Content uploaded by Chariton Charitonidis
Author content
All content in this area was uploaded by Chariton Charitonidis on Oct 12, 2014
Content may be subject to copyright.
ProofsProofs
Colour verbs in Modern Greek: A cognitive
approach
1
Chariton Charitonidis
Abstract
This paper gives an outlook on how the coordination between phonology and
semantics in verb derivation may look, and which constraints the distinction between
possible/established, possible/novel, and not possible/forced verbs can be mapped
onto. The object of investigation are the colour -ı
´zo verbs in Modern Greek. The
analysis is mainly based on the studies by Berlin & Kay (1969), Kay & McDaniel
(1978), and Kay & Maffi (1999) on basic colour terms. The colour verbs in Modern
Greek are compared with the English colour verbs. It is shown that systematic gaps
linked to the causative/inchoative paradigm of -ı
´zo verbs are due to the semantics
of the base rather than deficiencies in the base’s morphophonological character. The
lower the evolutional stage that a base can be fitted to, the more likely an -ı
´zo derivative
will be produced.
1. Introduction
2
Phonology and semantics are closely related in the study of derivation, cf. the analysis
of English (EN) verb derivatives in Plag (1999, 2003). However, we are confronted
with a chicken-or-the-egg dilemma if we seek to assess semantics or morphophonology
as the determinants of derivation: does phonology ornament specific semantic spectra
which were previously (fully) expressed by means of derivation, or are specific
semantic spectra the artefact of the attested morphophonological patterns? Without an
independent cognitive theory it is very difficult to decide the issue: this theory should
refer to different perceptual modalities, e.g. vision, audition, sensorimotor, etc. so that
the cognitive component can be directly related to the derivational mechanisms of the
language.
3
Only in this way we can finally know whether specific semantic spectra are
regularly expressed (or exhausted). If not, then the morphophonological component
must be regarded as more relevant.
Word Structure 7.2 (2014): 125–152
DOI: 10.3366/word.2014.0063
#Edinburgh University Press
www.euppublishing.com/word
PLEASE CONSULT THE FINAL VERSION BEFORE QUOTING
ProofsProofs
Ralli (2005: 157) argues that in word formation ‘though the structural [e.g. blocking,
A/N] and semantic restrictions do not have exceptions in general, the other restrictions
[i.e. lexical constraints such as –native base, etc., A/N] can be violated for different
reasons’. In line with Ralli’s assessments, I accept semantic restrictions as the most
powerful, to the extent that they sufficiently explain possible/established, possible/
novel, and/or not possible/forced formations. This study will address this three-stage
paradigm with focus on verb derivation.
The object of investigation are the colour -ı
´zo verbs in Modern Greek (MG).
4
In
the following, I give a brief morphological description of the -ı
´zo derivatives and of the
COLOUR subclass.
According to Efthymiou (2011) the MG suffix -ı
´zo developed mainly from the
Ancient Greek (AG) suffixes -ı
´zdo
¯and -o
¯(stressed). In MG, the vast majority of -ı
´zo
derivatives are derived from nouns. Phonologically, -ı
´zo attaches primarily to
consonant-final bases.
5
Charitonidis (2005) examined -ı
´zo derivatives on a semantico-syntactic basis. In this
study -ı
´zo shows up with six lexical conceptual structures (LCSs; cf. Jackendoff 1983,
1990, etc.), whereby the derivation base refers to a great variety of thematic roles, i.e.
agent, theme, goal, instrument/means, etc. (see Charitonidis 2005: 59, 2007: 32). This
versatility of -ı
´zo results into its high productivity (see Charitonidis 2007: 35).
6
Colour
verbs in -ı
´zo are allocated to the distinctive alternation class 1a/b_9, i.e. causative
active / auto active
7
_passive participle. The thematic role of the base is always GOAL,
cf. kitrinı
´zo ‘become yellow’ (kı
´trino ‘yellow’).
In order to address the chicken-or-the-egg dilemma mentioned at the beginning
of this section we need a semantic framework onto which the morphophonological
patterns of the MG colour verbs can sufficiently be mapped. Accordingly, the
framework used in this paper mainly refers to the studies by Berlin & Kay (1969), Kay
& McDaniel (1978), and Kay & Maffi (1999) and is presented in section 2. Section 3
deals with the study by Androulaki et al. (2006) on basic colour terms (BCTs) in MG. In
this section, the MG BCTs are presented and the problematic status of the BLUE category
is discussed. Section 4 deals with the linguistic relevance of colour studies. In
particular, in section 4.1 the usefulness of predefined stimuli in colour studies is
underpinned. Section 4.2 underscores the need for examining colour categories in
different grammatical categories. Section 4.3 discusses the role of colour in natural
language semantics with special reference to pragmatic operations in verbal derivation.
Section 5 deals with the colour verbs in MG. In particular, section 5.1 presents the
general patterns of the MG colour terms and their assignment to evolutional stages. The
special order of the five terms for blue in the stage sequence is discussed. Section 5.2
refers to the chronology and use of the MG colour terms. As regards the -ı
´zo verbs, a
decline in use from the lower to the higher stages is evident. Section 5.3 deals with the
morphophonology of the MG colour adjectives and their verbal derivatives. The
tripartite morphological pattern of derivation bases is linked to the evaluation of the
verbal derivatives as ‘possible/established’, ‘possible/novel’, or ‘not possible/forced’.
In section 5.4 five exceptional verbs in the evolutional sequence of the MG colour terms are
discussed, with reference to historical, cultural, and/or morphophonological factors.
126 C H A R I T O N C H A R I T O N I D I S
ProofsProofs
Section 6 scales the results presented in section 5 in the domain of EN verbal derivation.
In particular, section 6.1 deals with the chronology and use of the EN colour terms. As
in the case of MG, the use of the EN verbal derivatives drops from the lower to the higher
stages. Section 6.2 examines the morphophonology of the EN verbal derivatives whereby
exceptional cases are linked to specific thresholds in the evolutionary sequence of BCTs.
Section 6.3 compares the morphophonogy of the EN with the MG colour verbs and
provides evidence for an asymmetrical levelling in the respective languages. Section 7
concludes that both BCTs and derived colour verbs emerge over stages in a cognitive
domain. Systematic gaps linked to the causative/inchoative paradigm of -ı
´zo verbs are
due to the semantics of the base rather than deficiencies in the base’s morphophonological
character. The lower the evolutional stage that a base can be fitted to, the more likely an
-ı
´zo derivative will be produced.
2 Framework
Berlin & Kay’s (1969) account on BCTs is a seminal work on the relation between form
und meaning in language. In the following I largely comply with the comprehensive
description of this study in Lo
¨bner (2002: 163–167).
By studying 100 languages from all over the world, Berlin & Kay (1969) found that
languages vary in the number of BCTs they have from two to eleven. In (1) the
temporal-evolutionary ordering of these eleven BCTs is given, taken from Berlin & Kay
(1969: 4). The possible types of systems form a sequence – starting with a contrast
between WHITE/WARM and BLACK/COOL – in which the higher systems are extensions
of the lower systems (Lo
¨bner 2002: 166–167).
(1)
wh
bl
I II
hite [red
lack
IIIa/IIIb
d] [green]
[yellow
b IV
[yellow]
w] [green]
V
[blue]
VI
[brown]
VII
purple
pink [B
orange
g
re
y
Berlin & Kay 1969: 4 ]
In Berlin & Kay (1969) a set of eleven focal colours for the BCTs in all the different
languages are stipulated. The focal colours are white, grey, black, red, orange, yellow,
green, blue, purple, pink and brown. These are the foci of the EN BCTswhite, grey,
black, red, orange, etc. and of the corresponding words in all other languages with
eleven BCTs. If a language has fewer BCTs, their foci are nonetheless among these eleven
focal colours.
An important development of Berlin & Kay’s (1969) model is the distinction
between ‘primary’ and ‘derived’ BCTs in Kay & McDaniel (1978): ‘Primary terms are
the first six terms on the hierarchy and their best examples are “perceptual primitives”
underpinned by “fundamental neural responses”. Derived BCTs are fuzzy set
intersections of pairs of primaries . . . For instance ORANGE is RED-YELLOW and
PURPLE is RED-BLUE.’ (Kay & McDaniel 1978; reported in Androulaki et al. 2006: 5).
As Al-Rasheed et al. (2011) note, Kay & McDaniel’s (1978) conjecture about the
intersectionality of the derived BCTs turned out not be true (cf. Jameson & D’Andrade
C O L O U R V E R B S IN MO D E R N GR E E K :A CO G N I T I V E AP P R O A C H 127
ProofsProofs
1997, etc.), but the greater prevalence of primary over derived terms across languages
holds. By examining languages from 110 non-industrialized societies, Regier et al.
(2005) show that best-example choices for colour terms cluster near the prototypes for
EN white, black, red, green, yellow, and blue, and (ii) best-example choices cluster more
tightly across languages than do the centers of category extensions.
