A holistic, socio-cognitive model of language and language change: a diachronic semasiological story of 'bedlam' (1).

ABSTRACT

This paper (2) presents a holistic, socio-cognitive model of describing language and explaining its change. The primary domain of the study is lexical semantics, both synchronic and diachronic. The object of the study are lexical expressions, which, like language as such, are contemplated on three dimensions--intersubjective, interactive and cognitive. The object is approached from three perspectives, namely theoretical, synchronic and diachronic. The latter two standpoints are not neatly detachable, because language in its perpetual change is history in the making. The panchronic perspective, in turn, is indispensable for the advancement of a holistic, socio-cognitive model of describing language and explaining its change, which is generated via a critically made synthesis of three investigative approaches, namely, cognitivism, Anlageteleologie, and invisible-hand theory. The authors to whom I am particularly indebted for engendering in my mind a fruitful capacity for wonder, resulting in the present model, are Adamska-Salaciak, Keller, Langacker, Itkonen, Lakoff and Johnson. The paper climaxes in a conjectural story unfolding the history of English 'bedlam', (3) thereby illustrating a practical application of the model.

1. A holistic, socio-cognitive model of language and meaning

Before advancing the model of explaining and understanding the socio-cultural evolution of language in general and meaning in particular, it is essential that a theoretical delimitation of the object of study be proposed.

From the vantage point of what might be called socio-cognitivism, language and, by implication, meaning are approached as dynamic, three-dimensional epiphenomena of human (re)cognition, specific communicative context, and historical socio-cultural context (cf. Keller 1994: 64, 87; Schonefeld 2001: 151). The three interdependent dimensions are dubbed cognitive or subjective, (4) interactive, and intersubjective. (5) In this construal, language is a historical, socio-cultural institution, a "phenomenon of the third kind" (Keller 1994: 57), (6) which acquires its functional potency via its embedment in a network of social relational acts performed by speakers, who connect this otherwise powerless abstract system of signs and formal rules to their experientially and interactively conceived conceptualizations, i.e., concepts grounded in neurolophysiologically determined conscious and unconscious cognition (e.g., Johnson and Lakoff 1997, 2002). The raison d'etre of language is to be found on the interactive level, and it is the exertion of influence, via verbalization of our conceptual experience, namely meaning (see Langacker 1988a: 6), upon our interlocutors (Keller 1994: 85). Meaning is therefore primarily the communicative means toward attaining the social goal of affecting one's addressee, and the product of what can be described as contextual semiosis, namely online meaning construction. The "contextual", "emergent" (Langacker 1987: 157) structure, negotiated in interaction, is meaning on the move--the first to break free from the conventional synchronic ranks, pulling the established structure to destinations unknown.

2. A dynamic construal of language and lexical meaning 

On the interactive dimension, language is tangibly a process. The inherent resilience and constant evolving of language, contingent upon micro-dimensional human action, provide for its optimal functionality and untrammeled subsistence--the super-goal striven for unconsciously and attained inadvertently and epiphenomenally in relevantly similar intersubjective behavior. (7) This is so because owing to its processualism, language can respond to the changing needs of its users, dictated by alterations in the world, as perceived by humans, and in conceptualizations thereof. And the existence of language is guaranteed as long as there are people who need the institutional framework of conventions and intuitive guidelines for innovative behavior, within which language operates and evolves, and which results from and reacts to what the attending users do, why and how they do it. Hence, the constant motion of language is far from chaotic. It is fueled by the cumulative, rational verbal acts of human subjects, who, under the influence of multifarious variables, strive for a variety of more or less idiosyncratic goals, few of which are conscious, and who nonetheless give rise to a macro-level structure which they have neither aimed at nor even thought of (see Keller 1994: 65). It is in their minds that any change-effecting motion within the system originates, and in their subsequent interaction that it either spreads via massive spiral circulation, engendering variation, and, sometimes, change, or dies a natural death of disuse.

