Polysemy and metaphor in perception verbs: a cross-linguistic study

your hands and This matter is in your hands, the link is carried out by means of 
4
See Lakoff (1987: Ch.2) and Taylor (1995: Ch.3.2.) for discussion on early development 
research into categorisation. See Ungerer and Schmid (1996) for an overview on ‘Prototypes and 
Categories’ (Ch.1) and ‘Levels of Categorisation’ (Ch.2). 
5
i.e. the central-prototypical sense. 

---

B. Iraide Ibarretxe-Antuñano 
Polysemy and metaphor in perception verbs 
21
metaphor; whereas in They are taking new hands, the link is metonymical (part for 
whole). 
Another consequence of the primacy of cognitive abilities is that there is no strict 
distinction between encyclopaedic and linguistic knowledge. Objectivists differentiate 
between these two different epistemological types of knowledge. On the one hand, 
‘linguistic’ or ‘definitional’ knowledge that “corresponds to the essential properties of 
the entities and categories that the words designate”; and on the other, ‘encyclopaedic’ 
knowledge “corresponds to the contingent properties of the entities and properties that 
the words designate” (Lakoff 1987: 172). This dictionary-encyclopaedia distinction 
leads objectivists to postulate a ‘meaning per se’ (Leech 1981: 70), independent of 
whatever the speaker may know about the states of affairs that he is referring to. This 
paradigm also induces the distinction between literal (objectively true or false) and 
figurative meaning (no direct correspondence to entities and categories in the real 
world). 
For Cognitive Linguistics, however, this distinction is not strict. Meanings are 
cognitive structures embedded in our patterns of knowledge and belief; conventional 
meanings arise from experience and knowledge and our complex conceptual structures 
are invoked in language use and comprehension
6
. The fact that our experience-based 
knowledge is present in linguistic meaning at every level implies that there is not a strict 
distinction between lexicon and grammar
7
, between semantics and pragmatics, between 
synchrony and diachrony. This is possible because the same social, functional and 
cognitive motivation present in historical changes is also observable in ongoing changes 
(Barcelona 1997: 11). 
This continuum between language and experience has led cognitive linguists to 
study how conceptual structures or cognitive models are reflected in language. 
According to Langacker (1987: 147ff.), most concepts invoke other concepts and 
without making an explicit reference to them, they cannot be adequately defined. 
Consider for instance the concept of [M
ONDAY
]. If we ask ourselves about the meaning 
6
That is why meaning is claimed to be ultimately pragmatic. 
7
As Langacker (1987: 3) states “Lexicon, morphology, and syntax form a continuum of symbolic 
structures, which differ along various parameters but can be divided into separate components only 
arbitrarily”. 

---

B. Iraide Ibarretxe-Antuñano 
Polysemy and metaphor in perception verbs 
22
of the word Monday, we will probably say that it is a day of the week; but again what is 
the meaning of [
WEEK
]? Weeks are imaginative creations of the human mind. The kind 
of imaginative structure required for the description of concepts such as [M
ONDAY
] are 
what Langacker (1987: 150) calls ‘abstract domains’: “any concept or conceptual 
complex that functions as a domain for the definition of a higher-order concept”. These 
abstract domains are equivalent to Lakoff’s (1987) ‘Idealised Cognitive Models’ (ICMs) 
and Fillmore’s (1982, 1985) ‘frames’. These abstract domains give structure to what 
Langacker (1987: 148) refers to as ‘basic domains’, i.e. primitive representational fields, 
not reducible to another; they occupy the lowest level of conceptual complexity. These 
basic domains are what Lakoff (1987: 281), following Fauconnier’s (1985) terminology, 
calls ‘mental spaces’
8
, mediums for conceptualisation and thought. In this case, the basic 
domain of the concept of [
WEEK
] would be [
TIME
]. 
In some cases, one abstract domain or ICM on its own cannot define the meaning 
of words. The latter may need the characterisation of several ICMs simultaneously; this 
is what Lakoff (1987:74) calls ‘cluster models’ (or domains).
Another consequence of this primacy of general cognitive abilities is the essential 
role of imagination. For many people, the word imagination is related to subjectivism, 
idealism, and relativism. Since the Enlightenment
9
, imagination has been despised in 
many theories of language, because it is regarded as a non-rational, unruly, and 
idiosyncratic play of ideas, and therefore, unsuitable for scientific research. In Cognitive 
Linguistics, imagination is considered a basic human cognitive ability, central to human 
meaning and rationality. As Johnson (1987: 172) explains, the way we reason and what 
we can experience as meaningful are both based on structures of imagination that make 
our experience what it is. We make sense of our less directly apprehensible experiences 
on the basis of more directly apprehensible experiences. For instance, Lakoff and 
Johnson (1980: 14ff.) have shown how we project part of our bodily experience of three-
dimensional space onto our experience of happiness, when we say My spirits rose; or 
onto our experience of sickness and death, as in He came down with flu. 
8
‘Mental spaces’ are “constructs distinct from linguistic structures but built up in any discourse 
according to guidelines provided by the linguistic expressions” (Fauconnier 1985: 16). 
9
See Johnson (1987: Ch. 6) for an account on the history of Imagination. 

---

B. Iraide Ibarretxe-Antuñano 
Polysemy and metaphor in perception verbs 
23