Kay & Maffi (1999) undertake a number of important revisions to Berlin & Kay’s
(1969) model, which comply with the current state-of-the-art (Biggam 2012: 85). In (2)
a summary of these revisions can be found, taken from Stanulewicz & Pawłowski
(2011: 107). ‘Macro-white’ and ‘macro-black’ refer to light and dark colours,
respectively. ‘Macro-red’ refers to more than one of the warm colours such as orange,
red, and yellow, and ‘grue’ refers to more than one of the cool colours, i.e. green, blue,
and grey.
8
As explicitly stated in (2), in Kay & Maffi (1999) grey shows up as a wild-
card occurring at any stage from III to VII. This information is relevant for the analysis
to follow. The evolutional route as regards the primary BCTs is displayed in Figure 1.
9
(2) Stages Berlin & Kay (1969) Kay & Maffi (1999)
I white and black macro-white and macro-black
II addition of red separation of macro-red and white
III addition of green or yellow separation of grue and black
IV green and yellow separation of yellow and red
V addition of blue separation of blue and green
VI brown
VII pink, purple, orange and grey pink, purple and orange, with
grey as a wild-card occurring at
any stage from III to VII
Let us now see how the research on MG BCTs fits the findings presented in this section.
3 Basic colour terms (BCTs) in Modern Greek (MG) and the study by
Androulaki et al. (2006)
The most profound survey on MG colour terms is the study by Androulaki et al. (2006).
In this survey, four naming studies with varying stimuli, lighting, instructions, and
informants were carried out. The main aim of the experimenters was to establish
Figure 1. Main line of evolutionary development of basic colour lexicons (Kay & Maffi 1999:
750).
128 C H A R I T O N C H A R I T O N I D I S
ProofsProofs
the BCTs of MG with particular consideration of the status of the two blue terms,
ble ‘blue’ and ghala
´zio ‘light blue’. In the following, I give a shortened version of
Androulaki et al.’s (2006: 7–10) outline of the experiments.
As regards measures, Androulaki et al. (2006) first analysed frequency of colour terms
in texts as an indicator of basicness (Study 1). It was expected that BCTs would occur
more often than non-BCTs, and primary terms more often than derived terms. In
addition, they conducted four naming studies from which they derived a range of
measures of salience as further indicators of basicness (Studies 2–5). In all naming
studies they reported two measures for each term: frequency of use across informants
and frequency of use across colour samples. The latter measure was used to examine
levels of agreement among informants. BCTs should be used frequently and with high
levels of agreement, particularly for the best example of a term.
The experimenters used various operational measures to locate the foci of BCTs. For
instance, in Study 3, tiles named quickly and with absolute agreement across
informants were taken as focals, etc.
As regards instructions, Androulaki et al. (2006) varied the demand characteristics of
their tasks, either explicitly, through instructions, or implicitly, by varying the range of
stimuli. Instructions were to: use simple everyday terms (Studies 2–3); name the
stimuli with no restrictions (Study 4); use simple necessary terms (Study 5). It was
expected that the use of BCTs would increase as the restrictions increased and that BCTs
would occur in compound terms in the least restricted condition. The two restricted
sets of instructions, particularly those requiring necessary terms, provided a way of
testing the relationship between the two blue terms. The implicit instruction referred
to the use of 40 stimuli from just the green-blue-purple region (Study 4). This was
done to map the domain of the two blue terms more precisely, but also as a further
check on their robustness.
As regards colour stimuli, the naming studies used stimuli drawn from a number of
commercial colour-order systems such as the Natural Colour System, Color-Aid,
Munsell, and the Optical Society of America.
10
Provided colour-space is sampled
adequately, it was expected that it would make little difference which system is used.
As regards lighting, daylight, CIE illuminant C (“north light”) and fluorescent were
used. It was expected that variations in the illuminant would lead to naming changes
for colours close to category boundaries, but they are unlikely to do so for category foci
and their near neighbours. Accordingly, stability across illuminants would be a
characteristic of basic terms.
The informants were bilingual Greek-English students and monolingual Greek
speakers from Crete.
The results in Androulaki et al. (2006) suggest that MG has twelve BCTs, i.e. the
eleven BCTs identified by Berlin & Kay (1969; see (1)) and the additional BCT ghala
´zio
‘light blue’. In particular, the MG BCTs are WHITE a
´spro,BLACK ma
´vro,RED ko
´kino,
GREEN pra
´sino,YELLOW kı
´trino,BLUE ble, BROWN kafe
´,ORANGE portokalı
´,PINK roz, PURPLE
mov, GREY gri, and LIGHT BLUE ghala
´zio or ghalano
´.
In the following, I give a brief etymological and grammatical profile of the MGBCTs,
largely following the description by Androulaki et al. (2006).
C O L O U R V E R B S IN MO D E R N GR E E K :A CO G N I T I V E AP P R O A C H 129
ProofsProofs
The term a
´spro ‘white’ originates from the Latin, ma
´vro ‘black’ from AG. The terms
ko
´kino ‘red’, pra
´sino ‘green’, kı
´trino ‘yellow’, and portokalı
´‘orange’ are derived from
plant names, but only the latter points to its derivation base explicitly (N portoka
´li
‘orange’).
11
All these terms are both nouns (i.e. the colour itself) and adjectives (when
accompanying a noun as modifiers). As nouns, they have eight forms, i.e. two numbers
(singular and plural) and four cases for each number (nominative, genitive, accusative,
vocative). As adjectives, they have in addition three genders (masculine, feminine,
neuter).
The terms ble ‘blue’, kafe
´‘brown’, roz ‘pink’, mov ‘purple’ and gri ‘grey’ come
from French and are not inflected.
12
In addition, the terms kafe
´‘brown’ and gri ‘grey’
have the derivatives kafetı
´and grı
´zo, respectively, which are used as nouns and
adjectives. As adjectives, they are inflected for all genders, numbers and cases.
13
The
BCTskafetı
´and portokalı
´follow a very productive pattern, since the derivation of colour
adjectives by means of the inflectional and derivational suffix -ı
´is the standard process
for adding terms for special hues, i.e. non-BCTs, to the MG lexicon (see Anastasiadis-
Simeonidis 1987: 397–398, Filopoulos 1994: 200). It should be noted that, in the
adjectival use, the neuter -ı
´form is regarded as basic and is often used as is, i.e.
uninflected, for all genders and cases both in singular and plural NPs (Holton et al.
1997: 79).
14
Androulaki et al. (2006) found out that the use of BCTs was reasonably stable
across variations in methods, stimuli, lighting, and informants. The two BCTs for blue,
i.e. ble and ghala
´zio, differ mainly in lightness. As Androulaki et al. (2006: 7) argue,
‘adding another blue term [in the case of MG, the term ghala
´zio/ghalano
´,A/N] appears
to be the most common way that languages move beyond stage VII’.
The distinction between blue and light blue in MG is similar to the equivalent
divisions in Maltese (c
˙elesti ‘sky blue’ and blu ‘dark blue’; Borg 2011), Russian
(sinij ‘dark blue’ and goluboj ‘light blue’; Corbett & Morgan 1988, Davies and
Corbett 1994, Paramei 2007), Turkish (mavi ‘blue’ and lacivert ‘dark blue’;
O
¨zgen & Davies 1998, Ra
¨tsep 2011), etc. It should be noted, however, that the
stipulation of two blue terms is not always attainable. As regards Turkish, O
¨zgen
& Davies (1998) and Ra
¨tsep (2011) conclude that only mavi can be considered
as a BCT with safety. Similarly, Al-Rasheed et al. (2011) conclude that Arabic
has eleven basic colour terms that correspond to Berlin and Kay’s (1969) universal
set. Samawee ‘light blue’ and khuhlie ‘dark blue’ are not BCTs but only azrock
‘blue’.
15
Nonetheless, the results in Androulaki et al. (2006) explicitly justify the status
of ghala
´zio as a BCT. As will be argued in section 5.4, the prominence of the term
ghala
´zio in the MG vocabulary is reinforced by historical, environmental, and cultural
factors.
4 The linguistic relevance of colour studies
The focus of the study of BCTs has always been within the domain of psychology and
psycholinguistics. However, as Sutrop (2011) points out, a colour field is regulated
130 C H A R I T O N C H A R I T O N I D I S
ProofsProofs
semiotically in any language, whereby the history of language and culture plays
a major role.
16
According to Biggam (2012) ‘as much detail as possible should be recorded
about the context of each occurrence of a colour term, and this requirement
may cause the researcher to be concerned with matters such as the age, sex and
social position of both speaker and listener, details of the environment, type of
event and social mood of each occasion of use and much more valuable data.’
(Biggam 2012: 7)
In a series of works, Lucy (1991, 1992, 1997, etc.) criticizes the aims and methods of
colour studies.
17
For example, Lucy (1992: 177) questions the criteria for identifying a
‘basic colour term’ introduced in Berlin & Kay (1969: 5–7). In Lucy’s view, these
criteria are circular and not theoretically justified.
Lucy’s (1997) study is an in-depth analysis of the linguistic relevance of colour
studies. According to this study ‘meaning is not reducible to denotation’ (Lucy 1997:
328). In natural language, referents are picked out according to a category’s
characteristic referential range and its formal distributional potential. Both factors
work together to give a form its semantic value (Lucy 1997: 322).
18
In the following,
I present the main points of criticism in Lucy’s (1997) study and discuss them.
