology is full of semantic descriptions: We talk about
possessives, adverbials of
manner, interrogative sentences,
causal clauses, etc., and many linguists have underlined the close relationship between grammar and meaning. Langacker states that «…grammar is symbolic in nature, consisting in the conventional symbolization of semantic structure»
| [23] | Langacker, Ronald W. (1987). Foundations of Cognitive Grammar, Volume I: Theoretical Prerequisites. Stanford U.P. |
[23]
(p. 2), and that «...all valid grammatical constructs have some kind of conceptual import»
| [24] | Langacker, Ronald W. (1991). Foundations of Cognitive Grammar. Volume 2: Descriptive Application. Stanford: Stanford UP. |
[24]
(p. 282). According to Matthews
(p. 53), «...syntactic relations are in part semantic relations». Dixon
| [7] | Dixon, R. M. W. (2010). Basic Linguistic Theory. Volume 1: Methodology. Oxford: Oxford UP. |
[7]
(p. 28) says that «There is a semantic basis to each part of every grammar», and Bouchard
(p. 247) that «Every syntactic combination reflects a semantic combination». Regarding
word classes, Haugen
| [15] | Haugen, Tor Arne (2012). Polyvalent Adjectives in Norwegian: Aspects of their Semantics and Complementation Patterns. Doctoral dissertation, The University of Oslo. |
[15]
(p. 25) states that «...it is hard to imagine why categories like noun and verb should emerge in the first place if they did not have a semantic basis».
Thus, meaning is fundamental in grammar and can largely explain grammatical properties, both morphological and syntactic. Why aren’t nouns like milk or silver normally used in the plural, and why can’t they be used with numerals or the indefinite article? Because they denote an unbounded substance that occurs in a certain quantity and not in a certain number, and therefore cannot be counted. Why are adjectives inflected in degree and used with degree modifiers like more/most/very/too/as (elegant)? Because they denote qualities (or states) that usually exist in varying degrees. For the same reason, adjective phrases typically function as either modifiers or predicatives to a noun phrase and describe the referent of the noun. Why can normally only nominals function as subjects or objects? Because nominals denote entities that can be participants with specific roles in an action or a state. Why is give a three-place verb that in the active voice normally requires a subject, a direct object and an indirect object? Because it denotes an action involving three participants: a giver, a gift, and a recipient, denoted by each nominal. Ignoring such aspects makes grammar a mystery.
7. Lexicon and Grammar: The Sign Level of Language
Saussure’s definition of the sign is usually taken to mean that also
language as a whole has two «levels» or «sides». For instance, Hjelmslev
| [17] | Hjelmslev, Louis (1973 [1947]). The basic structure of language. In Louis Hjelmslev: Essais linguistiques II [Linguistic Essays II] (Travaux du cercle linguistique de Copenhague XIV), 119–153. København: Nordisk sprog- og kulturforlag. |
[17]
(p. 126) calls language «a two-sided structure, involving
content and expression», and asserts (p. 127) that also a
text consists of expression and content, which must be analyzed separately. And Harder
| [14] | Harder, Peter (1996). Linguistic structure in a functional grammar. In Elisabeth Engberg-Pedersen, Michael Fortescue, Peter Harder, Lars Heltoft & Lisbeth Falster Jakobsen (eds.): Content, Expression and Structure. Studies in Danish Functional Grammar, 423–452. Amsterdam: John Benjamins. |
[14]
(p. 445) states that «the most basic property of language» is «its division into a content and an expression side». If these two «sides» are all there is, all linguistic units must belong to either one side or the other, i.e., they must be either
expression units or
content units. It seems obvious that
meanings are content units and
sounds or
written characters expression units, but what about
lexical and
grammatical units, which have
both expression
and content?