Interaction is an intrinsically teleological, goal-oriented process, whose main, albeit unconscious, telos, is the exertion of social influence, from the perspective of the speaker, and contextually correct interpretation of the message expressed, from the standpoint of the hearer (see Adamska-Salaciak 1992: 34). In the hermeneutic enterprise, the decipherer is aided by the intersubjective substructure, the context, and the fact that whatever innovations should arise, they are never haphazard formations, but are the outcome of the working of various factors.

The variables causing (non-nomologically) particular behavioral patterns in communication via the activation of certain tendencies are characterized by a high degree of idiosyncrasy and stasis and are held to correspond to the traditional causes of change (Adamska-Salaciak 1986: 111; 1988: 468-469). Teleologen, a group of German scholars, discriminated between conditions that: (1) relate to the material aspects of language; (2) arise in our interacting psyche and soma; (3) derive from culture, society and nature (Adamska-Salaciak 1986: 111). Naturally, given the fact that the phenomenal world is accessible to us only insofar as we can conceptualize it in the mind, synchronized with our corporeal architecture, and that language is but a byproduct of the mind, we must confer priority on the mind-related conditions of change, which overlap with some of the active triggers of change.

The universal and dynamic teleologies governing human communicative behavior and originating in the esoteric human mind are mercurial and highly unpredictable, as are their consequences (Adamska-Salaciak 1986: 111-112). They are largely unconscious, which is why, as recognized by the Teleologen, it is unconscious Anlageteleologie that is applicable to linguistic considerations (Adamska-Salaciak 1986: 93, 108). These driving forces of change which, when actualized on a larger scale by different conditions, determine the direction of change, can be categorized, as shown by the Teleologen, into six classes of tendencies toward (1) clarity; (2) emotional discharge; (3) beauty of expression; (4) economy of effort; (5) order; and (6) social conventions (Adamska-Salaciak 1986: 112-113). Naturally, it must be borne in mind that as regards the instantiations of the tendencies, foreknowledge is out of the question.

We have thus described the environment conditioning human verbal behavior, governed by a variety of active teleologies. The overlap between the tendencies operative in a speech community and the unintended generation of a structured phenomenon are possible because the agents are exposed to a similar context, thus being more likely to tend toward parallel goals and choose the same 'solutions', or accept those chosen by their more creative co-speakers (cf. Ullmann-Margalit 1978: 270-271; Itkonen 1983: 37; Keller 1994: 90-91).

3. A holistic, socio-cognitive model of explaining semasiological change

In the preceding sections, we proposed a dynamic construal of language. Let us now see how it relates to the diachronic model, whose advancement is the aim of the present paper.

The author has resolved to name the model holistic and socio-cognitive. It is holistic for a number of reasons. First, it treats language on all its dimensions as part of a larger whole (cf. Adamska-Salaciak 1986: 64-65), within the context of which its evolution must be elucidated--on the subjective level, the whole is constituted by our 'embodied' (Johnson and Lakoff 2002) cognition; on the interactive, it is the communicative situation(s), wherein language use is considered; finally, on the intersubjective level, language, a socio-cultural institution, is contextualized by the entire society of language users and their collectively taken individual verbal acts (interactive level), which are possible due to the cognitive plane, and which shape and sustain the abstract, social level. Second, the model recognizes the inescapable multidimensionality of language and approaches it accordingly, that is, it recognizes the fact that all the three dimensions form a whole and that the understanding of the parts is not possible without reference to the whole and vice versa. Third, it seeks to generate a (conjectural) story, thereby answering not so much the 'why' as the 'how' question. This ensures that language history reads like a true story--there really is some narrative to it, which adds to the speculative reconstruction a genuine dimension of gestalt. The motivation of the compound modifier 'socio-cognitive' is again to be sought in the three planes of language. The 'socio-' part covers not only the intersubjective, public character of language as a phenomenon of the third kind, but also the teleological interactive plane, with the recognition of the desire on the part of communicators to exert social influence and, as a result, achieve social success, which motivates language use. The model is also partially cognitive in that language is, in the last analysis, a human phenomenon, and anything that is human is made possible by the embodied human mind. Overall, the model seeks to reconstruct language change by telling a comprehensive, if 'only' probabilistic, story, which shows how the explanandum could have been effected. Insofar as it concentrates on both the social and the cognitive individual planes, the explanatory model under consideration can be further specified into two consecutive phases, namely, teleological and invisible-hand.

