1

Prototype Theory in Cognitive Linguistics

Daniel Aberra

This paper reflects on the understanding and the use of prototype theory of concepts in cognitive
linguistics. Particularly, it focuses on the understanding and misunderstanding by the adherents,
its primary use and extended use in the field. To assess prototype’s life and role in cognitive
linguistics a group of selected literature both from cognitive psychology mainly of Rosch, 1978;
Smith & Medin, 1981; Cohen & Murphy, 1984; Murphy & Medin 1985; Komatsu, 1992:
Murphy, 2002 and cognitive linguistics mainly the writings of Taylor, 1990; 1991; 1994; 1995;
2002; 2003; 2004; Geeraerts, 1989; Wierzbicka, 1990, Newmeyer, 1998 and Grezga, 2003 have
been read. As a result, it is observed that: (1) There are some misunderstandings of prototype’s
basic tenets and confusing them with other models of the theory of concepts. (2) Prototype is
considered to be ‘a catchall theoretical device’ instead of an analytical tool.(3) Like in cognitive
psychology prototype model is primarily used for the study of lexical semantics specifically
nouns and prepositions of Indo-European languages, and (4) Unlike in cognitive psychology,
prototype has exhibited an extended use and application in phonology, syntax and diachronic
lexical semantics. This effected into introducing new ideas in prototype like paragon, generic
neutral

, real world versus ideal world prototype distinction and what Grzega (2003) calls

conceptual recategorization

.

Introduction
In order to make sense of the world, human beings carve nature at its joints, dividing it in to categories
endlessly. What humans categorize in the sense of psychology, according to Smith (1990), are objects that
are believed to belong together. Humans beings, however, do not stop there. They form concepts. A
concept may mean several things ranging from a mental representation corresponding to the category, a
group of objects; a kind of mental glue (Murphy, 2002) to a representation of a chunk of knowledge that
can be used to categorize and understand a domain of objects, events, or processes (Cohen & Murphy,
1984). Cognitive linguistics, for that matter linguistics in general, is concerned with categorization on two
levels. Linguistic categories like words, morphemes, nouns, etc that are used to describe the objects of
investigation, and categories which are described by the linguistic categories (Taylor, 1995). Both
disciplines assume that categories and concepts are structured and organized for instance, from simple to
complex and taxonomically (Murphy, 2002).

Cognitive psychology has been successful in formulating and experimenting a battery of models of
theory of concepts and categorization with a key assumption that human’s theories of the world embody
conceptual knowledge and that their conceptual organization is partly represented in their theories. One
model among the theory of concepts is the prototype theory.

1

Prototype view is a product of cognitive psychology of the 1970s mainly due to Rosch’s ground breaking
research of the internal structure of categories (Murphy, 2002; Geeraerts, 1989). It came into existence in
cognitive psychology at the juncture when there was a growing dissatisfaction of the cognitive linguistics
classical –definitional

theory of necessary and sufficient conditions of a category. In the same vein there

was the growing pain and dissatisfaction with the inapplicability of the feature based componential

1

I use model, view and theory interchangeably in my discussion and naming of prototype.

2

analysis

of Chomskian school except for words of the closed class which belong to the same semantic

field or domain, like kinship terms (Geeraerts, 1989). Prototype gave a blow (Murphy, 2002) for both the
classical

and componential analysis theories at the same time. Cognitive psychology has advanced much

since then in experimenting and formulation of alternative models and theories and tackling the main
problem of concept study, i.e., conceptual combination. How ever, cognitive linguistics lag behind in
theorizing and experimenting except using some unsystematic anecdotal cases as evidence. In short,
neither the prototype is properly explored, nor a new theory is adopted and formulated. In actuality,
cognitive linguistics clinges to prototype theory till now as it is considered by almost all students of
cognitive linguistics as one of the three cognitive linguistics’s fundamental tenets with schemas and basic
level categories

(Ungerer & Schmid,1996; Taylor, 1995; 2002; 2004). The initial success of cognitive

linguistics as it manages to bridge the gap between formal syntax and morphology and relate the semantic
aspects of grammar with their common conceptual basis is reported by Ungerer & Schmid (1996).
Prototype, has still persists in a chaotic form despite protests, reviews and cautions raised as early as 1989
(Geeraerts, 1989; Wierzbicka, 1990; Nuyts, 1993). This is partly due to its properties like fuzzy
boundaries, family resemblance, central and peripheral members which give room for throwing out
inadequate analysis to readers. It seems that for every exception in linguistics, prototype is used as a fire
fighter, face saving mechanisms. This creates chaos and confusions for students of linguistics who
recently join the exploration. So the review though, brief help in understanding prototype theory by
comparing and contrasting it with cognitive psychology prototype view of the study of concepts.