The answer is that grammar, especially syntax (the lexicon is usually not mentioned), is placed variously on either side. An example of placing grammar on the
content side is Kofoed
| [22] | Kofoed, H. A. (1971). Sprog og sprogvidenskab. En oversigt over synspunkter og metoder i moderne, strukturel lingvistik [Language and Linguistics. A Survey of Point of Views and Methods in Modern Structural Linguistics], 2nd edn. Bergen: Universitetsforlaget. |
[22]
(p. 15), who states that «content study» is the study of «morphemics and semantics». But if morphology (and perhaps grammar altogether) is placed on the content side along with semantics, then what is the difference between morphology (or grammar) and semantics? Another example is Allerton
| [2] | Allerton, D. J. (1979). Essentials of grammatical theory. A concensus view of syntax and morphology. London: Routledge & Kegan Paul. |
[2]
(p. 42): «...language is a two-level semiotic system, being analyzable separately for content and expression units: it has a set of meaningful content units (signs) and a set of meaningless expression units (figurae)». And the study of «meaningful content units» is divided into «...grammar (or syntax), lexis (or vocabulary), and semantics». But if signs are «content units», what kind of units are
meanings? We must distinguish not only signs and
expression units, but also signs and
content units (meanings), which are
parts of the signs. And grammar, lexicology, and semantics are not aligned: Only the first two can be said to consist of «meaningful units» (signs) and thereby
comprise meaning.
Others place grammar or syntax on the
expression side or on
both sides separately, with the somewhat surprising result that we get
two syntaxes. An example is the Swedish reference grammar
| [35] | Teleman, Ulf, Staffan Hellberg & Erik Andersson (1999). Svenska Akademiens Grammatik [The Grammar of the Swedish Academy], vol. 1. Stockholm: Svenska Akademien. |
[35]
, which divides grammar into the «meaning side» (p. 41) and the «expression side» (p. 46), and where the grammatical structure is discussed under the expression side. Another example is the Danish reference grammar
| [12] | Hansen, Erik, & Lars Heltoft (2011). Grammatik over det Danske Sprog [A Grammar of the Danish Language]. København: Det Danske Sprog- og Litteraturselskab. |
[12]
, which divides (p. 28) syntax into an «expression syntax» and a «content syntax» instead of a single syntax consisting of (complex) signs. And by the expression side of sentences the authors understand (p. 101) «...the traditional construction of the sentence in syntagms [probably meaning ‘constituents’] (units that can be moved and substituted)», which means that the sentence constituents are considered to be expression units.
But if grammatical units are regarded as expression units, then what is the difference between
morphemes and
phonemes? Unlike sounds or letters, grammatical units, from morphemes to sentences, also have a
content. How can they be placed on the expression side? This way the term
expression is used in two quite different meanings: partly about
grammatical and
lexical units, which are meaningful, and partly about
phonological or
graphological units, which are not. When Sundman compares sentence constituents to
phonemes and suggests that «...also sentence constituents should have a
meaning-distinguishing function»
| [33] | Sundman, Marketta (1980). Existentialkonstruktionen i svenskan [The Existential Construction in Swedish]. Åbo: Publications of the Research Institute of the Åbo Akademi Foundation. |
[33]
(p. 253–254, my emphasis), she overlooks a basic difference between sentence constituents and phonemes, namely that sentence constituents are meaning
ful (signs), not meaning-
distinguishing (expression units).
Dividing language into an expression side and a content side means that there is no place for
signs as wholes, i.e. for neither
lexicon nor
grammar. Fabricius-Hansen
| [10] | Fabricius-Hansen, Cathrine (1981). Emner og modeller i sprogbeskrivelsen [Subjects and models in the description of language]. Norskrift 33, 43–89. http://urn.nb.no/URN:NBN:no-39783 |
[10]
(p. 52) points this out for syntax in glossematics: «When the content and expression planes are described separately [...], there is really no place for syntax, if one understands syntactic relations as relations between entire signs and not just content or expression units». And this is just how we should understand syntactic relations, as Harder
| [13] | Harder, Peter (1990). Tense, semantics and layered syntax. In Jan Nuyts, A. M Bolkestein and Co Vet (eds.): Layers and Levels of Representation in Language Theory. A Functional View, 139–163. Amsterdam/Philadelphia: John Benjamins. |
[13]
(p. 141) states: «Syntax deals with the combination of linguistic items into more complex linguistic items. Since the items combined are signs, syntax deals with both expression and content at the same time». None the less, also Harder
| [13] | Harder, Peter (1990). Tense, semantics and layered syntax. In Jan Nuyts, A. M Bolkestein and Co Vet (eds.): Layers and Levels of Representation in Language Theory. A Functional View, 139–163. Amsterdam/Philadelphia: John Benjamins. |
[13]
(p. 142) distinguishes between «content syntax» («how content elements are combined into larger wholes») and «expression syntax» («the rules for combining expression elements into complex expressions») instead of one syntax that combines signs. A theory of language that has no place for syntax or has to operate with two syntaxes, has a major shortcoming.