4.1 Basic colour terms (BCTs) and predefined stimuli
According to Lucy (1997), the Munsell colour array, used for the first time in
Berlin & Kay (1969), refers to a set of pre-defined colour hues. It does not include
variation in luster, luminosity, or reflectance. The same array dictates in advance
the possible meanings terms could have since no other meanings are embodied in
the samples. In addition, it is unrealistic since it forces informants to make
comparisons and judgements about colour variations not encountered in daily life.
Obviously, Lucy’s criticism can be generalized over colour studies using pre-defined
stimuli other than the Munsell colour array (cf. various studies in Biggam et al.
2011, etc.).
Contrary to Lucy’s (1997) assumptions, several studies underscore the reduced
relevance of the different stimuli used in colour research (Androulaki et al. 2006, Al-
Rasheed et al. 2011, etc.). As already mentioned in section 3, Androulaki et al. (2006)
employ different stimuli, and different illuminants in their experiments. They
recognize the involvement of lightness in specifying the domains of most BCTs, but
at the same time note that precise control over the different variables ‘is not crucial
in field studies aimed at establishing a language’s BCTs’ (Androulaki et al. 2006: 39).
Al-Rasheed et al. (2011) report that the results from tasks with predefined stimuli point
to the same BCTs as colour naming, etc.
The aforementioned findings point to the existence of a categorial, and perhaps also
pre-linguistic, perception of colour (cf. Sandford 2011: 216).
C O L O U R V E R B S IN MO D E R N GR E E K :A CO G N I T I V E AP P R O A C H 131
ProofsProofs
Concluding, the use of predefined stimuli is an effective means for the identification
of a language’s BCTs.
4.2 Formal distribution of colour terms
According to Lucy (1997), distributional facts relating to paradigms are usually ignored
in colour studies. The examination of the content of lexical items in isolation, for
example colour adjectives, is not really systematic. It may be the case that terms having
some reference to colour are a heterogeneous collection such as the derivational
paradigm of the EN colour adjectives. Accordingly, the distributional patterns of colour
terms must be taken into account in order to generate a typology of ‘colour systems’
across languages. At the same time, premature judgements about ‘deficient’ colour
systems can be avoided.
In line with Lucy’s arguments, I argue that there can be no ‘deficiency’ in colour
denotation, but only lexical or morphological accommodation of an underlying colour
system. Gaps in derivational paradigms may point to fundamental colour hierarchies,
such as the distinction between primary and non-primary BCTs (section 2). For
instance, established -ing forms in EN are mainly attested for the primary colours white,
black, red, green, yellow, and blue. Only the gerund browning derived from the first
non-primary BCT brown (Stage VI) does not follow this pattern (see Table 15.2 in Lucy
1997: 329), etc.
Verbs may refer to a broader colour gamut than nominals since they incorporate a
notion of continuation or measurement (Charitonidis 2012a, 2012b, 2013). In section
6.2 I argue that a putative verb form ?bluen ‘become/make blue’ may be regarded as a
derivation of the GRUE category and thus maintain the coherence of the -en paradigm
up to Stage III.
These and similar patterns point to the existence and evolutional dynamics of an
underlying colour system. Further distributional facts can be found in the remainder of
this paper which focuses on MG and EN verbal derivation.
4.3 The role of colour in natural language semantics
Lucy (1997) argues that colour is not central to the semantic organization of any
language that he knows of. ‘Unlike agency, time, number, or other such categories, it is
never grammaticalized’ (Lucy 1997: 330). He reports Conklin’s (1955 [1964: 189])
statement that ‘Colour, in a western technical sense, is not a universal concept and in
many languages such as Hanuno
´o there is no unitary terminological equivalent’ (Lucy
1997: 332). At another point, Lucy argues: ‘Although colour concepts tend to be
treated as adjectives when a language has such a category, this is by no means universal
either’ (Lucy 1997: 337). In other words, the absence of a uniform linguistic or
semantic category for colour in many languages challenges the assumption that a
universal colour system underlies language as a whole.
I agree with Lucy’s position that colour is not central to language since language
is a means for accomplishing complex communication tasks which refer to colours
132 C H A R I T O N C H A R I T O N I D I S
ProofsProofs
only marginally. However, colour can play a significant role in particular subsystems
of specific languages, e.g. in morphology.
By way of example, an unrestricted causative/inchoative structure for the colour
verbal derivatives in MG would overgenerate verbs. Non-primary BCTs and non-BCTs
tend to yield no causative verbs in this language (section 5). Similarly, both causative
and inchoative verbal derivatives are extremely rare for the non-BCTs in EN (section 6).
These restrictions must be taken into account as semantic conditions on the [ +Loc]
(or end-state) argument in the respective causative/inchoative structures – see (3)
and (4) below.
In the structure for the MG colour verbs in (3), three disjoint conditions are
assumed, i.e. [ +BCT] [ +PRIMARY], [ +BCT] [–PRIMARY], and [–BCT][ +CHAR].
[–PRIMARY] refers to the distinction between primary and non-primary BCTs
in Kay & McDaniel (1978). [ +CHAR] stands for ‘characteristic colour’ and refers
to colour adjectives derived from the names of objects having a characteristic
colour. The options [ +BCT][–PRIMARY] and [–BCT][ +CHAR], will additionally
drop the first causative part of the denotational skeleton, indicated with an
underline.
19
(3) MG colour verbs
[+dynamic ([
volitional-i
], [
j
])];
[+dynamic ([
i
], [ +dynamic, +IEPS ([
j
], [ +Loc ])])])], <base >]
[
+
BCT
][
+
PRIMARY
]
[
+
BCT
][
-
PRIMARY
]
[
-
BCT
][
+
CHAR
]
(4) EN colour verbs
[+dynamic ([
volitional-i
], [
j
])]; [+dynamic ([
i
], [ +dynamic, +IEPS ([
j
],
[+Loc[
+
BCT
] ])])])], <base>]
Concluding, colour denotation is a powerful generative component restricting
overgeneration – at least in verbal derivation (cf. the discussion about grammaticality
judgments in sections 5.3 and 6.2).
20
To address again the last point in Lucy’s argumentation, the fact that a
specific concept such as colour has no terminological equivalent in some languages
does not suspend the linking of these languages to an underlying colour system.
By way of comparison, truth conditions in logic refer to denotational functions
which are only inconsistently expressed in natural language, such as the inclusive
OR (Allwood et al. 1977: 35–37). This fact does not impair the capability of such
conditions to address basic semantic aspects in natural language constructions
universally.
Let us now turn to the analysis of the MG verbal derivatives in -ı
´zo in the COLOUR
field.
C O L O U R V E R B S I N M O D E R N G R E E K :A C O G N I T I V E A P P R O A C H 133
ProofsProofs
5 Colour verbs in Modern Greek (MG)
5.1 General patterns
Table 1 gives the general patterns of the MG colour terms – colour adjectives together
with their verbal derivatives. The lines dividing the table in subparts indicate
evolutionary thresholds (with the exception of sections 5 and 6, both referring to Stage
VIII +). In particular, the first group (Stages I–V) refers to primary BCTs, the second
group with only one member (Stage VI) refers to the non-primary BCT kafe
´/kafetı
´
‘brown’, and the third group (Stage VII) refers to the rest of the non-primary BCTs.
The fourth group (onset of Stage VIII +) refers to the sole BCT beyond Stage VII, i.e.
ghala
´zio/ghalano
´‘light blue’, identified as a proper BCT in Androulaki et al. (2006).
21
The fifth group refers to a representative sample of non-BCTs for special hues. The
sixth group (Stage VIII +) refers to non-BCTs as well. The terms xriso
´‘gold’/‘golden’,
porfiro
´‘Tyrian purple’/‘in the colour of porphura’, asimı
´‘silver’ and xrisafı
´‘golden’
contained therein are derived from the names of precious metals or dyestuffs
(for details about the terms in the sixth group see sections 5.2 and 5.4).
In the third row of Table 1 an -ı
´zo verb derivative for each colour adjective in the
first row is given. Only gri/grı
´zo ‘grey’ has no derivative in -ı
´zo but this may be due to
the wild-card behaviour of this term which, as already pointed out, occurs at any stage
from III to VII (see (2) in section 2).
22
In other words, the absence of a unique
assignment of the base to an evolutionary stage restrains derivation.
In the fourth and fifth row indications of use for the causative/inchoative alternation
can be found, i.e. ‘in common use’ (3), ‘rare’ (†), ‘very rare’ (), and ‘non-existent’ (*).
For these indications Google’s search engine was used. The Hellenic National Corpus
(HNC) was not employed because it currently contains only about 47.000.000 words
(accessed October 3, 2012). At the same time, very fresh neologisms – which abandon
in blogs, discussion groups, etc. – cannot be accessed by using HNC.
In the sixth row, alternate verb forms for the -ı
´zo verbs are given. Again, there are
indications of use for the causative and/or inchoative variant defined by using Google.
The chronological indications for alternates, bases, and verbal derivatives in the last
three rows were based on the two major dictionaries for MG, i.e. the Dictionary of
Common Modern Greek (DCMG) and the Dictionary of Modern Greek Language
(DMGL).