Teleology, or rather Anlageteleologie, typifies both language use and language acquisition (Adamska-Salaciak 1992: 30). Naturally, the largely unconscious goals fueling verbal behavior are entertained by the agent(s), not by language, being thus actual constructs in the agent's mind (Itkonen 1983: 39-40).

These teleological tendencies are conjecturally reconstructed in order to make the explanandum "teleologically or finalistically understandable" (Von Wright 1971: 2-3; Adamska-Salaciak 1992: 30), rather than nomically explainable (see Itkonen 1983: 203-205). Hence, both the origin of change, characterized by creativity (innovation engendering variation), and its subsequent spread, tending toward regularity and adoption, need to be studied severally to secure comprehensiveness of the explanation and thereby full comprehension of the phenomenon examined (Adamska-Salaciak 1986: 116). (8) At both these stages the role of the human factor is undeniable, but its character changes: whereas at the innovative phase the individual speaker is the most important actor, at the stage of spread the emphasis shifts to the social level of interaction. In order for language evolution to be feasible, the unconscious goal-directedness needs to acquire a society-binding character. But how does it happen that, without any metalinguistic dialog between the agents engaged, the individual idiosyncratic acts and largely unconscious, success-oriented choices have a bearing on what surfaces on the intersubjective plane?

To answer this question and account for the spread of an innovation, we need a conceptual device that can help us understand how what has started as a mere innovation spreads, via rational social filtering, across the whole community of language users, or, at least, across a particular socially (or otherwise) distinct group. Such a theoretical tool is supplied by Keller in the form of an "invisible-hand explanation", "a conjectural story", endeavoring to answer the "how" of language change (Keller 1994: 38, 68), which, despite Keller's denial, clearly subsumes Anlageteleologie on the micro-level. Methodologically speaking, it should describe the explanandum, namely "the causal consequence of individual intentional actions", (9) and the explanantia, namely the conditioning static and triggering environment, and the process of the emergence of the phenomenon (Keller 1994: 70-71).

To conclude, it is essential to point out that only when the invisible-hand theory is combined with the Anlageteleologie of rational actions and with the cognitive embodied construal of the mind and its epiphenomena can we claim to advance a comprehensive model of explaining the socio-cultural evolution of language. Anlageteleologie closely correlates with the postulates of cognitive linguists, who put a lot of emphasis on unconscious cognition and its influence on behavior. It is the micro-plane of individual acts that necessitates reference to cognition and teleology, as the effectiveness of social behavior is possible due to the various psychological and social dispositions affecting the choice of the routes best fitted to attain the overt and/or covert goal(s) of one's action. The macro-level is the unintended and little cared-for result of what is initiated on the subjective plane and what spreads on the interactive dimension in the wake of our unconsciously teleological actions. An innovation must go through the sieve of social selection to gain the status of a variant and be available for acquisition. Naturally, the process of diffusion, fuelled by interaction, the rivalry whose trophy is social success and whose byproduct is language change, is usually long and arduous. What ends up as a change, which is only a phase, whose duration depends on the actions of speakers on the micro plane, may start as a 'one-season fad' catching on with a particular group of interactors, but, if deemed functional by more and more speakers, it may gradually win its way into the intersubjective system, wherefrom it is amenable to acquisition.