Now, of course expression and content can be considered separately, abstracted from its companion. We may be interested in the pronounciation or spelling of a word or in its meaning, or in the meaning of a sentence or a whole text. But we should not forget that neither of them occurs in isolation, but always together in signs. So if we call the expression and the content «sides» or «levels», we must also recognize a third and basic level: the sign or lexicogrammatical level, where the units are signs of various types, from morphemes to sentences. As Langacker’s «content requirement» [23] (p. 53–54) underlines, the linguistic sign means that there are three kinds of linguistic units: expression units, such as sounds or letters; content units, i.e. meanings or parts of meanings (semantic factors), and signs, which consist of the first two. And there must be a level for each of them.
8. Neither Languages nor Texts Consist of Expression and Content
The fact that signs consist of expression and content doesn’t mean that also languages and texts do, because neither languages nor texts are signs. Languages are sign systems and consist of signs, not expression and content. More precisely of signs organized in subsystems such as lexicon and grammar, which makes it possible to combine minimal signs into complex ones and construct an unlimited number of signs. In addition, there usually is an independent expression system like the sound system or the alphabet. Only minimal signs, i.e. morphemes, consist of expression and content. Complex signs like complex words (derivations or compounds), complex word forms (e.g. inflections) and syntagms like sentences or prepositional phrases consist of smaller signs, often hierarchically, and must be analyzed in constituents at multiple levels, as is usually done in grammatical analysis.
Texts cannot be signs because they are individual products and follow only “regulative” or functional conventions – pragmatic, stilistic, rhetorical etc. – that govern the use of language, how to be eloquent or a good communicator. They cannot be right or wrong, only good or bad (e.g. incomprehensible or incoherent), and are not what we learn when we learn a language. It would simply be wrong, and probably impossible, to analyze a text, e.g. this one, in expression and content. Texts consist of parts defined by their content, most clearly marked in certain written texts: chapters, sections and subsections, paragraphs, periods and utterances – the minimal text and the maximal unit of grammar.
However, each utterance in a text (mostly sentences) could be called a sign – usually a complex sign, consisting of smaller signs. Though also utterances are individual products, they follow “constitutive” or structural linguistic conventions – graphological or phonological, lexical and grammatical – which may be broken (e.g. by spelling mistakes), resulting in a faulty or “ungrammatical” utterance. So utterances could be regarded as the maximal sign, and must be analyzed hierarchically in smaller signs (constituents), not in expression and content.
What the readers can see in written utterances such as these, is the expression – letters and other characters. Using a language as sender means producing the expression of the chosen signs (and choosing the utterance type, e.g. a statement). But knowing the sign system that is used – here: standard written English – readers recognize the expression units as expressions of signs that they know, connected to meanings: morphemes and word forms separated by spaces (e.g. see or sign-s), and sentences or other utterances marked by punctuation, and can hopefully understand both each utterance and the semantic connection between them, i.e. the text as a whole.
9. Conclusion
The conclusion is that spoken language, alphabetical written language and gestural language consist of two main systems: a
sign system and an
expression system – the «double articulation». The sign system consists of a
lexicon that, according to Dixon
| [7] | Dixon, R. M. W. (2010). Basic Linguistic Theory. Volume 1: Methodology. Oxford: Oxford UP. |
[7]
(p. 22), «classifies things», and a
grammatical system that «organizes things». The lexicon consists of two main classes: an open class of
lexical words and a closed class of
grammatical words, both of which can be subdivided. For instance, nouns can be divided into
mass nouns and
countable nouns, and the latter into
proper nouns naming individuals and
common nouns naming classes.
In verbal language the «things» organized by grammar are words and word forms, and grammar can be divided into two systems: a morphological system that classifies words (in word classes), makes complex words (stems) by compounding and derivation and in many languages inflected word forms by inflection, and a syntactic system for combining word forms hierarchically in syntagms of various types (constructions) such as prepositional phrases or sentences. So also grammar classifies, e.g. words or constructions. However, lexicon and grammar overlap since there are grammatical words and word classes are regarded as grammatical organization and treated in morphology. Moreover, fixed expressions with a syntactic structure but a holistic meaning, e.g. figure out or put up with, must be regarded as both grammatical and lexical units.