23
The indication ‘neologism’ (NL) in Table 1 is based on the absence of the
verb forms in DCMG and DMGL, the appearance of these forms in various discussions
groups, forums, etc. on World Wide Web (WWW), and in some ambiguous cases on the
linguistic intuition of the author as a native speaker of (Athenian) MG.
As can be seen in Table 1, five terms for blue show up, i.e. ble ‘blue’, ghala
´zio/
ghalano
´‘light blue’, bla
´vo ‘dark blue’, and thalası
´‘sea blue’, ‘light blue’. ble is assigned
to the first group (Stage V) because it is the most general and standard BCT for blue in
MG (Androulaki et al. 2006). The terms ghala
´zio/galano
´‘light blue’ are assigned to the
fourth group (onset of Stage VIII +) because they are identified as non-primary BCTs in
Androulaki et al. (2006). bla
´vo ‘dark blue’ is contained in the fifth group (Stage VIII +)
because it denotes a special non-BCT hue.
24
thalası
´‘sea blue’, ‘light blue’ is also
134 C H A R I T O N C H A R I T O N I D I S
ProofsProofs
Table 1. Modern Greek colour terms.
Base
Stage
(B&K)
-ı
´zo
verb
Inchoative
(Google)
Causative
(Google)
Alternates
(Google)
Base
(DCMG,
DMGL)
-ı
´zo verb
(DCMG,
DMGL)
1a
´spro ‘white’ I asprı
´zo 3 3 HL MD
ma
´vro ‘black’ I mavrı
´zo 3 3 HL MD
ko
´kino ‘red’ II kokinı
´zo 3 3 HL HL
kı
´trino ‘yellow’ IV kitrinı
´zo 3 3 kitrinia
´zo (":†) [NL, A/N] HL MD
pra
´sinoble ‘green’ IV prasinı
´zo 3 3 AG HL
‘blue’ V bledhı
´zo 3†MD +(fr. bleu) NL (A/N)
2kafe
´/
kafetı
´
‘brown’ VI kafetı
´zo 3MD/MD +NL (A/N)
3gri/grı
´zo ‘grey’ VII – – – griza
´ro (#/"), grizo
´no (#/":)
[NL, A/N]
MD +(fr. gris) NL (A/N)
mov ‘purple’ VII movı
´zo 3MD +(fr. mauve) NL (A/N)
portokalı
´‘orange’ VII portokalı
´zo 3MD +NL (A/N)
roz ‘pink’ VII rozı
´zo 3†?roza
´ro (":) [NL, A/N] MD +(fr. rose) NL (A/N)
4ghala
´zio ‘light blue’ VIII +
(onset)
?ghalazı
´zo †ghalazo
´no (#:†,") [NL, A/N] MD NL (A/N)
ghalano
´‘light blue’ VIII +
(onset)
?ghalanı
´zo †ghalano
´no (#:,":†) [NL, A/N] MD ?MD
5bla
´vo ‘dark blue’ VIII +blavı
´zo 3†MD MD +
bez ‘beige’ VIII +?bezı
´zo †* beza
´ro (#:1, ") [NL, A/N] MD +(fr. beige) NL (A/N)
ladhı
´‘olive’ VIII +?ladhı
´zo * ladho
´no (#/": spatial) [MD] MD +NL (A/N)
thalası
´‘sea blue’, ‘light blue’ VIII +?thalası
´zo * thalaso
´no (LX) [MD +] MD +NL (A/N)
visinı
´‘berry’, ‘cherry red’ VIII +visinı
´zo 3* MD +NL (A/N)
6xriso
´‘golden’ VIII +xrisı
´zo 3 3 xriso
´no (#: mainly spatial) [AG] AG/MD AG
porfiro
´‘Tyrian purple’ VIII +porfirı
´zo 3†porfiro
´no (#,":†) [?] AG/MD +HL/MD +
asimı
´‘silver’ VIII +asimı
´zo 3 3 asimo
´no (#,#+spatial, LX, etc) [MD] MD +NL (A/N)
xrisafı
´‘golden’ VIII +xrisafı
´zo 3†MD +NL (A/N)
#
°
*
in common use
rare
very rare
non-existent
causative
A/N
AG
B&K
HL
inchoative
author’s note
Ancient Greek: 1400–300 B.C.
Berlin & Kay (1969)
Hellenistic Greek: 300 B.C.–400~500 A.D.
LX
MD
MD+
NL
lexicalization
Medieval Greek: 500–1800 A.D.
1800–
neologism (last decades)
C O L O U R V E R B S I N M O D E R N G R E E K :A C O G N I T I V E A P P R O A C H 135
ProofsProofs
contained in the fifth group not only because it denotes a special non-BCT hue but also
because in Androulaki et al. (2006) it shows up as a marginal term replacing ghala
´zio
only sporadically.
5.2 Chronology and use
The chronological indications in the last two rows of Table 1 are quite uniform
and support an evolutionary view of both the derivation bases and the -ı
´zo derivatives.
In general, MD verbs show up after HL bases, NL verbs show up after MD +bases.
At Stages I–IV a
´spro ‘white’, ma
´vro ‘black’, ko
´kino ‘red’, and kı
´trino ‘yellow’
constitute the most essential core as HL bases paired mainly with MD derivatives. The
adjective pra
´sino ‘green’ at Stage IV is opposed to this chronological order by being an
AG word (Anastasiadis-Simeonidis 1987: 411). This may be a hint for assigning pra
´sino
back to Stage III, i.e. before kı
´trino. Another peculiarity is the old chronology of both
the adjectives xriso
´‘golden’ and porfiro
´‘Tyrian purple’ and their verbal derivatives
which associates these terms with Stages I–IV (see section 6 in Table 1). These terms
ought to follow the patterns of the other Stage-VIII +verbs, i.e. MD +base – NL
derivative. The indications of use for the derivatives associate these terms with Stages
I–IV, as well. As will be argued in section 5.4 these patterns are due to the cultural
prominence of the entities referred to by xriso
´and porfiro
´.
As regards the indications of use for the -ı
´zo verbs in Table 1, the verbs at Stages
I–IV are in common use in both the inchoative and causative variant. From Stage V
and beyond, however, the inchoative variant is more widely used than the causative
one. A decreasing pattern from the primary to the non-primary BCTs and from the BCTs
to the non-BCTs shows up. First, there is a drop in the use of the causative variants of
the derivatives (primary BCTs vs. non-primary BCTs). Second, there is a drop in the use
of inchoatives, accompanied by total diminishing of causatives in some cases (BCTs vs.
non-BCTs). This decreasing pattern is not always perfect, cf. asimı
´zo ‘become/make
silver’ (ası
´mi ‘silver’/non-BCT) being in common use in both the inchoative and
causative variant, etc.
25
Leaving aside these few exceptions, however, the general
patterns of use are quite uniform.
5.3 Morphophonology
The derivation bases at Stages I–IV are native parisyllabic adjectives in -os, -i, -o (for
masculine, feminine, and neuter, respectively)
26
and are stressed on the penultimate,
see a
´spro ‘white’ and ma
´vro ‘black’, or on the antipenultimate syllable, see kı
´trino
‘yellow’, ko
´kino ‘red’, and pra
´sino ‘green’.
27
ble ‘blue’ at Stage V is the unadjusted
French word bleu ‘blue’. The exceptional morphophonological pattern of this term
introduces Stages VI and VII, which mainly contain phonologically unadjusted bases
of French origin, see kafe
´‘brown’, gri ‘grey’, mov ‘purple’, and roz ‘pink’ (the variants
kafetı
´for kafe
´,and grı
´zo for gri are phonologically adjusted).
28
In other words,
ble stands immediately before Stage VI, referring to a major shift in the morphological
make-up of the derivation bases. The exceptional portokalı
´‘orange’ at Stage VII is
136 C H A R I T O N C H A R I T O N I D I S
ProofsProofs
a native Greek term. It refers again to a major (second) shift in the morphological
make-up of the bases, i.e. -ı
´derivation, and, in a similar way as ble, introduces the
next stages.
As already mentioned in section 3, most bases at Stage VIII +are adjectives in
-ı
´,stressed on the final syllable, cf. asimı
´‘silver’, ladhı
´‘olive’, thalası
´‘azure’, visinı
´
‘cherry red’, xrisafı
´‘golden’, etc.
29
It should be noted that the suffix -ı
´zo almost
perfectly matches the morphophonological shape of the adjectival/nominal ending
-ı
´than the other verb forming suffixes in MG because it contains a stressed -i-.
In the following, I use the labels ‘possible/established’, ‘possible/novel’, or ‘not
possible/forced’ for the verbal derivatives, whereby the attestation of these verbs in the
www always calls for their characterisation as possible.