This paper focusing on categorization of objects at a simple concept level argues that in cognitive
linguistics prototype application;
(1) There are some misunderstandings of basic tenets of prototype and confusing them with other theory
of concepts.
(2) Prototype is considered as ‘a catch all theoretical construct’.
(3) Prototype are primarily used in lexical semantics for discussing nouns, such as in cognitive
psychology and prepositions in Indo-European languages, and
(4) Unlike in cognitive psychology, prototype has exhibited an extended use and application in areas of
phonology, syntax and diachronic lexical semantics. Specifically in the diachronic lexical semantics, this
effected into introducing new ideas in prototype like paragon, generic neutral, real world versus ideal
world

prototype distinction and what Grzega (2003) calls conceptual recategorization.

To substantiate the arguments proposed selected and representative literature from both disciplines:
Rosch, 1978; Smith & Medin, 1981; Cohen & Murphy, 1984; Murphy & Medin 1985; Komatsu, 1992:
Murphy, 2002 i.e. from cognitive psychology and the writings of Taylor, 1990; 1991; 1994; 1995; 2002;
2003; 2004; Geeraerts, 1989; Wierzbicka, 1990, Newmeyer, 1998 and Grezga, 2003 from cognitive
linguistics have been referred.

(1) Prototype model is a widely used and a partially misunderstood concept in cognitive linguistics due to
overlap with a schema model. From it’s use as a model in the theory of concepts and covering natural
objects and nouns in cognitive psychology, it ascends to the status of a theory and covers a wider
language phenomenon from phonology, syntax, prepositions, semantics to the study of diachronic lexical
semantics in cognitive linguistics. Unlike its wider coverage it lacks methodlogy and procedure of data
gathering and interpretation. In this section I will try to disentagle at least one misuderstanding in the

3

conceptualization of prototype vis-á-vis schema. Reviewing features and charactersitics of prototype and
schema how prototype and schema overlaps will be discussed.
What is a prototype of a category?
One answer is, the clearest cases of category membership defined operationally by people’s judgements
of goodness of membership in the category (Rosch, 1978). She also warned us of the two sources of
confusion in the discussion of prototype. (1) The notion of prototype has tended to beome reified as
though it meant a specific category member or mental structure. (2) The empirical findings about
prototypicality have been confused with theories of processing – that is, there has been a failure to
distinguish the structure of categories from theories concerning the use of that structure in processing.
Moreover, typicality differences for Rosch is an empirical fact of people’s judgements about category
membership.