In recent linguistic theory grammar, especially (in generative grammar, exclusively) syntax, has received most attention. But among ordinary speakers, it’s rather
words that matter, as Dixon
| [7] | Dixon, R. M. W. (2010). Basic Linguistic Theory. Volume 1: Methodology. Oxford: Oxford UP. |
[7]
(p. 20) points out: «All over the world, speakers conceive of a language as consisting of its vocabulary, with little regard paid to grammar». The speakers may have a point here. Since both morphology and syntax concern word forms, words can be considered the basic part of language and the
prerequisite for grammar: Plural markers presuppose countable nouns, auxiliary verbs presuppose lexical verbs, etc. According to Boye and Harder
(p. 6–7), grammatical units have «…an ancillary communicative purpose as secondary or background elements». They presuppose the words they organize.
Additionally, grammatical units arise from lexical units through grammaticalization of lexical words and syntagms (like English going to + verb) and the combination of word forms into constructions. So language must have started with words, and grammar came later, when there were enough words and words of different classes, at least nouns and verbs, which could be combined meaningfully. We can imagine a language with limited grammar (like pidgin languages), but not one without a lexicon. Such a language would also lack grammar because there would be nothing for grammar to organize. The importance of words in communication is shown by the difficulties we all experience, especially in old age, with «finding the right word» to express what we want to say. Grammar is seldom a problem for healthy people, but words often are. And without words, grammar is useless.
Author Contributions
Eric Papazian is the sole author. The author read and approved the final manuscript.
Conflicts of Interest
The author declares no conflicts of interest.
References
| [1] |
Aitchison, Jean (2003). Words in the Mind. An Introduction to the Mental Lexicon. Oxford: Blackwell.
|
| [2] |
Allerton, D. J. (1979). Essentials of grammatical theory. A concensus view of syntax and morphology. London: Routledge & Kegan Paul.
|
| [3] |
Bouchard, Denis (2013). The Nature and Origin of Language. Oxford: Oxford UP.
https://doi.org/10.1093/acprof:oso/9780199681624.001.0001
|
| [4] |
Boye, Kasper, & Peter Harder (2012). A usage-based theory of grammatical status and grammaticalization. Language 88 (1), 1–44.
https://doi.org.ezproxy.uio.no/10.1353/lan.2012.0020
|
| [5] |
Bybee, Joan L. (2010). Language, Usage and Cognition. Cambridge: Cambridge UP.
https://doi.org.ezproxy.uio.no/10.1017/CBO9780511750526
|
| [6] |
Crystal, David (2007). How Language Works. New York: Avery.
|
| [7] |
Dixon, R. M. W. (2010). Basic Linguistic Theory. Volume 1: Methodology. Oxford: Oxford UP.
|
| [8] |
Faarlund, Jan Terje (2001). The notion of oblique subject and its status in the history of Icelandic. In Jan Terje Faarlund (ed.): Grammatical Relations in Change, 99–135. Amsterdam: John Benjamins.
|
| [9] |
Faarlund, Jan Terje (2005). Revolusjon i lingvistikken. Noam Chomskys språkteori [Revolution in Linguistics. Noam Chomsky’s Theory of Language]. Oslo: Det Norske Samlaget.
|
| [10] |
Fabricius-Hansen, Cathrine (1981). Emner og modeller i sprogbeskrivelsen [Subjects and models in the description of language]. Norskrift 33, 43–89.
http://urn.nb.no/URN:NBN:no-39783
|
| [11] |
Gil, David (2000). Syntactic categories, cross-linguistic variation and universal grammar. In Petra M. Vogel and Bernhard Comrie (eds.): Approaches to the Typology of Word Classes, 173–216. Berlin: Mouton de Gruyter.
|
| [12] |
Hansen, Erik, & Lars Heltoft (2011). Grammatik over det Danske Sprog [A Grammar of the Danish Language]. København: Det Danske Sprog- og Litteraturselskab.
|
| [13] |
Harder, Peter (1990). Tense, semantics and layered syntax. In Jan Nuyts, A. M Bolkestein and Co Vet (eds.): Layers and Levels of Representation in Language Theory. A Functional View, 139–163. Amsterdam/Philadelphia: John Benjamins.
|
| [14] |
Harder, Peter (1996). Linguistic structure in a functional grammar. In Elisabeth Engberg-Pedersen, Michael Fortescue, Peter Harder, Lars Heltoft & Lisbeth Falster Jakobsen (eds.): Content, Expression and Structure. Studies in Danish Functional Grammar, 423–452. Amsterdam: John Benjamins.
