Methodological basis
Both the conception of the functional domains of language and the conception of the hierarchical system of constructions are grounded in general comparative grammar. The form they take in the description of a particular language is essentially a specified subset of either general framework.
Since a language description obeys the categorical imperative, the theoretical framework which it applies does not result by pure deduction from a theory of language. Instead, it is also the result of inductive analysis and typological comparison.
The set of categories of grammatical constructions, including nominal, adjectival, adverbial etc. constructions, their internal conformation in terms of formatives, head-dependent syntagmas, coordinative syntagmas etc., the kinds of relations holding such syntagmas together, including relations of phora, government and modification, all of this is the result of empirical work which gathers data from many languages, analyzes these pursuing the heuristic question ‘what is the pattern underlying this (kind of) expression’, and systematizes them in terms of a framework of structural devices used by natural languages.
Likewise, the set of notions, including substantive and predicative notions, of associated operations of conversion among these, including hypostasis and predication, of relations, including participant roles, local relations, possessive relations etc., the set of modal operations, of illocutions and many more, all of this is the result of inductive work which analyzes the functions fulfilled by constructions, formatives and other structural means in any given language, pursuing the heuristic question ‘what is being done by these means in terms of cognition and communication’, compares these functions across languages, recognizes isofunctionality and systematizes the notional categories, relations and operations in terms of a framework of functional domains which embody what people do with language and which provide grammatical meanings.
The systematization applied to the findings of empirical work is, of course, theoretical work. This presupposes a theory of cognition and communication and a semiotic theory which delineates the form of a semiotic system used under the circumstances of the conditio humana. Some portions of the structural and of the functional framework are deducible from such theories, but most of the details are not. For these, the theory only provides a consistent conception, but not a logical necessity.
Both the framework of structural devices and the framework of functional domains are what have been called ‘maximum models’.1
- For the system of structural devices, this means that a set of them is universally available to structure linguistic units; but no language uses all of them, and every language integrates the devices used into a specific system of grammatical structure whose categories are like, but not the same as in other languages. For instance, the framework provides for infixes. Some languages have them, others lack them. Those which have them manifest variation. Some have infixes only in verb roots, others in morphemes of other categories. Some languages position them after the onset of the first syllable, others in front of the coda of its last syllable. Such differences may become definitional properties of language-specific categories. The grammar of a specific language is then an instantiation of the language-independent structural framework.
- For the system of functional domains, this means that, for each of these domains, every language devotes part of its grammar to a subset of the functions constituting it; and any function fulfilled by some structural device of some language finds its place in the framework. In this sense, the framework is complete. For any specific function, a language chooses whether to fulfill it by the language system, be it in the grammar or in the lexicon, or to leave its fulfillment to inference. For instance, the framework provides for the semantic role of the beneficiary, a pan-human notion used in the interpretation of a situation. It may be coded in the grammar, e.g. by an applicative construction indicating a benefactive function for its undergoer Y; it may be coded in the lexicon, e.g. by a locution like in favor of Y; or it may be left to inference, e.g. in a clause expressing that X repairs some property of Y's. The first of these cases may constitute a language-specific benefactive construction. The benefactive role figures in the functional framework, specifically in the functional domain of participation, because there are at least some languages which code it in their grammar. In this sense, the grammar of a language is a language-specific instantion of the language-independent functional framework.
Theoretical basis
The present website on linguistic description operates under the more general heading ‘linguistic methodology’. It does not pretend to expound a theory of language. The semasiological and onomasiological frameworks proposed are meant to fulfill a practical purpose, viz. to offer a template for structuring a comprehensive and consistent description of a language. Despite such restraints, the conception proposed needs to be theoretically sound. To assess it in this respect, it will here be contrasted with two existent theoretical approaches between which the present conception steers a middle course.
The present conception does not incorporate what has been called a universal grammar (Hauser et al. 2002). The grammar of a language is a specific mapping of devices chosen from the structural framework onto functions chosen from the functional framework. The mapping is a central aspect of the continuous ‘systematization’ involved in language as an activity (Coseriu 1978: 272). It obeys universal constraints embodied in the above-mentioned theories of cognition, communication and of a semiotic system; but it is free enough to produce diverse grammatical systems. There is, among other things, no universal set of word classes; beside nouns and verbs, languages choose from a large set. Nor does syntactic structure need to be shaped by constituency, based on contiguity; syntactic relations may instead be based on dependency, this in turn being based on semantic relationality and phora.
Nor does the present conception endorse radical relativism (Evans & Levinson 2009). It is not the case “that languages could differ from each other without limit and in unpredictable ways” (Joos [ed.] 1957: 96). Such a doctrine runs the risk of denying the unity of mankind. All natural languages are fundamentally alike both in what they achieve in terms of cognition and communication and in the expressive means they employ for it. This is because they all obey some universal constraints of the kind mentioned above. These include the functional domains: All languages can refer and predicate and specify the illocutionary force of an utterance. Some of the constraints on linguistic structure have been formulated in terms of implicational universals. A subset produces absolute universals, too, such as the distinction of two classes of sounds, vowels and consonants, or the distinction between two word-classes, nouns (not necessarily common nouns, but necessarily pronouns and proper names) and verbs.
Future theoretical work will enable the deduction of functional and structural constraints on linguistic activity and will allow us to give a more consistent shape to the two frameworks here proposed. Future empirical work will discover additional structural devices and grammatical meanings which must be assigned their place in the frameworks or will modify the frameworks themselves. For the time being, the frameworks are just offers for those seeking something applicable.
1 This conception appears already in Roman Jakobson's theory of a universal set of features which form the sounds of natural languages. The set suffices to describe all occurring speech sounds; but hardly any language makes use of all of the features. Every language chooses a subset of them to construct its sound system and adapts the features used to the specific structure of this system.
References
Coseriu, Eugenio 1978, Sincronía, diacronía e historia. El problema del cambio lingüístico. Tercera edición. Madrid: Gredos (Biblioteca Románica Hispánica; II. Estudios y Ensayos, 193).
Evans, Nicholas & Levinson, Stephen C. 2009, “The myth of language universals. Language diversity and its importance for cognitive science”. Behavioral and Brain Sciences 32(5):429-448.
Hauser, Marc D. & Chomsky, Noam & Fitch, W. Tecumseh 2002, “The faculty of language: What is it, who has it, and how did it evolve?” Science 298:1569–1579.
Joos, Martin (ed.) 1957, Readings in linguistics. The development of descriptive linguistics in America since 1925. New York: American Council of Learned Societies.