LINGUIST List 24.379

Tue Jan 22 2013

Review: Cognitive Science; Semantics; Syntax: Panther & Radden (2011)

Editor for this issue: Monica Macaulay <>

Date: 05-Dec-2012
From: Daogen Cao <>
Subject: Motivation in Grammar and the Lexicon
E-mail this message to a friend

Discuss this message

Book announced at

Editor: Klaus-Uwe PantherEditor: Günter RaddenTitle: Motivation in Grammar and the LexiconSeries Title: Human Cognitive Processing 27Publisher: John BenjaminsYear: 2011

Reviewer: Daogen Cao, Zhejiang University of Finance and Economics


The present volume is a sequel to the volume “Studies in LinguisticMotivation,” edited by the two editors in 2004, and so serves as a continueddiscussion on linguistic motivation. It is demonstrated by the contributorshow language-independent factors such as bodily experience, emotion,perception, action, social/communicative interaction, and culture, whenmediated through cognition, motivate grammar and the lexicon in a variety oflanguages such as English, German, French, Italian, Hungarian, Russian,Croatian, Japanese, and Korean.

The volume is divided into two parts plus an introduction. The first partcontains nine articles addressing issues of motivation pertaining to grammar,and the second part comprises five articles that reveal and demonstratemotivating factors in the lexicon.


The Introduction authored by the two editors presents a sketch of the overallframework of the interaction among human systems as well as a summary of thecontributions to the volume.

In Part One, various aspects of grammar are explicated in terms oflanguage-independent motivational factors. According to Ronald Langacker, theEnglish auxiliaries are motivated by their crucial role as existential verbs.They schematically predicate the existence of a relationship to be negotiatedby the interactive system. The default existential predicate is ‘do’; itindicates the unqualified existence of a process for negotiating purposes.‘Do’ is, therefore, needed with e.g. yes-no questions (‘Did she wait?’),negative statements (‘She didn’t wait’), and affirmative statements (‘She DIDwait’), in which the grounded process is negotiated between the interlocutors.In questions, for example, the hearer’s position is elicited. By contrast, inpositive statements the speaker takes the validity of the proposition forgranted, i.e. assumes that it is not to be negotiated. Therefore, theexistential ‘do’ is not needed and hence not expressed. As a result thesentences ‘*She does will wait’ and ‘*She did wait’ with the unstressed formof ‘did’ are ungrammatical.

Working within the cognitive linguistics theoretical framework, Rong Chen putsforward a Ground-before-Figure (GbF) model that is instantiated by theinversion of subject and lexical verb in sentences like ‘In the room was aunicorn’, as well as the existential construction (e.g. ‘There is a unicorn inthe room’). Both the constructions are rooted in the gestalt-perceptualprinciple of figure and ground. It is argued, however, that the existentialconstruction is more versatile (cf. ‘*In the room was no sign of life’ but‘There was no sign of life in the room’), and is therefore a more typicalinstance of the GbF model because the word there in the existentialconstruction designates mind as a default ground where virtually anything canor cannot exist. The perceptual principles of figure-ground gestalt thusmotivate the GbF model and its linguistic manifestations.

Within Langacker’s Cognitive Grammar framework, Cristiano Broccias discussesparticipant-oriented uses of adverbs, which exhibit cognitive and perceptualmotivations. Two types of participant-oriented adverbs are identified, viz.“manner” and “transparent” adverbs. The former is exemplified by ‘Fred ate thesausages ravenously’, in which the adverb not only indicates manner but alsoassigns a property to the subject referent; the latter is exemplified by‘Sally painted the house beautifully’, in which the adverb only indicates theproperty of the object noun. It is claimed that the difference between the twotypes of adverbs results from a difference in viewing arrangement, inparticular, the vantage point of the conceptualizer.

Dangling participles are considered poor expressions in written StandardEnglish, but their use cannot be eradicated in speech, as testified by thedata from the British National Corpus. Naoko Hayase argues that this isbecause the construction is ecologically well motivated by its specificcommunicative function within the grammatical system of English. Hayase showsthat constructions with a dangling participle describe a coherent “cognizancescenario”, a scenario that is based on our common experience of noticingsomething while engaged in some activity. The dangling participialconstruction evokes a conceptualizer who conceives the situation described inthe main clause. The conceptualizer is typically the speaker or a virtual orgeneric person. The dangling participial construction is thus highlysubjective.