|
| [15] |
Haugen, Tor Arne (2012). Polyvalent Adjectives in Norwegian: Aspects of their Semantics and Complementation Patterns. Doctoral dissertation, The University of Oslo.
|
| [16] |
Hjelmslev, Louis (1966 [1943]). Omkring sprogteoriens grundlæggelse [Prolegomena to a Theory of Language]. København: Akademisk forlag.
|
| [17] |
Hjelmslev, Louis (1973 [1947]). The basic structure of language. In Louis Hjelmslev: Essais linguistiques II [Linguistic Essays II] (Travaux du cercle linguistique de Copenhague XIV), 119–153. København: Nordisk sprog- og kulturforlag.
|
| [18] |
Hockett, Charles F. (1958). A Course in Modern Linguistics. New York: MacMillan.
|
| [19] |
Hockett, Charles F. (1960). The Origin of Speech. Scientific American 203 (3), 88–97.
https://www.jstor.org/stable/24940617
|
| [20] |
Hurford, James R. (2007). Language in the Light of Evolution, vol. 1: The Origins of Meaning. Oxford: Oxford UP.
|
| [21] |
Jackson, Howard (1995). Grammar and Meaning. A Semantic Approach to English Grammar. London: Longman.
|
| [22] |
Kofoed, H. A. (1971). Sprog og sprogvidenskab. En oversigt over synspunkter og metoder i moderne, strukturel lingvistik [Language and Linguistics. A Survey of Point of Views and Methods in Modern Structural Linguistics], 2nd edn. Bergen: Universitetsforlaget.
|
| [23] |
Langacker, Ronald W. (1987). Foundations of Cognitive Grammar, Volume I: Theoretical Prerequisites. Stanford U.P.
|
| [24] |
Langacker, Ronald W. (1991). Foundations of Cognitive Grammar. Volume 2: Descriptive Application. Stanford: Stanford UP.
|
| [25] |
Langacker, Ronald W. (2010). How not to disagree. The emergence of structure from usage. In Kasper Boye and Elisabeth Engberg-Pedersen (eds.): Language Usage and Language Structure, 107–143. Berlin: De Gruyter Mouton.
|
| [26] |
Langacker, Ronald W. (2013). Essentials of Cognitive Grammar. Oxford: Oxford UP.
|
| [27] |
Lyons, John (1981). Language and Linguistics. An Introduction. Cambridge UP.
|
| [28] |
Martinet, André (1965). La linguistique synchronique [Synchronic Linguistics]. Paris: PUF.
|
| [29] |
Matthews, Peter (1991). Morphology, 2nd edn. Cambridge: Cambridge UP.
https://hdl-handle-net.ezproxy.uio.no/2027/heb08448.0001.001
|
| [30] |
Matthews, Peter (1995). Syntax, semantics, pragmatics. In F. R. Palmer (ed.): Grammar and Meaning. Essays in honour of Sir John Lyons, 48–60. Cambridge: Cambridge UP.
https://doi-org.ezproxy.uio.no/10.1017/CBO9780511620638
|
| [31] |
Saussure, Ferdinand de (1967 [1916]). Cours de linguistique générale [A Course in General Linguistics]. Paris: Éditions Payot et Rivages.
|
| [32] |
Stokoe, William C. (1960). Sign Language Structure: an Outline of the Visual Communication Systems of the American Deaf. Buffalo, N. Y.
|
| [33] |
Sundman, Marketta (1980). Existentialkonstruktionen i svenskan [The Existential Construction in Swedish]. Åbo: Publications of the Research Institute of the Åbo Akademi Foundation.
|
| [34] |
Sundman, Marketta (1987). Subjektval och diates i svenskan [Subject choice and Diathesis in Swedish]. Åbo: Åbo Akademis Förlag.
|
| [35] |
Teleman, Ulf, Staffan Hellberg & Erik Andersson (1999). Svenska Akademiens Grammatik [The Grammar of the Swedish Academy], vol. 1. Stockholm: Svenska Akademien.
|
| [36] |
Wierzbicka, Anna (1998). The semantics of English causative constructions in a universal-typological perspective. In Michael Tomasello (ed.): The New Psychology of Language. Cognitive and Functional Approaches to Language Structure, 113–153. Mahwah, New Jersey: Lawrence Erlbaum Associates.