The established derivatives asprı
´zo ‘become/make white’, mavrı
´zo ‘become/make
black’, kokinı
´zo ‘become/make red’, kitrinı
´zo ‘become/make yellow’, and prasinı
´zo
‘become/make green’, at Stages I–IV are regarded as the most natural colour verbs, as
opposed to bledhı
´zo ‘become/make blue’, kafetı
´zo ‘become/make brown’, movı
´zo
‘become/make purple’, and rozı
´zo ‘become/make pink’ at Stages V–VII which are
possible/novel. The derivatives ?ladhı
´zo ‘become olive-green’,
30
?thalası
´zo ‘become
sea-blue’,
31
visinı
´zo ‘become cherry-red, asimı
´zo ‘become/make silver’,
32
and xrisafı
´zo
‘become/make golden’
33
at Stage VIII +are possible/novel, as well. These derivatives
have bases denoting material entities with a characteristic colour. Ultimately, verbs
derived from adjectives in -ı
´whose bases denote an object without a characteristic
colour, cf. !trapezı
´zo (trape
´zi ‘table’) or !potirı
´zo (potı
´ri ‘glass’), etc., are definitely ‘not
possible/forced’.
Concluding, the noun xro
´ma ‘colour’ and its -ı
´zo derivative xromatı
´zo ‘to colour’
serve as hyperonyms for all colour adjectives/nominals and their verbal derivatives,
respectively. This pattern points to the relevance and coherence of the COLOUR field
in MG.
5.4 Exceptional cases: cultural prominence
There are five derivatives in -ı
´zo, i.e. ?ghalazı
´zo ‘become/make light blue’,
34
?ghalanı
´zo
‘become/make light blue’,
35
blavı
´zo ‘become/make dark blue’, xrisı
´zo ‘become/make
golden’, and porfirı
´zo ‘become/make cerise’, whose patterns of use in the inchoative
and/or causative variant, together with the morphological and evolutional make-up
of their bases, do not seem to conform to the general patterns presented in sections
5.1–5.3.
A. The bases ghala
´zio ‘light blue’ and ghalano
´‘light blue’ of ?ghalazı
´zo ‘become/
make light blue’ and ?ghalanı
´zo ‘become/make light blue’, respectively, were assigned
to the onset of Stage VIII +according to Androulaki et al. (2006; see section 3).
However, these bases follow the pattern of the colour adjectives at Stages I–IV. Why
does this happen? As already mentioned, the difficulty to define a BCT for blue is due to
historical and cultural reasons since the status of the BLUE category in Greek was
traditionally uncertain (see n. 14). In view of this situation, the domination of the
colour ‘light blue’ in the Greek cultural area has favoured a regular morphology for
C O L O U R V E R B S I N M O D E R N G R E E K :A C O G N I T I V E A P P R O A C H 137
ProofsProofs
the base adjectives ghala
´zio and ghalano
´(cf. Wierzbicka’s 1990, 1996 description of the
influence of swaths of colours in the natural environment).
B. The base bla
´vo ‘dark blue’ of blavı
´zo ‘become/make dark blue’ was assigned
to Stage VIII +after ghala
´zio and ghalano
´. Again, this base follows the pattern of the
colour adjectives at Stages I–IV. As already mentioned in n. 22 bla
´vo is a loan word
from Italian (blavo). Its phonological form is explicitely similar to that of the native
Greek colour terms at the early stages. At the time of its occurrence in the Greek
vocabulary the standard BCT for blue, i.e. ble, was not yet used. Therefore, bla
´vo was
used in juxtaposition with the two terms for light blue, i.e. ghala
´zio and ghalano
´.
Accordingly, the frequent use of its derivative blavı
´zo has been favoured – in this case
in the prototypical inchoative variant.
C. As already mentioned in section 5.2, the old chronology and the common use of
the derivatives xrisı
´zo ‘become/make golden’ and porfirı
´zo ‘become/make cerise’ refer
these verbs to Stages I–IV. These verbs are derived from the non-BCTsxriso
´‘golden’
and porfiro
´‘Tyrian purple’, respectively. Like ghala
´zio/ghalano
´and bla
´vo, xriso
´and
porfiro
´follow the morphological pattern of the colour adjectives at Stages I–IV
(Androulaki et al. 2006). I assume that this cross-reference of use for the derivatives
and of form for the bases relies on the cultural prominence of the underlying entities,
i.e. xriso
´s‘gold’ and porfı
´ra ‘porphura’.
36,37
6 English (EN) colour verbs
The MG derivational patterns presented up to this point would only be suggestions
if they could not be scaled up in a different language. To that end, I will present
the corresponding patterns from EN, a language whose system of verbal derivation is
considerably different than that of MG.
In EN, the suffixes -ate, -ify, -ize, and -en are used to derive verbs from other
categories (mostly nouns and adjectives; Plag 1999, 2003: 92–94). New derivatives in -
en are extremely rare. In OED (1994) only two causative neologisms in -en are
attested (crispen, outen), which also have earlier zero-derived equivalents (Plag 1999:
219). eN-seems to be the only clearly category-changing prefix in EN verbal derivation.
It attaches mainly to nouns to form verbs with a spatial interpretation (envision,
emplane; Plag 1999: 93, 218; for other verb-deriving prefixes see Plag 1999, 2003).
On the other hand, conversion is the most productive verb-deriving process in EN.
It can create verbs from nouns, e.g. to boot from boot, and sometimes verbs from
adjectives, e.g. to cool from cool (Lieber 2004: 89–95).
Table 2 gives the general profile of the EN colour adjectives or nominals
together with their verbal derivatives. The lines dividing the table indicate the same
evolutionary thresholds as in Table 1 for MG (with the exception of sections 5, 6, and 7,
referring to Stage VIII +). The first six sections of Table 2 contain terms equivalent to
the MG terms. Section 4 contains no terms because there are no English BCTs at the
onset of Stage VIII +. In section 5 there is no one-word term corresponding to the MG
term bla
´vo ‘dark blue’. azure and sea correspond to the MG term thalası
´‘sea-blue’, and
the terms burgundy, maroon, and crimson correspond to the MG term visinı
´‘cherry-red’.
138 C H A R I T O N C H A R I T O N I D I S
ProofsProofs
These correspondences are only suggestions and should be regarded as a working
hypothesis.
38
In section 6 there is no one-word term corresponding to the MG term
porfiro
´‘Tyrian purple’, ‘in the colour of porphura’.
There is an extra seventh section at the bottom of Table 2 which contains
the historically first non-BCTs in the system of EN language, taken from Casson (1997:
233–234).
39
The addition of this additional sixth section reduces arbitrariness in
the description of the immediately preceding sections 5 and 6 which contain terms
equivalent to the MG terms.
In the third row of Table 2 verbal derivatives for the colour terms of the first row
are given, in particular -en derivatives or zero-derived (converted) verbs.
In the fourth and fifth row there are indications of use for the causative/inchoative
alternation, i.e. ‘in common use’ (3), ‘rare’ (†), and ‘non-existent’ (*). For these
indications the British National Corpus (BNC; http://corpus.byu.edu/bnc/) and the
Corpus of Contemporary American English (COCA; http://corpus.byu.edu/coca/)
were consulted.
40
In the search for converted verbs the sequence ‘[form].[*v]’ was
mainly used.
41
In the sixth row alternate verbs for the derivatives in the third row are given,
together with indications of use. The chronological indications in the last three rows
were taken from the online version of Oxford English Dictionary (OED; www.oed.com).
In the seventh row are indicated the lexical category of the base and in the eightieth
row the transitive/intransitive use of the verbs contained in the third row, taken again
from OED. The reason for the inclusion of the transitivity indications in
the eightieth row is that in some cases the search in BNC/COCA did not yield any
verbs. As will become clear in section 6.1, this information is crucial for the
comparison of the EN with the MG colour verbs. The indications of use in Table 2 were
validated by using WWW, e.g. by looking up the relevant terms in discussion forums
such as the ‘English Language and Usage’ forum at http://english.stackexchange.com,
etc. In addition, two native speakers of EN were interviewed on the current status of the
verbal derivatives at the Campus of the University of Cologne, Germany.
6.1 Chronology and use
As in the case of MG, the chronology of the EN colour terms is quite uniform and
supports an evolutionary view of both the derivation bases (adjectives or nouns)
and the verbal derivatives. In general, verbs from late Old English (c.900–c.1150)
or Middle English (c.1150–1500) show up after bases from early Old English
(c. 650–900), verbs from early Modern English (1500–1700) show up after bases from
(late) Middle English (see Table 2).
Similarly as in MG, at Stages I–IV the terms white, black, red, yellow and green
constitute the most basic core. These terms come from early Old English and are
paired with derivatives from early Old English to Middle English. The term grey at
Stage VII seems to violate the chronological BCT-sequence strongly. It showed up in
early Old English (a700) before the terms blue (1366) and brown (a1000). This
exceptionality, however, should not surprise us. As already pointed out in section 2,
C O L O U R V E R B S I N M O D E R N G R E E K :A C O G N I T I V E A P P R O A C H 139
ProofsProofs
Table 2. English Colour Terms.