It is observed by Mitsuko Narita Izutsu and Katsunobu Izutsu that lexicalitems such as English ‘while’, ‘where(as)’, German ‘während’, Japanese‘-nagara’, ‘tokoroga’, and Korean ‘-myeonseo’ have all undergone a semanticshift from the meaning of TEMPORAL/SPATIAL OVERLAP to the meaning ofCONTRAST/CONCESSIVE. Underlying the semantic change are motivational factorsthat are both cognitive (metonymic inference) and perceptual. Experiments showthat metonymic inference is motivated by temporal/spatial overlap, whichlargely corresponds to perceptual overlap in Langacker’s viewing arrangementof two situations.

Through carefully designed experiments, Teenie Matlock shows that imperfectivedescriptions of past events trigger more inferences than their perfectivecounterparts. This result conforms with the semantics of the imperfective andperfective aspect: the imperfective aspect provides an internal perspective ofa situation and focuses on its ongoingness, while the perfective aspectprovides an external perspective of a situation and focuses on its completion.The “more action” effect of the imperfective is motivated by our ability tomentally simulate events: in taking an internal view of an ongoing situation,our subjective experience of the action increases and engages us in “moment tomoment processing”.

Certain types of the caused-motion construction involve non-motion verbs, e.g.‘He gazed me out of the club’. When such non-motion verbs are inserted intothe caused-motion construction, they are coerced into expressing a change oflocation by virtue of metaphors like EXPERIENTIAL ACTION IS EFFECTUAL ACTION.Using a decompositional approach, Annalisa Baicchi is able to account for thechange in valency of non-motion intransitive verbs and theirconstruction-coerced change in meaning.

The subject (doer) of clauses containing English ‘must’ is invariably markedwith nominative case but the subject of clauses containing Hungarian ‘kell’has either nominative or dative case marking. The grammaticalnominative/dative alternation of the doer in Hungarian is accounted for byPéter Pelyvás with the dual role attributed to the doer participant in theconceptual structure: the agent-like role in performing the imposed act and apassive role in the non-autonomous obligation portion of the event modelinvolving deontic modality.

The highly developed honorific systems in Korean and Japanese are functionallysimilar but differ with respect to non-subject referent honorifics, whichindicate the speaker’s deference toward a non-subject referent participant inthe event described. Satoshi Uehara argues that the relatively higherproductivity of non-subject referent honorifics in Japanese is motivated bytwo socio-cultural factors: the egocentric viewing arrangement which is basedon the Japanese self-humbling nature of deference, and the distinction between‘uchi’ (‘inside’) and ‘soto’ (‘outside’) which refers to the omnipresentboundary between the in-group and the out-group.

Part Two of the volume is devoted to motivational accounts of certain aspectsof the lexicon. Elena Tribushinina argues that the conceptual motivation inadjectival semantics lies in cognitive reference points. The reference-pointreasoning is actually a pervasive cognitive phenomenon intrinsic to theinterpretation of dimensional adjectives. It is argued that a multitude ofreference points may be used to anchor conceptual specifications ofadjectives, prototypes being only a special case of the reference-pointmechanism. For example, dimensional adjectives may be interpreted vis-à-vis anaverage value of the property (norm), endpoints of the scale (as with themeasure phrase ‘five feet seven inches tall’) and dimensions of the human body(EGO) (as in the sentence ‘Giraffes are tall’).

The socio-cultural motivation of the use of the metonymy CAPITAL FORGOVERNMENT is investigated by Mario Brdar and Rita Brdar-Szabó, who find thatthe metonymy is used more frequently in English and German than in Hungarianand Croatian newspapers. For the latter two languages, they find that themetonymy is more frequent in weekend editions than in workday editions. Thisreceives a motivational explanation given a cultural model which contrasts aweekend frame of mind with an ordinary workday frame of mind.