Geeraerts (1989) states four characterstics of a prototype that are frequently mentioned in the literature.
(1) Prototypical categories cannot be defined by means of a single set of criterial (necessary and
sufficient) attributes. (2) Prototypical categories exhibit a family-resemblance structure, or more
generally, their semantic structure takes the form of a radial set of clustered and overlapping meanings.
3. Prototypical categories exhibit degrees of category membership; not every member is equally
representative for a category.
4. Prototypical categories are blurred at the edges.
Komatsu (1992) has listed five characteristics of prototype under the name of a family resemblance view:
(1) centrality of typicality, (2) abstractness, (3) weighted attributes,(4) independence and additive
combination of weights: linear separability, and (5) Retention of central tendencies. Concerning schema
Komatsu (1992) explicitly states first that it is a hybrid of prototype and exemplar models and about the
relationship between the prototype and schema, that a schema is a single structure that captures
characterstics of both the family resemblance view (by storing information that is abstracted across
instances) and the instances approach (by retaining information about actual instances). He states that
some people used them interchangeabily. Others made a clear distinction between the two and still others
used schema more generally, not suggesting any more than that schemata are networks of associations
that inlude representations of specific instances. Murphy’s (2002) discussion of the relation between
prototype and schema shows where their ties are. One development that is tied to the prototype view is
the use schemata to represent concepts. It is an improvement on the feature list of prototype. It structures
a representation that divides up the properties of an item into dimensions and values on those dimensions.
On the other hand, in cognitive linguistics the relations between the two vary. Langacker, (1987b) stated
the relations as a synthesizing factor of categorization. A schema is an abstract template representing the
commonality of the structures it categorizes, which thus elaborate or instantiate it. Like Murphy,
Langacker also states that a schema differs from a list of criterial attributes in being an integrated concept
in its own right. It is simply characterized with lesser specificity and detail than its instantiations. Where
as for Taylor (1990; 1995; 2002) who claims to follow prototype is an instance of a schema i.e., a subset
of a schema. To conclude the relation between schema and a prototype is a mixed one and leads to
confusion, especially the summary representation of a protoype category and a schematic representation is
really blurred and need clear lines to be used as a model for understanding the relation between categories
and concepts.

(2) Prototype is considered to be a ‘catchall device’

4

Prototype model as a theory of concept doesn’t escape strong criticism from both disciplines. In the circle
of cognitive psychology one thorny problem is it does not have any explanatory mechanism or does not
have an answer on how complex concepts are constructed out of simple(r) concepts (Komatsu,1992). It
also fails to represent intra and inter-concept relations and more general world knowledge (Murphy &
Medin, 1985). Prototype as a similarity-based model does n’t have a mechanism to show coherence and
relationship among features of a category.

In the linguistics circle, prototype due to the historical accident, i.e., matching in time with the
development of cognitive linguistics in the mid 1980s, and the 1970’s publications of Rosch and her
colleagues exclusive use of a language category- noun, was mystfied to cover or said to cover large area
of the discipline. However, two major drawbacks have been mentioned as early as 1990s. First, on the
technical practical side, prototype model in linguistics lacks methodological principles and procedures of
data gathering and interpreting. Except studies on prepositions, there is no clear replicable procedure or
technical guide to test its applicability on a new data. On the other hand, prototype model is considered as
‘a catchall theoretical device’ (Geeraerts, 1989; Wierzbicka, 1990; Hilferty, 1997; Grzega, 2003). Just to
cite some, prototype model has been unjustly enlarged or too effusively used as an explanation before all
relevant aspects (Grzega, 2003). The notion prototype should not be turned into a catchall theoretical
construct (Hilferty, 1997). Wierzbicka (1990) dramatically narrates the ‘prototype save’ attitude of the
believers. She states that, in semantics, the actual usage of words is too messy, too unpredictable to be
accounted for by definitions. But fortunately, semanticists don’t have to worry about it any longer: they
can now deploy the notion of prototype for all residues and unsolved problems. She, instead, suggests the
usefulness of the prototype model as a specifc, probably powerful analytical tool, and not as ‘a universal
thought-saving device’. In other words it is a useful tool for semantic description but not a useful
fullfledged semantic theory by its own. It has been observed in cognitive psychology literature that
prototype model has no inbuilt mechanism to handle relations and strength or weight of features or
combinations.

Eventhough there has been strong criticisms of the adherents misuse and sloppiness and weaknesses of
the prototype model, it is observed as a very productive descriptive device for word classes especially for
nouns in cognitive psychology and prepositions in cognitive linguistics.

(3) Like in cognitive psychology prototype is primarily used in linguistics for the study of lexical
semantics specifically of nouns and applied further to preposition studies.

The 1970s literature of cognitive psychology primarily that of Rosch, (1973; 1975a; 1975b; 1975c;1978);
Rosch & Mervis, (1975); Rosch et al, (1976) among others , in addition to sealing off effectively the
classical view

from serious considerations (Murphy, 2002) the literature heralds a new era of using

experimental results to back up models of theory of concepts. In the words of Smith & Medin (1981), in
addition to identification of people’s judgement of categories, models, built based on the results, bridge
the direct results of categorization to conceptual representation and process. We will also use the models
for the interpretations of the direst experimental results. According to Smith & Medin, this point has been
overlooked in recent research on natural concepts, by the assumption taken that categorization data
directly inform us about the nature of concepts.