|
| [37] |
Åfarli, Tor A. & Laila Sakshaug (2006). Grammatikk: Syntaks og morfologi med norsk i sentrum [Grammar: Syntax and Morphology with Norwegian in Center]. Oslo: Det Norske Samlaget.
|
Cite This Article
-
APA Style
Papazian, E. (2025). The Linguistic Sign and the Subsystems of Language. International Journal of Language and Linguistics, 13(1), 11-18. https://doi.org/10.11648/j.ijll.20251301.13
Copy
|
Download
ACS Style
Papazian, E. The Linguistic Sign and the Subsystems of Language. Int. J. Lang. Linguist. 2025, 13(1), 11-18. doi: 10.11648/j.ijll.20251301.13
Copy
|
Download
AMA Style
Papazian E. The Linguistic Sign and the Subsystems of Language. Int J Lang Linguist. 2025;13(1):11-18. doi: 10.11648/j.ijll.20251301.13
Copy
|
Download
-
@article{10.11648/j.ijll.20251301.13,
author = {Eric Papazian},
title = {The Linguistic Sign and the Subsystems of Language},
journal = {International Journal of Language and Linguistics},
volume = {13},
number = {1},
pages = {11-18},
doi = {10.11648/j.ijll.20251301.13},
url = {https://doi.org/10.11648/j.ijll.20251301.13},
eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijll.20251301.13},
abstract = {The point of departure is that the traditional linguistic «components» or subsystems like phonology, semantics, lexicon and grammar are based on the linguistic sign (or symbol), and the article discusses the relation between these components: Are they independent and supplementary, or are some systems part of others? And what is their relation to the sign and its two parts, expression and content (meaning)? First, the linguistic sign is presented and discussed, and defined 1) as a general unit with any expression, not limited to speech, 2) non-mentalistically, in a manner compatible with a usage-based approach to language, namely as a social or conventional product that also comprises physical entities like sounds or letters. Then some problems concerning the relation between the sign and the subsystems are pointed out and discussed in the following sections. The main conclusions are: 1) Lexicon and grammar include (e.g. phonological) expressions, phonology deals with the expression system. 2) There is no semantic system – meaning is part of lexicon and grammar. 3) Grammar is not «autonomous», but consists of (complex) signs and includes meaning. 4) Lexicon and grammar are neither expression nor content, but represent the sign or lexicogrammatical level. 5) Languages and texts are not signs and do not consist of expression and content, but of subsystems and parts. 6) Words are the basic part of language, presupposed by grammar.},
year = {2025}
}
Copy
|
Download
-
TY - JOUR
T1 - The Linguistic Sign and the Subsystems of Language
AU - Eric Papazian
Y1 - 2025/01/22
PY - 2025
N1 - https://doi.org/10.11648/j.ijll.20251301.13
DO - 10.11648/j.ijll.20251301.13
T2 - International Journal of Language and Linguistics
JF - International Journal of Language and Linguistics
JO - International Journal of Language and Linguistics
SP - 11
EP - 18
PB - Science Publishing Group
SN - 2330-0221
UR - https://doi.org/10.11648/j.ijll.20251301.13
AB - The point of departure is that the traditional linguistic «components» or subsystems like phonology, semantics, lexicon and grammar are based on the linguistic sign (or symbol), and the article discusses the relation between these components: Are they independent and supplementary, or are some systems part of others? And what is their relation to the sign and its two parts, expression and content (meaning)? First, the linguistic sign is presented and discussed, and defined 1) as a general unit with any expression, not limited to speech, 2) non-mentalistically, in a manner compatible with a usage-based approach to language, namely as a social or conventional product that also comprises physical entities like sounds or letters. Then some problems concerning the relation between the sign and the subsystems are pointed out and discussed in the following sections. The main conclusions are: 1) Lexicon and grammar include (e.g. phonological) expressions, phonology deals with the expression system. 2) There is no semantic system – meaning is part of lexicon and grammar. 3) Grammar is not «autonomous», but consists of (complex) signs and includes meaning. 4) Lexicon and grammar are neither expression nor content, but represent the sign or lexicogrammatical level. 5) Languages and texts are not signs and do not consist of expression and content, but of subsystems and parts. 6) Words are the basic part of language, presupposed by grammar.
VL - 13
IS - 1
ER -
Copy
|
Download
Share This Article
Twitter
Linked In
Facebook