3.3. A diachronic story of 'bedlam' from a holistic, socio-cognitive perspective

As an illustration, let us now present a holistic, socio-cognitive explanation of the semantic changes undergone by the lexeme 'bedlam' (cf. Nerlich and Clarke 2001: 256). The word was generated via the process of ellipsis, to which the nominal phrase 'the Hospital of St. Mary of Bethlehem in London' was subject. Thus entering the vocabulary of the English language in the 16th century, it added to the lexicon (lexical change). Before we ponder on the etiology or rather teleology behind the semantic evolution of the lemma, let us say a few words about the history of the hospital in question:

"The priory of St. Mary of Bethlehem outside Bishopsgate was founded in 1247 and began to receive lunatics in 1377. It was given to the city of London as a hospital for lunatics by Henry VIII in 1547. In 1676 it was transferred to Moorfields and became one of the sights of London ..." (BDPF).

However, already in the fifteenth century "[f]or a modest fee, people could watch the inmates behind the bars, much as we view animals in the zoo today, except that onlookers would tease the poor souls with jeers and taunts" (DWPO).

"It was a place for assignations and one of the disgraces of seventeenth century London.

'All that I can say of Bedlam is this: 'tis an almshouse for madmen, a showing room for harlots, a sure market for lechers, a dry walk for loiterers' New Ward: The London Spy (1698).

In 1815 Bedlam was moved to St. George's Fields, Lambeth, the present site of the Imperial War Museum ... [I]n 1931 the occupants were moved to West Wickham" (BDPF).

"Inmates of Bedlam who were not dangerous were kept in the 'Abraham Ward' and occasionally allowed out in distinctive dress and permitted to beg. This gave an opportunity to many impostors" (BDPF).

So much on the history of the institution, which may be considered a condition concerning the material culture. Let us now return to the lexical item and its development. Why or what for could the process of 'verbal abbreviation' (Nerlich and Clarke 2001: 255) have taken place? The most plausible answer is that it took place because speakers referring to the disreputable asylum, whose name was rather lengthy, were driven by the universal human tendency toward economy, thus seeking the shortest possible way of referring to the place, with simultaneous optimal communicative efficiency, i.e. without risking unintelligibility. In addition, the fact that the hospital became a sightseeing spot in London must have considerably increased the frequency of occurrence of the lexeme in discourse, which made the process of accommodation of the word to the vocabulary of adult speakers easier. That being so, it was not long before the lemma was available as a conventional vocabulary item to new generations acquiring the language. Around the same time that the change caught on in the community, a further semantic development occurred, whereby the meaning of the lemma was extended metonymically to signify also a patient of the hospital. Why did that happen? The patients were physically linked to the hospital, and what happened there was believed to concern them directly. What is more, some of the less severely disturbed patients were allowed to leave the asylum and confront the society as beggars, being, nevertheless, stigmatized by "a tin plate on their left arm" (SOED), which let everyone know where they came from. Important psychological conditions that must have influenced this metonymic "polysemization" (Nerlich and Clarke 2001) are the contiguity obtaining between both real-world and conceptual categories 'asylum' and 'patient of an/the asylum', and simultaneously the salience accorded to the patient-in-what-kind-of-hospital feature. The following lines illustrate the new metonymically extended usage of 'bedlam':

"She roar'd like a Bedlam" (Swift); "Plaine bedlam stuffe" (Milton) (SOED);

"Let's follow the old earl, and get the bedlam/To lead him where he would ..." (King Lear III, 7, 103); "The country gives me proof and precedent/Of bedlam beggars who with roaring voices/Strike in their numb'd and mortify'd bare arms/Pins, wooden pricks" (Shakespeare) (DEL); "art thou bedlam?" (Henry V, V, 1, 20), "the bedlam brainsick duchess" (second part of Henry VI, III, 1, 51), "a bedlam and ambitious humour" (second part of Henry VI, V, 1, 132) (SL).