5

Moreover, the 1970s contents of the experiments are categorization and conceptualization of basic objects
in natural catgegories, cognitve reference points and cognitive represenations of semantic categories. All
these boil down to one, nouns. The contents of experiments are nouns, internal structure of nouns and
cognitive representations of nouns. These seminal publications on nouns, taxonomy of nouns and internal
structure of nouns i.e. publications on language internal issues gave the prototype model, an unconditional
acceptance in linguistics. Prototype model and its back up experimental research also lay the framework
for the emergence of cognitive linguistics.

Prototype is not only used as a model or analytical device, but it is considered as one of the three basic
principles of linguistic categorization with basic level concepts and schema (Taylor, 1985 (1995); 1990;
2002). Taylor, in turn, relies on Langacker (1987b). Langacker’s article not only states that categorization
by prototype and schemas are the bases of his theory- cognitive linguistics, but also two more issues of
concern here. I suspect these issues of concern resulted from cognitive psychology experimental
prototype theory publications on nouns and noun concepts.

1. Nouns have prototypical features and representations hence physical objects are prorotypical
nouns.
2. Nouns lend themselves to schematic semantic characterization.

The two concerns clearly show the extension of the prototype model to cover not only a group of nouns
that belong to the same taxonomic semantic class, but the whole class of nouns from the very inception
of the prototype model. In addition to barring other models like Exemplar, Theory Theory or Simulated
Situation Theory

, nowadays Prototype model is not only an uncommon to but has a wide currency in

linguistic publications. To give the practical feel of how prototype is used, I quote Givón (2001) verbatim.

1. The prototype of the class noun occupies that most time-stable end of the scale.
2. One important reason for the great temporal stability of prototype nouns is that they are multi-
featured

bundles of experience. Further, like all natural categories, they exhibit strong feature –

association

.

3. The time-stability of prototype nouns also owes much to the fact that they are concrete, and
made out of relatively-durable materials.
4. The fact that prototype nouns tend to be spatially compact rather than scattered all over the
perceptual space is in part a consequence of their spatial-coherence.
5. One important consequence of prototype nouns being compact, is that they also tend to be
relatively small, thus figures occupying a small portion of the total perceptual field or
ground

…prototype nouns tend to be countable, as against non-prototypical mass nouns (‘rain’,

‘love,’ ‘water’, ‘sand’, ‘freedom’)

To conclude another highly productive and successful area in prototype theory application is the study of
prepositions (Colombo & D’Arcais, 1984; Kishner & Gibbs, 1996;Myers, 1994;Rice, 1996; 2003; Rice,
Sandra & Vanrespaille, 1999; Sandra & Rice, 1995; Tyler & Evans,2001). These prepositions studies
range from English to other Indo-European languages. Chinese (Myers, 1994) is also observed. What
characterizes most of these prepositional studies, unlike the phonological, syntactic or their noun
counterparts, is their successful attempts in following cognitive psychology methodology. They follow
systematic data collection, experimentation, interpretation and presentation of data.

6

Albeit on a small scale, prototype model has been applied at other levels of linguistic catgeories,
phonology, syntax, semantics and diachronic lexical semantics. Especially the diachronic lexical
semantics literature show the introduction of nuances in the prototype model. Some new senses of
meaning shift and conceptual recatgorization have been observed.

(4) Prototype has exhibited an extended use and application in phonology, syntax and diachronic lexical
semantics. This effected into introducing new ideas in prototype theory like paragon, generic neutral,
real world

versus ideal world prototype distinction and what Grzega (2003) calls conceptual

recategorization

.

Taylor (1995) devoted half of the book reviewing the role and application of prototype theory in category
extension through the process of metaphor and metonymy, past tense, intonation, syntactic constructions
like the possessive genitive, acquisition of grammatical categories, phonemes and syllable constructions
in addition to nouns. Murphy’s (2002) summary best represents what can be said about the book and the
prototype model Taylor adheres to and the effect of linguistic based prototype study. I quote him
verbatim.