Base
Stage
(B&K) Derivative Inchoative Causative Alternates
Base
(OED)
Derivative
(OED)
1white I whiten 3 3 c888 (A) a1330
black I blacken 3 3 black (O: t) eOE (N-A) c1330
red II redden 3 3 ?red (O: t) eOE (N-A) 1552
yellow IV yellow 3 3 c700 (N-A) a1050
green IV green 3 3 eOE (N-A) eOE
blue V blue ? (amb) 3?bluen (O&M)
(O: 0, B/C: 0)
1366 1606 (t)
2brown VI brown 3 3 a1000 (A) c1300
3grey VII grey 3 3 a700 (N-A) c1400
purple VII purple 3 3 OE (N-A) ?a1475
orange VII * * * a1400 (N-A) *
pink VII pinken †(B: 1, C: 2) †(B: 0, C: 2) ?pink (O [1792]: t/i [ +up],
B: "[†], C: #/"[†])
1566 (N-A) 1890 (t/i)
4N/A VIII +(onset) N/A N/A N/A N/A N/A N/A
5beige VIII +* * * 1858 (N-A) *
olive VIII +* * * a1225 (N-A) *
azure VIII +azure * (B/C) * (B/C) a1330 (N-A) 1490 (t)
sea VIII +* * * c825 (N) *
burgundy VIII +* * * 1600 (N) *
maroon VIII +* * * 1594 (N-A) *
crimson VIII +crimson †(B/C) †(B: 0, C: 1) 1416 (N-A) 1609 (t/i)
6golden VIII +golden * (B/C) * (B/C) gold (N): c725
golden (A): c1300
1850 (t/i, †)
silver VIII +silver †(B/C) †(B/C) silverize (O [1605]:
#:†, B/C: 0)
c825 (N-A) c1440 (t/i)
140 C H A R I T O N C H A R I T O N I D I S
ProofsProofs
7violet VIII +violet * (B/C) * (B/C) 1370 (A) 1623 (t)
russet VIII +russet * (B/C) * (B/C) a1330 1592 (t)
ochre/ocher VIII +ochre/ocher * (B/C) * (B/C) 1364 (N-A) 1553 (t)
scarlet VIII +scarlet * (B/C) * (B/C) c1250 1553 (t, obs)
vermilion VIII +vermilion * (B/C) * (B/C) 1296 (N-A) 1606 (t)
#
*
A
a
amb
B [=BNC]
B&K
in common use
rare
non-existent
causative
inchoative
adjective [OED]
ante [OED]
categorial ambiguity (B/C)
British National Corpus
Berlin & Kay (1969)
C [=COCA]
c
e
i
N
N/A
O [=OED]
OE
O&M
obs
t
Corpus of Contemporary American English
circa [OED]
early [OED]
intransitive [OED]
Noun [OED]
not applicable
Oxford English Dictionary (online)
Old English (c.700-c.1150) [OED]
Onysko & Michel (2010: 3)
obsolete [OED]
transitive [OED]
C O L O U R V E R B S I N M O D E R N G R E E K :A C O G N I T I V E A P P R O A C H 141
ProofsProofs
in a later revision of Berlin and Kay’s (1969) model, i.e. in Kay & Maffi (1999), grey is
regarded as a wild-card occurring at any stage from III to VII.
The chronological indications in sections 5 and 6 of Table 2 (Stage VIII +) are not
directly relevant in the present context because the bases and the derivatives appearing
therein are included for comparison with the equivalent MG terms (sections 5 and 6
of Table 1). Nonetheless, in five of the nine cases, no verbal derivatives are attested – a
pattern which is in accord with the general drop of EN verbal derivatives at Stage
VIII +(see below). As an exception, the base silver and its verbal derivative silver follow
the chronological pattern at Stages I–IV (cf. n. 34).
Section 7 of Table 2 refers to the first non-BCTs for EN (Casson 1997; Stage VIII +).
In general, verbs from early Modern English show up after nouns or adjectives from
Middle English (OED).
According again to the indications in Table 2, the colour verbs in -en (Stages I & II)
are in everyday use, followed by converted verbs which up to Stage VII are also
common.
42
There is no verb *to orange at Stage VII, but this is due to the exceptional
and transparent association of this verb to an object having a characteristic colour.
43
Colour terms of this kind tend to emerge after Stage VII and yield no derivatives
in EN.
44
The derivative pinken at Stage VII is a rare verb in BNC and COCA and
an existing verb for OED. It is also a possible verb for some native speakers of EN.
According to the present analysis, this is the only morphologically dissonant case
within the group of the EN BCTs. It should be noted, however, that pinken is the
first verb derivative in the evolutionary sequence that according to BNC and COCA is
rare. From this stage and beyond (Stage VIII +), verbal derivatives are either rare
(see to silver and to crimson) or non-existent (cf. *to beige, *to olive, etc.). A
further peculiarity of pinken is that, according to OED, it is the newest verb among the
verbs in Table 2 (first token: 1890). This fact suggests the status of a near-neologism
for this verb, in accord with the neologism status of the MG verbal derivatives at the
same stage.
Most notably, the historically first non-BCTs (Casson 1997: 233–234) in the bottom
(seventh) section of Table 2 have no derivatives at all – again according to BNC and
COCA.
45
Only OED reports the existence of causative variants for these verbs – an
opposite pattern in relation to the MG verbal derivatives, which seem to develop the
inchoative variant first (cf. sections 4.3 and 5.2). Let us now recruit morphophonology
to account for these patterns more accurately.
6.2 Morphophonology
In EN, there are morphophonological restrictions on the -en verbs. Their bases are
monosyllables and end in a plosive, fricative or affricate (see Plag 1999, 2003). This
means that phonological factors prohibit the combination of vowels at the syllable
boundaries between stem and suffix. Accordingly, the bases of the established colour
verbs blacken, whiten, and redden end in a plosive. However, the novel verb ?bluen, with
a rear combination of vowels at the syllable boundaries between the stem and the suffix,
may occur in some contexts, as Onysko and Michel (2010: 3) report. We see thus that
142 C H A R I T O N C H A R I T O N I D I S
ProofsProofs
morphophonology points to the distinction between established verbs and
neologisms.
46
In a similar way as in MG, the bases of the attested EN colour verbs blacken, whiten,
and redden exhaust the Stage II of BCTs. Blue, the base of ?bluen does not appear until
stage V after yellow and green at Stage IV, respectively. According to the major later
revisions of Berlin & Kay’s (1969) model in Kay & Maffi (1999), at Stage III the
separation of grue and black takes place. ‘Grue’ may refer to more than one of the cool
colours, i.e. green, blue, and grey (see section 2). ?bluen can thus be regarded as derived
from a putative composite term ‘grue’ – cf. a similar discussion on the referential
status of colour verbs in Lucy (1997: 336–337).
The relatively new and rare verb pinken at Stage VII follows the morphophonological
pattern of whiten, blacken, and redden as the last -en verb for an EN BCT and as the last -en
verb for an EN colour-term in general (see also section 6.1). Overall, the cases of ?bluen
and pinken suggest that the -en paradigm is active throughout the BCT sequence.
In a similar way as in MG, there are verbs in -en as common hypernyms of all EN
colour verbs, i.e. the derivatives darken and whiten, whose bases refer to cool and warm
colours, respectively. This inclusion pattern supports the cognitive (categorial) and
evolutional analysis proposed in this paper.
Concluding, I would like to point out an issue of morphological levelling in both
MG and EN.
6.3 Morphological levelling
As the analysis in this paper has shown, MG colour verbs are mainly derived by means
of -ı
´zo. The base adjectives at Stages I–IV end in -os, -i, -o for masculine, feminine,
and neuter, respectively. Each of these endings is associated with a different
inflectional class (IC), i.e. IC1, IC3, and IC5, respectively, and is not stressed (see Ralli
2005: 119–120). After the exceptional Stages V–VII, mainly containing verbs of
French origin, and with a few exceptions at Stage VIII +,MG colour verbs follow
a highly uniform pattern. They have as bases adjectives in -ı
´,assigned to IC6.
47
This
ending is always stressed and at the same time derivational. Overall, at high stages a
very productive morphological pattern, i.e. the derivation of adjectives from nouns
by means of the ending -ı
´,is accompanied by the suffixation of a highly productive
element, i.e. -ı
´zo (Charitonidis 2005, 2011, Efthymiou 2011, Efthymiou et al. 2012).
48
In EN, at Stages I and II, EN colour verbs are canonically derived by means of the
suffix -en. The bases of these verbs meet specific conditions as regards the quality
of their end consonant (see section 6.2). At Stage V and beyond, EN colour verbs are
mainly derived by means of conversion.
49
This means that the most productive
morphological process for deriving verbs in EN (Clark & Clark 1979, Lieber 2004),
takes over an elaborate morphological process, i.e. -en suffixation.
I argue that these patterns of unification in both languages are special cases of
morphological levelling (cf. Haspelmath & Sims 2010: 273–274).
At the same time, an asymmetry is evident. In MG, the derivation bases conform to
a uniform and highly productive pattern at high stages (-ı
´ending). The verbal suffix
C O L O U R V E R B S I N M O D E R N G R E E K :A C O G N I T I V E A P P R O A C H 143
ProofsProofs
remains the same throughout the evolutional sequence (-ı
´zo). In EN, at later or higher
stages, a highly productive process (conversion) adopts a vastly varied set of bases.
In the latter case there is no direct reference to the morphological make-up of the
bases.
A further important difference between MG and EN is that, whereas in MG the
levelling directly refers to the inflectional paradigms of the colour adjectives and their
stress patterns, in EN the levelling involves the morphological process as a whole
without any reference to inflection.