This metonymic extension was a semasiological process, which was, however, followed at some point by an onomasiological one. The latter consisted in the employment of the internal formative process of derivation, which produced a lexical change in the form of another term for "an inhabitant of Bedlam; a madman", namely, 'bedlamite' (bedlam + -ite "forming names denoting natives of a country", here, of a place (NODE)). Its use can be illustrated by the following quotation (from DEL):

   If wild ambition in thy bosom reign
   Alas! Thou boast'st thy sober sense in vain;
   In these poor bedlamites thyself survey
   Thyself less innocently mad than they

(Fitzgerald).

Notice that the expressive value of the short text is positive as far as the attitude toward 'bedlamites' goes. One can sense in it sympathy or pity, but neither hatred nor mockery.

Returning to the metonymic change (container for content), it could have occurred because the speakers were driven by a mixture of tendencies toward least-effort (something short), and, more importantly, toward order (prototypically organized, i.e. Wittgenstein's family-resemblances-governed polysemization of a category, here, on the basis of conceptual and real-world contiguity). Bedlam patients seem to have formed a category of their own, whose special character was determined by the realities of this concrete asylum. The first two uses of the lexeme are likely to have coexisted for some time. Their frequency of occurrence may have gradually decreased when the word came to be used to designate any lunatic asylum. This meaning, marked in dictionaries as 'archaic' (e.g., NODE), is preserved till the present day.

The change was made possible by the cognitive figurative device of synecdoche and, to be more specific, of generalization, that is, going one level up the category 'hospital'. The process of generalization may have been collectively applied for three reasons: because speakers sought (1) an original, or (2) a new and less direct, that is, euphemistic, or (3) a more direct, dysphemistic, way of talking about mental hospitals. In each of these cases the speakers would have been unconsciously driven by the tendency toward emotional discharge. That people should search for a less direct way of referring to asylums is anything but strange, given that diseases in general and mental disorders in particular have always been a taboo--in Ullmann's (1962: 206-207) terms, 'a taboo of delicacy'. 'Bedlam' provided an indirect way of expressing the unpleasant concept. There was little, if any, risk of miscommunication because the place was well known. If this indirect, euphemistic reference really did occur, it must have lost its function rather quickly due to the stigma attached to the hospital (see the quotation from New Ward: The London Spy above). The opposite process, namely dysphemization, may also have been operative from the beginning in many micro-plane individual instances of usage, being motivated by the speakers' desire to express their negative or even mocking attitude toward such hospitals. The name of the hospital in question provided an interesting category for generalization due, on the one hand, to the place's localization in the capital, and, on the other, to its notoriety for squalid conditions and cruel treatment of inmates. Its popularity can be further evidenced by folk songs about bedlam and its patients dating back to the 16th and 17th centuries. One example is a song entitled Tom o'Bedlam--"[o]ne of the greatest of Elizabethan anonymous poems" (BDPF). In fact, 'Tom o'Bedlam' is a phrase, whose equivalent is 'Abram-man' or 'Abraham cove' (see above 'Abraham Ward'), meaning "a mendicant who levies charity on the plea of insanity" (BDPF).