Taylor (1995) has written an entire book on typicality effects in language (mostly in semantics,
but also in phonology and syntax).(This is in fact a very useful introduction to category structure
from a linguistic perspective. It is especially good at discussing cases other than noun concepts,
which are almost the only topic in the psychological literature). All these examples provide
evidence for the view that word meaning is represented in the conceptual system.

I observe the prototype concept and its features as a very useful analytical model and highly effective for
word class studies. The Indo-European languages preposition studies and other word classes enable one to
suggest that it has more value in word meaning study than the futile attempts at phonology and syntax.
The two subfields of linguistics do require explanation and better be handled by the theory theory than the
prototype model

. This is also compounded by for a simple methodological fact that phonological and

syntactic studies that involved prototype model did not exhibit a replicable research procedures. The data
they did use are handpicked and anecdotal. By virtue of their limited size and representing categories,
closed class lexical items, or words that belong to the same semantic field can be best studied as they can
overcome the anecdotal, selection or sampling problem as all members of the class are used. The worst
problem of using anecodotal data is the absence of systematic handling of the category or the concept
issue that underlie them. Cognitive psychologists like Rosch (1978), however, unequivocally argued that
human categorization should not be considered as an arbitrary historical accident, but it has to be the
result of psychological principles of categorization. Like any other research it has to be subjected to
investigation. The anecdotal evidence of linguistics in general does not seem to fit to this basic principles
of categorization and science i.e. the studies are not substantiable or replicable.

Diachronic lexical semantics focuses on comparing meaning change of a word or words of the same
taxonomic class through time. Meaning change of a word in one way or another is related to the concept
that the word represents. Hence, meaning change of a word may reflect conceptual or catgorization
change. Diachronic lexical semantics (alternately called historical lexicology) adopts prototype model to
understand and explain which features of a concept is affected when a meaning change occur in its
representation – a word.

7

Grzega (2003) identifies in his review article, details of prototype phenomena in the diachronic lexical
semantic field while observing a tendency of prototype overuse by the practitioners. Grzega observes at
least refinement of five senses or development of new definitions of prototype in meaning change or shift
of a word within its taxonomic class i.e. in relation to superordinates, coordinates (co-hyponyms) and
subordinates (Murphy, 2002): paragon, generic neutral, distinction of real world versus ideal world
prototype and conceptual recategorization.

Briefly, paragon refers to the highest valued member of a category (a minimal lexical class) in one’s
‘ideal’ (cognitive) world which need not necessarily refer to a category’s most typical member in the real
world; ‘meat’ replaces ‘victual’ as ‘a highly valued food’. A member of a category that accepts the
highest number of contexts which occur most in different contexts is labeled as a generic neutral i.e. it
functions or acquires the new function of representing the lexical class like a superordinate. French ‘rue’
covers wider contexts than it did in the past (see Grzega, 2003 for details). Real world prototypes and
ideal world (idealisitc)

prototypes are fairly transparent. When the category has a single prototype

member, that member may have an extension of meaning for new contexts. In the real-world prototype
case, the French chevaucher means ‘to ride on a horse’. It has now an extended meaning, and
‘chevaucher’ means ‘to ride any kind of animal’. An ideal- world example is the Gallo-Latin
‘formaticus’. It referred previously as ‘formed cheese’, now it stands for (all types of) ‘cheese’.

Grzega labeled the phenomenon conceptual recategorization where there is a reorganization of the entire
conceptual field. According to Grzega it can be any aspect of the prototype. As he says it may be
connected to a change in (the salience of) things, a change in the salience of specific aspects, or some sort
of lexical and/or conceptual fuzziness. Latin passer referring to ‘sparrow’ changes to the Rumanian
language pasare refering to ‘bird’. Old Englsh bryd ‘young bird’ to Modern English bird. In conclusion,
Grzega has reported that

BIRD

underwent a conceptual recategorization as speakers shifted the borderline

away from the size distinction. This, according to Grzega may be due to fuzziness of the feature size
within the internal structure of the category

BIRD

.