The final three chapters in Part Two focus on motivational processes withlinguistic sources, i.e. motivational links among lexical items. DanielaMarzo’s article investigates Italian native speaker judgments aboutmotivational relations in the lexicon, and develops Peter Koch’s conception ofmotivation by regarding a lexical unit as motivated if it is both formally andconceptually related to another lexical unit. Potential motivational relationsin the lexicon are also the subject of Birgit Umbreit’s article. In contrastto traditional studies, which view motivation as a unidirectional process froma “motivational base” to a more complex unit, Umbreit proposes amultidirectional network of motivated relations.

The issue of motivatability is central to Christina Sanchez-Stockhammer’sarticle, in which degrees of a lexeme’s motivatability are distinguished. Forexample, English ‘football’ is “fully motivatable”, ‘income’ is “partiallymotivatable”, ‘understand’ is “unmotivatable but transparent” and ‘leaf’ is“unmotivatable”. The analysis of the motivatability of the 2,500 most frequentwords of English and German conducted in the article confirms that Germanvocabulary is indeed more motivatable than English vocabulary. However, theassumption that this is due to the larger stock of Romance words in Englishcould not be confirmed.


As is demonstrated by this volume, motivation in grammar and the lexicon canbe established as a central theoretical construct in the study of naturallanguage. The collection of papers brought together in this volume pointsunmistakably to the fact that much that is done in the fields of cognitive andfunctional linguistics contributes to the study of motivation in language. Infact, an overriding task for linguistics is to address the issues of how muchlanguage is motivated and how it is motivated. This is consistent with thefact that motivation is in a large part a principle of language.

Although Radden & Panther (2004:42) once modestly claimed that “it isimpossible to provide conclusive evidence for or against the hypothesis thatall of language is motivated by language-independent factors”, as Ronald W.Langacker remarks in his contribution to this volume (p. 30), “virtuallyeverything [in language] is motivated”, and motivation affords some middleground which is an alternative to the two extreme positions, namely, fullpredictability of language structure and complete arbitrariness of languagestructure. In a similar vein, Heine (1997:3) argues that since “humanbehaviour is not arbitrary but […] driven by motivation”, language structure,which is a product of human behaviour, “must also be motivated”. It istherefore not implausible to see motivation in language as being normative,which is divergent from de Saussure’s (1916/1959) notion of relativemotivation in the sense that motivation is but a limiting case ofarbitrariness (Radden & Panther 2004). And if motivation in language is thenorm that reduces arbitrariness to the status of being the last resort inlanguage (Lakoff 1987), then explanatory adequacy as a standard laid down forthe formulation of linguistic theories is translatable as revealing themotivated correlation between language and language-independent factors, ofwhich cognition plays a central role (Introduction to this volume by Panther &Radden, p. 2). Linguistic inquiry should adopt a motivational approach tomost, if not all, linguistic phenomena.

At this junction, we would like to do some justice to generative grammar, inparticular, Chomsky’s Transformational Generative Grammar, which also has,among others, the theoretical aim of attaining explanatory adequacy but whichreceives much criticism from cognitive linguists for the thesis of theautonomy of grammar. For example, in his contribution to the present volume(p. 30), Langacker attacks the thesis once again by saying that generativegrammar as an autonomous or self-contained system (module) would imply fullpredictability internally (serving to predict/generate all and only thegrammatical sentences of a language) and essential arbitrariness externally(without being constrained by other factors such as cognition, communicationand social interaction).

We would like to point out, however, that the syntactic component in thehypothesized faculty of language now referred to as narrow syntax (Chomsky2000; 2001) is autonomous in that the computation of and through syntacticobjects or syntactic features that takes place in narrow syntax isencapsulated from other components or modules like morphology, phonology andsemantics. In other words, the narrow syntax computation does not see thecomponents external to the syntactic component, and the syntactic derivationoccurs without reference to meaning, discourse or language use. Once thenarrow syntax embarks upon a computation with the feature matrices retrievedfrom the lexicon, it will do the computing with the given features only, beingprecluded from either backtracking to morphology or looking ahead to phonologyor semantics until the Spell-Out operation that submits the product ofsyntactic computation to interpretational mechanics (the inclusivenesscondition). For all this, however, what the ultimate output of the syntacticcomputation is like is still shaped by what has been furnished by the lexiconor morphological component. For one thing, lexicon or morphology feedssyntactic computation; for another thing, it is the uninterpretable formalfeatures within the narrow syntax that trigger and drive certain syntacticoperations. Furthermore, what is yielded by the syntax will eventually besubject to evaluation by the sensori-motor system (phonological component) andthe conceptual-intentional system (semantic component) that both interfacewith narrow syntax. The evaluation is made with respect to the legibilitycondition and the principle of full interpretation.