7 Conclusion
The analysis in this paper has shown that it is the nature of human colour perception,
i.e. semantic/cognitive constraints, which in the first place call for the formation of
colour verbs in MG. The lower the BCT stage that a base can be fitted to, the more likely
an -ı
´zo derivative will be produced. A similar pattern is evident in the derivation of EN
colour verbs.
The distinction between primary and derived BCTs (Kay & McDaniel 1978) applies
in both MG and EN verbal derivation. The use of the verbal derivatives and the
grammaticality judgements of native speakers are in accord with this distinction.
Possible/established verbs address a primary categorial core (Stages I–V), cf.
MG kokinı
´zo ‘become/make red’ (ko
´kino ‘red’) and EN to redden ‘become/make red’,
etc. Possible verbs/neologisms are adjacent to this primary core, cf. MG movı
´zo
‘become/make purple’ (mov ‘purple’) and EN to pinken ‘become/make pink’, etc.
(Stages VI–VII). In the case of MG, very fresh neologisms are far away from a cognitive
core, cf. ladhı
´zo ‘become/make olive-green’ (ladhı
´‘olive-green’), etc. (Stage VIII +),
whereas not possible/forced verbs are cognitively dissociated, cf. !trapezı
´zo or !potirı
´zo,
whose bases trape
´zi ‘table’ and potı
´ri ‘glass’, respectively, refer to objects without
a characteristic colour, etc.
Table 3 summarizes the respective patterns.
Concluding, both BCTs and derived colour verbs emerge over stages in a
cognitive domain. Systematic gaps linked to the derivational causative/inchoative
Table 3. Evolutionary sequence of colour verbs in Modern Greek and English.
PRIMARY BCTs
(Stages (I–V)
Possible/established verbs
Primary cognitive core
MG
EN
kokinı
´zo ‘become/make red’
(ko
´kino ‘red’)
to redden ‘become/make red’
NON-PRIMARY
(DERIVED) BCTs
(Stages VI–VII)
Possible verbs/neologisms
Adjacent to a primary cognitive
core
MG
EN
movı
´zo ‘become/make purple’
(mov ‘purple’)
to pinken ‘become/make pink’
NON-BCTs
(Stage VIII +)
Very fresh neologisms
Far away from a cognitive core
MG ladhı
´zo ‘become olive-green’
(ladhı
´‘olive-green’)
N/A Not possible/forced verbs
Cognitively dissociated
MG !trapezı
´zo (trape
´zi ‘table’)
!potirı
´zo (potı
´ri ‘glass’)
144 C H A R I T O N C H A R I T O N I D I S
ProofsProofs
paradigm are due to the semantics of the base rather than deficiencies in the base’s
morphophonological character. To give an answer to the chicken-or-the-egg dilemma
mentioned in the Introduction: morphophonology does not define existing semantic/
cognitive spectra but simply ornaments them.
50
Acknowledgments
I would like to thank the anonymous referees of Word Structure for their helpful and
constructive comments. They contributed to a thorough reworking of the first
submission. My special thanks to the anonymous referee †2 for inciting me to adopt
a cross-linguistic view.
I would also like to thank Claudia Wild and Natasha Turner for evaluating the full
set of EN colour verbs as regards their current use.
Abbreviations
AAdjective
Aante
A/N Author’s note
AG Ancient Greek
amb Categorial ambiguity
B&K Berlin & Kay (1969)
BCT Basic colour term
BNC British National Corpus
Ccirca
CHAR Characteristic colour
COCA Corpus of Contemporary
American English
DCMG Dictionary of Common
Modern Greek
DMGL Dictionary of Modern Greek
Language
EN English
HL Hellenistic Greek
HNC Hellenic National Corpus
IC Inflectional classa
IEPS Inferable Eventual Position or State
LCS Lexical Conceptual Structure
LOC Location
LX Lexicalization
MD Medieval Greek
MD +1800– A. D.
MG Modern Greek
NNoun
N/A Not applicable
NCS Natural Color System
NL Neologism
NP Nominal phrase
OE Old English
OED Oxford English Dictionary
WWW World Wide Web
Notes
1. We would like to thank two anonymous referees whose comments and suggestions helped
us improve the original manuscript.
2. This paper is part of the self-funded project ‘The Integration of Socio-expressive Meaning
Into Verb Structures’ conducted by the author at the University of Cologne, Germany.
Parts of this paper can be found in a preliminary form in Charitonidis (to appear).
The revisions in the present paper concern, among others, a reordering of colour terms in
both Modern Greek and English to strictly fit the BCTs–non-BCTs sequence.
C O L O U R V E R B S I N M O D E R N G R E E K :A C O G N I T I V E A P P R O A C H 145
ProofsProofs
3. For example, Jackendoff (1983, 1990, etc.) proposes verb structures which are founded on
the visual capacity.
4. The derivational suffix is actually -ı
´z- without the inflectional vowel -o for the first person
singular – the standard citation form for MG verbs. In the following, I give all verbal forms
together with the inflectional ending.
5. Unfortunately, there are no systematic accounts on both the metrical form and the
phonological boundaries of morphological units involved in MG verbal derivation.
6. The high productivity of -ı
´zo is also pointed out in Efthymiou et al. (2012).
7. In Charitonidis (2005) the auto category is a subclass of the inchoative category.
8. This definition of macro-categories is taken from Biggam (2004: 21).
9. Figure 1 displays the most common route (Trajectory A). A more elaborate route with three
trajectories can be found in Kay & Maffi (1999: 751), Kay & Maffi (2005: 536).
10. For details about the Munsell, Color-aid, and Natural Colour System see Androulaki et al.
(2006: 44–45).
11. The suffix -ino has lost its representational strength in the frequently used BCTsko
´kino,
pra
´sino, and kı
´trino (Anastasiadis-Simeonidis 1987: 401; see also Alexandris 2009). It should
be noted that the bases kı
´tro ‘citron’ in kı
´trino and pra
´so ‘leek’ in pra
´sino are MG nouns in
common use – as opposed to the AG noun ko
´kos ‘pit’, ‘stone’ in the base of ko
´kino which
nowadays has a dissociated meaning.
12. A similar code switch for the expression of BCTs after yellow is evident in Maltese
vernacular. In this language, the first five colour terms come from Arabic. They are
followed by colour terms of Italian provenance (Borg 2011: 77); see also n. 44.
13. In Androulaki et al. (2006) the form kafetı
´is not discussed.
14. The default ending -ı
´is likely to come from -i- in the Turkish word mavi (DMGL,
p. 1996). In contemporary MG, the word mavı
´(approximately meaning ‘purple’) is only
marginally used. The masculine form of the ending -ı
´is -ı
´sand the feminine form is -ia
´.
It should be noted that the assignment of the adjectives in -ı
´s, -ia
´, -ı
´to a declensional
paradigm is not an easy task (DMGL, pp. 1995–1996; see also Anastasiadis-Simeonidis 1987:
397–398).
15. The stipulation of two BCTs for BLUE violates the non-inclusiveness criterion (criterion ‘ii’ in
Berlin & Kay 1969: 6) since LIGHT BLUE is included in BLUE – cf. a similar discussion about
the status of lacivert ‘dark blue’ in Turkish in Ra
¨tsep (2011: 144).
16. As regards the history of language and culture, the difficulty to define a BCT for blue in MG
is partially due to historical and cultural reasons since the status of the BLUE category
was traditionally uncertain in Greek (cf. sections 5.1 and 5.4). As Lyons (1999: 63) notes:
‘It is not at all clear, however, that Ancient Greek had a word for the sixth colour in the
BK-hierarchy [Berlin & Kay 1969, A/N]: blue. In fact, there are serious, and perhaps
insoluble, prob-lems relating to Ancient Greek words that denote colours in the
blue-purple area of the spectrum. At least three, and possibly four, words have to be
considered as basic level-1 colour terms: halourgos (usually translated “purple”), kuaneos
(... usually translated “dark blue”), orphninos (“violet”?) and possibly glaukos.’ (Lyons
1999: 63; in Borg 2011: 85). For more details on AG colour terms, see Smitherman (2013).
17. A summary of objections to early research methodology can be found in Biggam (2012:
84–85).
18. In a similar line of argument, Biggam (2004: 37) points out that the study of a particular
language should be led by evidence from the language itself and not by any theory of
categorization.
146 C H A R I T O N C H A R I T O N I D I S
ProofsProofs
19. The bipartite structures in (3) and (4) are general representations of causative/inchoative
verbs within Lieber’s (2004) framework. The content of a colour term shows up in the
[+Loc] slot and refers to an end state in a situation. ‘IEPS’ (‘Inferable Eventual Position or
State’) refers to a sequence of places or states (progression). The indices iand jco-index the
agent and patient arguments. For further details about the composition of the skeletons in
(3) and (4), see Lieber (2004: 81–86).
20. In Lieber’s (2004, 2007) theory, colour refers to ‘pragmatic body’, i.e. an extra semantic
representation parallel to grammatical ‘semantic skeletons’. Author’s ‘socio-expressive tier’
(Charitonidis 2012a, 2012b, 2013, etc.) refers to this pragmatic body as well.