Conclusion

This review is the first reflection to undersatnd the full breadth and length of prototype theory in
linguistics. Despite using prototype theory literature from cognitive psychology and following one or
another defintion or features of prototype model, this paper pinpointed two trends in the four arguments.
On the one hand, there is a misunderstanding and overuse of prototype theory. On the other hand there is
an extension or refinement of prototype model applications from the original study of noun concepts in
cognitive psychology to at least four areas of linguistics (preposition, phonology, syntax and diachronic
lexical semantics). By focuing on these two main trends, the paper discussed how categorization in terms
of prototype and schema have been understood, conceptualized, explicated, empirically supported and
extended vis-á-vis cognitve psychology.

Similarity-based theories of conceptual representations (Murphy, 2002; Komatsu, 1992) and
categorizations (Ungerer & Schmid, 1996; Nuyts, 1993; Deane, 1996) i.e., prototype and schema have
been observed since the 1970s in cognitive psychology and 1980s in cognitive linguistics (Lakoff, 1987;

8

Langacker, 1987a; Taylor, 1989 (1995). From a cursory look at the literature and the references compiled
and cited by the practitioners, it is clear that the study of categorization in cognitive psychology is
experiment based and on the lead. Linguists that ascribe to cognitive paradigm use these theoretical
concepts liberally as a catchall theoretical construct (Wierzbicka, 1990;Hilferty, 1997). Despite having
the same sources, the uses and interpretations of these concepts show differences in the two cognitive
disciplines.

There is also among linguists a growing awareness and uneasiness (Wierzbicka, 1990; Nuyts, 1993;Deane
1996) about the murkiness of the proper usage theories of concepts and unfamiliarity (Deane, 1996) with
the new models and theories of the related disciplines. Moreover, Nuyts (1993) noted the lack of good
will to account more of the considerable body of cognitive psychology literature on knowledge
representation and conceptualization by the cognitive linguistics proponents. In fact, he made an appeal in
his review of Langacker (1987a) and Lakoff (1987) to these two theoreticians, who originally formulated
cognitive linguistics theory, as their disciples follow them sheepishly.

Murphy (2002) noted the typicality effect of a la Rosch has been a predominant trend pursued in cognitive
linguistics as in Taylor (1995). There is also a commendable use of prototype theory from the study of
noun concepts to the study of semantics, prepositions, phonemes and syllable structures. What Murphy
(2002) did not mention the Indo-European languages concentration in the studies. With regard to focuing
on Indo-European languages, recently Medin & Altran (2004) have reported the importance of cross-
cultural research on identification of people’s conceptualization of objects and have noted the inadequacy
and failure of generalization to humanities at large based on research findings gathered from standard
populations in industrialized societies only.

Linguists like Geeraerts (1989); Wierzbicka (1990); Nuyts (1993) and Deane (1996) outcry and incisive
comments about prototype theory’s abuse and limitations seem to lead the proper use of prototype theory
in cognitive linguistics no where (cf. for opposite trends Taylor, 1995; 2002) with the exception like
Grezga’a (2003) work on diachronic semantics and the studies on prepositions in general. Those who
focus on diachronic lexical semantics or meaning changes and shifts not only identified the limitations or
prototype theory but also they have provided new elements, new concerns and new additions to prototype
theory, a focus on the semantics of lexical changes- which is also related to recategorization of concepts
in some cases.

References

Cohen, B. & Murphy, G.(1984). Models of concepts. Cognitive Science 8, 1-26.
Colombo, L. & D’Arcais, G. (1984). The meaning of Dutch prepositions: a psycholinguistic study of

polysemy. Lingustics 22, 51-98

Deane, P.(1996). On Jackendoff’s conceptual semantics.Cognitive Linguistics 7-1, 35-91.
Dirven, R. & Taylor, J.R. (1988). The conceptualization of vertical space in English: The case of T

ALL

. In

Rudzka-ostyn, Brygida (Ed.). Topics in Cognitive Linguistics. Amsterdam/ Philadelphia: John Benjamins. PP.379-
402.

Geeraerts, Dirk.(1989). Introduction: Prospectus and problems of prototype theory. Linguistics 27-4,587-

612.