The gradual disappearance of the first indexical meaning could have been motivated by the fact that (i.e. happened because) speakers wanted to avoid confusion as to whether the referent was general or specific (the social-conventions-tendency, especially, the Gricean principle of manner). Another reason may be related to the fact that the word functioning as a proper name was fossilized in the phrase 'Tom o'Bedlam' (compare: "what a shambles" in: Nerlich and Clarke 2001: 254). One might wonder why a similar generalization has not affected the second meaning. For some reason this has not happened, or, perhaps, has happened very sporadically, because it is recorded in a few dictionaries (e.g., DEL, SOED). It must be pointed out, however, that it is not at all clear whether the dictionaries mean any madman, or perhaps only the lunatics from the Bedlam. The reluctance of speakers to generalize the meaning of the lexeme to designate any patient of an asylum could have been motivated by the fact that the Hospital of St. Mary of Bethlehem had a very bad reputation. The infamy of the asylum, made even more public by the visits of outsiders, must have contributed to the pejoration of the word 'bedlam'. There was no socially perceived need to refer to patients, unfortunate enough to have been afflicted with a disease, by derogatory terms. Finally, also in the seventeenth century, 'bedlam' began to be used in a metaphorical way to mean "a scene of wild uproar", which is a more abstract concept. This change occurred because speakers in search of still more expressive means of communication (the tendencies toward beauty of expression and/or toward emotional discharge) perceived a similarity between prototypical scenes in an asylum and scenes of wild uproar that they witnessed outside 'bedlams' (tendency toward concreteness--the use of concrete concepts to stand for more abstract ones).

This is a conjectural story of how the word 'bedlam' has fought for survival against the changing circumstances and expressive needs of speakers. Let us close the story with the reiteration of the non-triggering conditions of the change, the dynamic tendencies, (10) and the tendency statements (see n. 13), while also presenting a network model (11) of the change.

Conditions:

a) A long name of an ill-famed lunatic asylum in the capital, notorious for its appalling treatment of the insane (a formal aspect of language--a linguistic condition);

b) The hospital was open to visitors seeking entertainment in observing the tortures of the inmates, which led to the increase in the place's popularity; this publicity must have been reflected in the frequency of use of the word (a social condition); (12)

c) The conditions and treatment procedures in asylums changed in the Renaissance; mental maladies were recognized as such, and were no longer treated as possession by the devil (a socio-cultural condition);

d) The general idea of how the insane behave has remained unchanged (a social condition);

e) A negative attitude toward the insane, amplified by the fact that some residents of Bedlam feigned madness to extort money through beggary (a social condition);

f) The taboo-status of anything concerning madness (a psychological condition).

Tendencies (operative at different stages):

T1 Speak in such a way that you are understood, with expending as little effort as possible, but without risking misunderstanding or unintelligibility;

T2 Speak in such a way that you can make use of any useful regularities that come up and that minimize your effort;

T3 Speak in such a way that you appear conspicuous and original;

T4 Speak in such a way that you sound polite, and your language beautiful;

T5 Speak in such a way that you make your feelings/opinions explicit.

Tendency Statements

TS1 Words that are often used tend to become shorter

TS2 Words that are used metonymically tend to save our production effort; (13)

TS3 Words that are used metonymically tend to form synchronic and diachronic chains (Nerlich and Clark 2001);

TS4 Words that are used figuratively tend to undergo polysemization;

TS5 Words that signify tabooed concepts tend to undergo pejoration.

The diagram above shows how the variably motivated process of socio-cultural evolution of the lemma 'bedlam' could have proceeded. In Early Modern English, the lexeme appears to have been more polysemantic than it is in Present-Day English, where, if listed at all, the meaning "asylum" is marked as 'archaic', and the figurative abstract sense functions as a prototype, whose position, given the (current) circumstances, is not in jeopardy. What will happen with the lexeme even in the foreseeable future remains as much a mystery as is the case with any other phenomena which are teleologically actuated and collectively, but inadvertently, brought about. The only certain thing is that the process-like nature of such intersubjective phenomena will always prevail, which is why whatever they seem to be at a given moment is not going to last. As epiphenomena of human actions, they must and will evolve.