The above account, sketchy as it is, suffices to indicate that Chomsky’smodular model of grammar is only in part or relatively autonomous. It is onlywhen feature assemblies are being processed that syntax can be said to beautonomous. Autonomy of syntax is conditional and principled to the extentthat syntax is susceptible to morphology prior to syntactic computation andevaluable posterior to syntactic computation in terms of semantics andlanguage use. In linguistic performances, there is likely to be insufficientor excessive morphological information or wrongly chosen morphologicalinformation entering narrow syntax, and in this case, just because of therestriction imposed by the stringent inclusiveness condition which is a reflexof narrow syntax autonomy, the narrow syntax would have no chance whatsoeverof mending the information (replenishing what is lacking or removing thesuperfluous or replacing the mistaken) so that the computation could be savedand the resultant structure would become possibly convergent. This constitutesa case of autonomy of syntax causing its own crash, indicating that thecurrent model of grammar envisaged by Chomsky in his Minimalist Program doesnot claim full predictability, thus escaping Langacker’s criticism. The escapeis possible because we have a different line of thinking over autonomy ofgrammar or syntax, given Chomsky’s idea of grammar.

On the other hand, though the very narrow syntax computation does not lenditself to influence from language-external factors such as cognition,communication or social interaction, the general computational principlesposited in Chomsky’s model of grammar for computational efficiency are stillmotivated by interpretational considerations such as economy orinterpretability/legibility or biological considerations, viz. theconsideration of perfection or optimality in language as a design solution tothe problem of how to make language both learnable and usable. This refutesLangacker’s view that autonomy of grammar rules out the predictability ofgrammatical rules on language-external grounds and consequently the rules takearbitrary forms.

As a final remark, it is worth noting that Chomsky has never been so ambitiousas to assert that syntax could lay claim to everything in language. In fact,he acknowledges that grammars are shaped in part by function: “Surely thereare significant connections between structure and function; this is not andhas never been in doubt. […] Searle argues that it is reasonable to supposethat the needs of communication influenced [language] structure” (Chomsky1975:56-58).


Chomsky, Noam. 1975. Reflections on Language. New York: Pantheon.

Chomsky, Noam. 2000. Minimalist inquiries: The framework. In R. Martin et al.(eds.), Step by step: Essays on minimalist syntax in honor of Howard Lasnik,pp. 89-155. Cambridge, Mass: MIT Press.

Chomsky, Noam. 2001. Derivation by phase. In M. Kenstowicz (ed.), Ken Hale: ALife in. Language, pp. 1-52. Cambridge, Mass: MIT Press

de Saussure, Ferdinand. 1916/1959. Course in General Linguistics. New York,Toronto, London: McGraw-Hill.

Heine, Bernd. 1997. Cognitive Foundations of Grammar. New York and Oxford:Oxford University Press.

Lakoff, George. 1987. Women, Fire, and Dangerous Things: What CategoriesReveal about the Mind. Chicago and London: University of Chicago Press.

Radden, Günter & Klaus-Uwe Panther. 2004. Introduction: Reflections onMotivation. In Radden, Günter & Klaus-Uwe Panther (eds.). Studies inLinguistic Motivation (Cognitive Linguistics Research 28), pp. 3-46. Berlin:Mouton de Gruyter.


Cao Daogen is a teacher teaching and studying linguistics in ZhejiangUniversity of Finance and Economics, English Department. He holds a Ph.D. inLinguistics (Fudan University, 2005). His research interests include syntax,semantics and cognitive linguistics.

Page Updated: 22-Jan-2013