Geeraerts, Dirk.(1992).Review article: Polysemy and prototypicality. Cognitive Linguistics 3-2, 219-231.
Givón, T.(2001).Syntax.Vol.I.Amsterdam/Philadelphia: John Benjamins.
Grzega, Joachim.(2003).On using (and misusing) prototypes for explanations of lexical changes. Word 54-

3,335-357.

9

Hilferty, Joseph. (1997). Mother, lies, and bachelors: A brief reply to Wierzbicka (1990). Word 48-1, 51-

59.

Jodlowiec,M.& Kwasniewicz, K.(1991). Review of John R. Taylor: Linguistic Categorization: Prototypes

in Linguistic Theory’. Cognitive Linguistics 2-3, 283-309.

Kishner, J. & Gibbs, R. (1996). How “Just” gets its meanings: Polysemy and context in psychological

semantics. Language & Speech, 39-1,19-36.

Komatsu, Llyod.(1992). Recent views of conceptual structure. Psychological Bulletin, 112-3, 500-526.
Lakoff, George.(1987). Women, Fire and Dangerous Things: What Categories Reveal about the Mind.

Chicago: The University of Chicago Press.

Langacker, Ronald.(1987a). Foundations of Cognitive Grammar: Theoretical Prerequisites. Vol.1.Stanford:

Stanford University Press.

Langacker, Ronald.(1987b). Nouns and verbs. Language, 63-1,53-94.
Medin, D.& Atran, S.(2004). The native mind: Biological categorization and reasoning in development and

across cultures. Psychology Review, 111-4,960-983.

Mondry, Henrietta & Taylor, John R.(1992). On L

YING

in Russian. Language and Communication 12-2,

133-143.

Murphy, Gregory.(2002). The Big Book of Concepts. Mass, Cambridge: The MIT press A Bradford book.
Murphy, G. & Medin, D. (1985). The role of theories in conceptual coherence.Psychological Review,92-

3,289-316.

Myers, D. (1994).Testing for prototypicality: The Chinese morphem gong, Cognitive Linguistics,5-3, 261-

280.

Nuyts, Jan.(1993). Cognitive linguistics. Journal of Pragmatics 20, 269-290.
Rice, S. (1996). Prepositional prototypes.In Pütz, M & Dirven, R. The Construal of Space in Language and

Thought. Berlin /New York: Mouton de Gruyter.PP.135-165.

Rice, S. (2003). Growth of a lexical network: Nine English prepositions in acquisition.Cuyckens, H. et al

(Eds.).Cognitive Approaches to Lexical Semantics. Berlin /New York: Mouton de Gruyter .PP.245-280.

Rice, S.,Sandra, D. & Vanrespaille, M. (1999).Prepositional semantics and the fragile link between space

and time.In Hiraga, M. et al (Eds.). Cultural, Psychological and Typological Issues in Cognitive Linguistics.
Amsterdam/ Philadelphia: John Benjamins. PP.107-127.

Rosch, Eleanor.(1973). Natural Categories. Cognitive Psychology 4,328-350.
Rosch, Eleanor.(1975a). Cognitive representations of semantic categories. Journal of Experimental

Psychology: General 104-3, 192-233.

Rosch, Eleanor.(1975b). Reply to Loftus Rosch’s cognitive representations of semantic categories. Journal

of Experimental Psychology: General 104-3,241-243.

Rosch, Eleanor.(1975c). Cognitive reference points. Cognitive Psychology 7,532-547.
Rosch, Eleanor.(1978).Principles of categorization. In Rosch, E. & Lloyd, B. (Eds.).Cognition and

Categorization. Hillsdale/ N.J. Lawrence Erlbaum.PP.27-48.

Rosch, Eleanor and Mervis, Carolyn.(1975). Family resemblances: Studies in the internal structure of

categories. Cognitive Psychology 7, 573-605.

Rosch, E. et al.(1976).Basic objects in natural categories. Cognitive Psychology 8, 382-439.
Sandra, D. & Rice, S. (1995).Network analyses of prepositional meaning: Mirroring whose mind-the

lingust’s or the language user’s? Cognitive Linguistics 6-1,89-130.

Smith, E.1990. Categorization.In D. Osherson & E. Smith (Eds.).Thinking: An Invitation to Cognitive

Science. Cambridge/ London:The MIT Press. VOL.3,33-53.