4. Concluding remarks

The synchro-diachronic model of language and meaning, built in this paper, presents the 'understandum' as a dynamic and multifaceted epiphenomenon, whose diachrony and synchrony form a whole by continuously merging into each other in a way that makes any strict separation artificial. The perpetually evolving structure of language is a socially filtered and structured byproduct of historical, rational actions of exerting influence via online meaning construction, determined by universal psychological and social (14) teleologies--inculcated into the embodied mind. Growing out of the past and weaving into the future, already altered by the threads of the present, language is part of human history, amenable to constant interpretation, which perpetuates its evolution. This context-dependent interpretation inherent in communication may sometimes turn out to be 'misinterpretation', resulting, if it is socially accepted, in a revaluation of the sign(s) involved. The potential macrostructural outcomes of language interpretation are unpredictable and transitional. That being so, the process leading to their emergence can only be reconstructed ex post facto via non-nomic, conjectural logical reasoning, which has been exemplified in 3.3.

To round off, let us say that the attendant philosophical and methodological assumptions of the panchronic model are incompatible with theories that posit a non-interdisciplinary character of linguistics, the autonomy of language, the subservience of meaning, or its disembodiment, homogeneity and stability. Irreconcilable with the model are also any local cynosures that treat language fragmentarily, without recognizing its multidimensionality, and those that fail to give due attention to the links obtaining between linguistic and extra-linguistic knowledge, or to the role of figurative processes in language, thought and understanding. Discordant with the model are also theoretical frameworks demanding that the deductive-nomological mode of logical reasoning be employed in explanations of linguistic facts or that the truth of theories be validated with certainty. Finally, any theory that disregards the process-like character of language, and neglects the vital interrelations between language synchrony and diachrony, is also incongruous with the standpoint espoused here. This stance, climaxing in the model of explaining language change, is grounded in the conviction that the dynamic and multifaceted nature of language must be reflected in any comprehensive study of this social phenomenon grounded in and enclosing history.

[FIGURE 1 OMITTED]

REFERENCES

Adamska-Salaciak, Arleta

1986 Teleological explanations in diachronic phonology. [Unpublished Ph.D. dissertation, Adam Mickiewicz University, Poznan.]

1988 "The Teleologen on language change", Folia Linguistica Historica 8/1-2: 457-480.

1992 "Wyjasnianie w jezykoznawstwie historycznym" [Explanation in historical linguistics], Biuletyn Polskiego Towarzystwa Jezykoznawczego 47, 48: 27-42.

1993 "Language change as a phenomenon of the third kind", Folia Linguistica Historica 12/1-2: 159-180.

Ferguson, Adam

1767 An essay on the history of civil society. Edinburgh. [No indication of publisher.] Harris, E. Errol

1959 "Teleology and teleological explanation", The Journal of Philosophy 56/1: 5-25. Itkonen, Esa

1983 Causality in linguistic theory. London: Croom Helm. Johnson, Mark--George Lakoff

1997 Philosophy in the flesh (floppy disk version, [C] George Lakoff and Mark Johnson).

2002 "Why cognitive linguistics requires embodied realism", Cognitive Linguistics 13/3: 245-263.

Keller, Rudi

1994 On language change: The invisible hand in language. London--New York: Routledge.

Krawczak, Karolina

2004 Towards a holistic, socio-cognitive model of explaining semasiological change. [Unpublished M.A. thesis, Adam Mickiewicz University, Poznan.]

Langacker, W. Roland

1987 Foundations of cognitive grammar. Vol.1. Theoretical prerequisites. Stanford: Stanford University Press.

1988a "An overview of cognitive grammar", in: Brygida Rudzka-Ostyn (ed.), 3-48.

1988b "A usage-based model", in: Brygida Rudzka-Ostyn (ed.), 127-161.

Lyons, John

1995 Linguistic semantics: An introduction. Cambridge: Cambridge University Press. Nerlich, Brigitte--David D. Clarke

2001 "Serial metonymy. A study of reference-based polysemization", Journal of Historical Pragmatics 2/2: 245-272.

Popper, Karl R.