Smith, E. & Medin, D.(1981). Categories and Concepts. Cambridge/ London: Harvard University Press.
Taylor, John R. (1988). Contrasting prepositional categories: English and Italian. In Rudzka-ostyn, Brygida

(Ed.). Topics in Cognitive Linguistics. Amsterdam/ Philadelphia: John Benjamins Publishing Company. PP.299-
326.

Taylor, John R. (1989). Possessive genitives in English. Linguistics 27-4, 663-686.
Taylor, John R. (1990). Schemas, prototypes, and models: in search of the unity of sign. In Tsohatzidis,

S.L.(ed.).Meanings and Prototypes: Studies in Linguistic Categorization. London/ New York: Routledge.PP.521-
534.

Taylor, John R.(1991). Things, places, and directions (squib). Cognitive Linguistics 2-4, 357-360.
Taylor, John R.(1992).Old problems : Adjectives in cognitive grammar. Cognitive Linguistics 3-1, 1-35.

10

Taylor, John R.(1993).’ Some pedagogical implications of cognitive linguistics’. In Geiger, Richard A &

Rudzka-Ostyn, B (Eds.). Conceptualizations and Mental Processing in Language. Berlin/ New York: Mouton de
Gruyter. PP.

Taylor, John R.(1994a). “Subjective” and “objective” readings of possessor nominals. Cognitive

Linguistics 5-3, 201-242.

Taylor, John R. (1994b). Possessive and Topicality. Functions of Language 1-1:67-94.
Taylor, John R.(1994c). The two level approach to meaning. Linguistische Berichte 149, 3-26.
Taylor, John R.(1994d).Fuzzy categories in syntax: The case of possessives and compounds in English.

Revista di Linguistica 6-2,327-345.

Taylor, John R. (1995). Linguistic Categorization: Prototypes in Linguistic Theory (2

nd

ed.). New York:

Oxford University Press.

Taylor, John R. (1996a). On

RUNNING

and

JOGGING

. Cognitive Linguistics 7-1, 21-34.

Taylor, John R.(1996b).Possessives in English; An Exploration in Cognitive Grammar. Oxford: Clarendon

Press.

Taylor, John R. (2002a).Near synonyms as a co-extensive categories:

HIGH

and

TALL

revisited. Language

Sciences 25,263-284.

Taylor, John R.(2002b).Cognitive Grammar. Oxford: OUP.
Taylor, John R.(2003a).Polysemy’s paradoxes. Language Sciences 25,637-655.
Taylor, John R. (2004). The ecology of constructions. In Radden, Gηnter and Panther, Klaus-uwe (Eds.).

Studies in Linguistic Motivation. Berlin/ New York: Mouton de Gruyter.PP.49-73.

Taylor, John R, Kasabov, Nikla & Kilgour, Richard.(2000). Modeling the emergence of speech sound

categories in evolving connectionist systems. The information science discussion paper series.

www.divcom.otago.ac.nz/infosci/publctus/complet..downloaded on 23/02/05

Taylor, John R, Cuyckens, H. & Dirven, Rene.(2003). Introduction: New directions in cognitive lexical

semantic research. In Taylor, John R, Cuyckens, H. & Dirven, Rene (Eds.) Cognitive Approaches to Lexical
Semantics. Berlin/New York: Mouton de Gruyter.PP.1-28.

Tyler, A. & Evans, V.(2001). Reconsidering prepositional polysemy networks; The case of over.Language,

77-4,724-765.

Ungerer, Fridrick and Schmid, Hans-Jorg.(1996).An Introduction to Cognitive Linguistics. Harlow:

Addison Wesley Longman Ltd.

Wierzbicka, Anna.(1990). Prototype save: On the uses and abuses of the notion of ‘prototype’ in linguistics

and related fields’. In Tsohatzidis, Savas (Ed.). Meanings and Prototypes: Studies in Linguistic Categorization.
London:Routledge.PP.347-367.

Winters, Margaret.(1989). Diachronic prototype theory: On the evolution of the French subjunctive.

Linguistics 27-4,703-730.