1974 Objective knowledge. An evolutionary approach. Oxford: Clarendon. Rudzka-Ostyn, Brygida (ed.)

1988 Topics in cognitive linguistics. Amsterdam: John Benjamins. Schonefeld, Doris

2001 Where lexicon and syntax meet. Berlin--New York: Mouton de Gruyter. Stubbs, Michael

2001 Words and phrases: Corpus studies of lexical semantics. Oxford: Blackwell. Ullmann-Margalit, Edna

1978 "Invisible-hand explanations", Synthese 39/2: 263-291. Ullmann, Stephen

1962 Semantics. Oxford: Blackwell. Von Wright, Georg Henrik

1971 Explanation and understanding. New York: Cornell University Press.

DICTIONARIES

Evans, Ivor H. (ed.)

1986 Brewer's dictionary of phrase and fable (BDPF). (2nd edition). London: Cassell. Johnson, Samuel (ed.)

1828 A dictionary of the English language (DEL). 2 vols. Heidelberg: Engelmann. Little, William--H.W. Fowler--J. Coulson (eds.)

1956 The shorter Oxford English dictionary (SOED). Oxford: Clarendon Press.

Morris, Mary--William Morris (eds.)

1966 Dictionary of word and phrase origins (DWPO). New York: Harper and Row Publishers.

Pearsall, Judy (ed.)

2001 The new Oxford dictionary of English (NODE). Oxford: Oxford Univesity Press. Schmidt, Alexander (ed.)

1902 Shakespeare lexicon (SL). 2 vols. (3rd edition). Berlin: Georg Reimer.

KAROLINA KRAWCZAK

Adam Mickiewicz University, Poznan

(1) I would like to thank professor Arleta Adamska-Salaciak for her constructive comments on earlier drafts of the present paper.

(2) The present paper is an abridged version of Krawczak 2004.

(3) Throughout the present paper, the punctuation conventions used will be those employed by Lyons (1995: 24-26), namely, single quotation marks for lexical expressions, italics for forms and double quotation marks for meanings.

(4) In this context, 'subjective' is used in the sense of being peculiar to, or experientially generated by an individual human subject.

(5) Cf. Keller (1994: 133-135), Stubbs (2001: 233-235), Popper (1974: 106-108).

(6) A "phenomenon of the third kind" is understood by Keller as a socio-cultural institution, and more specifically as "the result of human action, but not the execution of any human design" (Ferguson 1767 cited in: Keller 1994: 37). It is initiated inadvertently by a multiplicity of verbal actions marked by certain parallelisms (Keller 1994: 91), which are severally insufficient, but jointly pertinent to their corollary (Keller 1994: 64).

(7) Cf. Harris (1959: 10), Itkonen (1983: 211) and Keller (1994: 91).

(8) The ensuing considerations in the present paragraph are based on Adamska-Salaciak (1986: 116).

(9) The law-like generation that Keller proclaims to be peculiar to the emergent phenomenon is highly dubious, which is why it is more appropriate to refer to the 'laws' as "tendency statements" (Adamska-Salaciak 1993: 169).

(10) The way in which the tendencies are formally couched is based on the pattern set by Keller (1994: 101).

(11) The "network model", known also as "usage model", has been developed by Langacker (e.g., 1988b). It presents the multiple subsenses of a lexeme as "nodes in a network, linked to one another by various sorts of 'categorizing relationships'" (Langacker 1988b: 134). There are three kinds of such relations, namely, 'extension', elaboration of 'the schema', and discerning 'mutual similarity' (Langacker 1988b: 134). Langacker graphically represents the three kinds of relations by dashed, solid, and dashed double-headed arrows, respectively. In addition, the global prototype is marked with bold lines, and so is the local one, with the difference that the lines of the latter are comparably thinner.

(12) As Zipf's law states, the more frequently a word is used, the shorter it tends to get.

(13) This means that "metonymy is a strategy used to extract more information from fewer words", it is "a conceptual and semantic abbreviation device" (Nerlich and Clarke 2001: 253).

(14) These subsume communicative principles, including that toward being understood/exertion of influence upon the other; and the tendency toward beauty.