21. All stages beyond Stage VII are henceforth referred to as ‘Stage VIII +’.
22. Efthymiou (2011) argues that -ı
´zo avoids attaching to bases ending in /z/. See, however,
?bezı
´zo in the fifth section of Table 1 whose base bez ‘beige’ ends in /z/ (http://www.
mammyland.com/forum-neo/viewtopic.php?f=29&t=19747&start=700&view=print,
http://www.teleiosgamos.gr/forum/viewtopic.php?f =8&t =11158&start =765&hilit =2013,
etc. [found in www by using Google; accessed September 4
th
, 2012]).
23. In both DCMG and DMGL, the chronological indication of the entries is inconsistent.
The periods AG, HL, MD, and MD +indicated in Table 1 were mainly defined according to the
introductory sections in DMGL (pp. 20–21) and DCMG (pp. xxi–xxiii). MD +(‘after medieval
Greek’) refers to the indication neo
´teri perı
´odhos ‘later period’ in DCMG.
24. bla
´vo is an adaptation of the old Italian word blavo (DCMG). It should be noted that bla
´vo
refers to a difference in lightness conditions like a
´spro and ma
´vro.
25. To this effect, asimı
´zo follows the pattern of xrisı
´zo, cf. the preceding discussion.
26. The stem of these adjectives ends in a consonant. This pattern is in accord with
Efthymiou’s (2011) statement, that -ı
´zo attaches exclusively to consonant-final base stems.
The same pattern cannot be regarded as a condition on the derivation of COLOUR verbs
since the number of vowel-final adjectival stems in MG is very small (see Holton et al. 1997:
73–89).
27. The elaboration of inflection for gender and number in the primary BCTs is also evident in
Maltese vernacular. In this language, causative verb forms are available exclusively for
the root morphemes byd ‘white’, swd ‘black’, h
_mr ‘red’, and h
_dr ‘green’ (Borg 2011: 81; cf.
n. 44).
28. It should be noted that there are no native Greek nouns or adjectives ending in -v, or -z.
29. The existence of such a derived adjective is not an absolute requirement for the emergence
of a colour verb, cf. the existing novel formation ?ghalatı
´zo ‘have a milky hue’ derived
directly from N gha
´la ‘milk’ and not from an alleged A *ghalatı
´(‘milky’), etc. It should be
noted that Berlin and Kay (1969) and various studies thereafter do not regard as BCTs words
derived from the name of an object having a characteristic colour (cf. the criticism on this
method in Lucy 1997).
30. http://www.rdeco.gr/xtizw/chroma, http://www.snhell.gr/testimonies/content.asp?id=165&
author_id=103, etc. (found in WWW by using Google; accessed November 16
th
, 2012).
31. http://www.lykeionellinidon.gr/portal/View_object.aspx?id=14298&lang=gr, http://www.
modelclub.gr/forums/index.php?topic =1632.0, etc. (found in WWW by using Google;
accessed November 16
th
, 2012).
32. As an anonymous referee noted, asimo
´no is a well attested verb, meaning ‘to cause to
shine like silver’ or ‘to cover in silver’. The former reading shows up only in literary or
metaphorical contexts. The latter reading refers to a strong [ +Location] component in -o
´no
which standarly promotes themes (see Charitonidis 2011: 34). As opposed to asimı
´zo,
C O L O U R V E R B S I N M O D E R N G R E E K :A C O G N I T I V E A P P R O A C H 147
Revised in the
final version!
Revised in the
final version!
ProofsProofs
asimo
´no has no inchoative variant – a pattern which sets apart this verb from the other
colour verbs which are primarily inchoatives. The same argumentation applies to xriso
´no,
the rival form of xrisı
´zo, meaning ‘to cause to shine like gold’ or ‘to cover in gold’.
33. xrisafı
´zo is derived from the adjective xrisafı
´‘gold-coloured’ and often used parallel to
xrisı
´zo ‘become/make golden’ (sections 5.2 and 5.4).
34. http://forum.scooterclubhellas.gr/viewtopic.php?f=23&t=23407, http://www.avclub.gr/
forum/showthread.php/99761-The-Kenwood-K%CE%91–305-Adventure/page2, etc.
(found in www by using Google; accessed November 16
th
, 2012).
35. http://julia-sevenstars.blogspot.de/2008_12_01_archive.html, http://digitalschool.minedu.
gov.gr/modules/document/file.php/DSGYM-B120/HTML/g02-web/htms/sel_food_
2c.htm (found in WWW by using Google; accessed November 16
th
, 2012).
36. OED definition of porphura: ‘A shade of crimson, spec. (also Tyrian purple) the colour of a
dye obtained from various gastropod molluscs and traditionally used for fabric worn by
people of imperial or royal rank’ (in OED online, entry ‘purple’).
37. Colour terms referring to precious metals or substances tend to appear as BCTs in many
languages. By examining type modification in Modern English colour adjectives, Steinvall
(2002) argues: ‘there are the six Primary BCTs and to a certain degree also brown, grey, golden
and silver’ (Steinvall 2002: 122).
38. An extensive internet search for the corresponding EN terms and the consultation of
bilingual dictionaries such as OGELD were not conclusive. One reason for this may be the
different lightness conditions between the North European (cf. English) and the Southern
European (cf. Greek) area which make a literal translation of the colour terms impossible.
Most notably, such equivalence issues arise overproportionally within the non-BCTs (Stage
VIII +; cf. the discussion on best-example choices in section 2). Filopoulos (1994: 197–198)
underlines the absence of absolute correspondence of colour terms in different languages,
while pointing out the influence of environmental and cultural factors on the semantic value
of colour terms.
39. Casson (1997: 233) reports that ‘the first secondary terms were incorporated into the
[English, A/N] language in the late Middle English period (between 1350 and 1500). Twelve
terms developed hue senses at this time, of which nine were names for dyestuffs, pigments,
and textiles (and earlier entities from which they derived): gold, silver, violet, azure, crimson,
russet, ocher, scarlet, and vermillion.’ [Casson’s italics].
40. BNC contains ca. 100.000.000 words (1980s–1993) and COCA contains ca. 450.000.000 words
(1990–2012); www.corpus.byu.edu (as of November 15, 2012).
41. In the results it was not always possible to distinguish between the use of a term as an
adjective or a verb, cf. sentences such as her lips blue (BNC/A fatal inversion. Vine, Barbara.
London: Viking, 1987, pp. 19–117), etc. In such cases, OED was consulted for a more precise
categorial assignment of the terms.
42. I have some reservations about the actual use of the inchoative variant of blue in common
EN (see n. 39).
43. The verb *to orange would be a possible verb according to the ‘Innovative denominal verb
convention’ in Clark & Clark (1979: 787). As an internet user in the ‘English Language
and Usage’ forum reports, ‘I could see an orange grower saying, “The trees should orange
by next month,” meaning their current green fruit will shine bright orange in a few weeks
or so...’ (http://english.stackexchange.com/questions/80770/to-orange-and-to-pink).
44. The sixth criterion in Berlin & Kay (1969: 6) for the exclusion of a word as a BCT is as
follows: ‘Colour terms that are also the name of an object characteristically having that
148 C H A R I T O N C H A R I T O N I D I S
ProofsProofs
colour are suspect, for example, gold, silver, and ash. This subsidiary criterion would
exclude orange, in English, if it were a doubtful case on the basic criteria (i–iv).’ [B&K’s
italics].
45. With a few exceptions, the same applies to the verbs in sections 5 and 6 of Table 2.
46. As Onysko and Michel (2010: 3) argue, the rule of deadjectival verbal suffixation in
EN cannot explain why ?bluen is not acceptable as the other established colour verbs in -en.
47. The inflectional paradigm of the neuter adjective xrisafı
´‘golden’ in (i) below serves as a
representative example. As already mentioned in section 3, the -ı
´form is more widely
used – in many cases uninflected for all genders, in singular and plural NPs. Similar
restrictions on use apply to the other IC6 colour-adjectives. For more details about the
ending -ı
´,see sections 3 and 5.3.
(i) SINGULAR PLURAL (IC6; see Ralli 2005: 120)
NOM xrisafı
´xrisafi-a
´
GEN xrisafi-u´ xrisafi-o
´n
ACC xrisafı
´xrisafi-a
´
VOC xrisafı
´xrisafi-a
´
48. As already mentioned in section 5.3 the suffix -ı
´zo is phonologically more akin to the
endings -ı
´s, -ia
´, -ı
´than the other verb-forming suffixes because it contains a stressed -i-.
49. The verbs *to orange and *to pink (Stage VII) are not attested in OED. For the verb
*to orange see n. 41. For the exceptional verb to pinken see sections 6.1 and 6.2.
50. This observation is akin to the doctrines of Distributed Morphology which allocates a great
amount of non-generative morphological operations to the phonological component (PF)
after the main syntactic operations (see Embick & Noyer 2006).
References
Alexandris, Katerina 2009. The morphology of colour terms in Modern Greek. In Proceedings of
the 7th Conference ‘Greek Language and Terminology’, October 22–24, Athens.
Allwood, Jens, Lars-Gunnar Andersson & O
¨sten Dahl 1977. Logic in linguistics. Cambridge:
Cambridge University Press.
Al-Rasheed, Abdulrahman S., Humood H. Al-Sharif, Mohammed J. Thabit, Norah S.
Al-Mohimeed & Ian R. L. Davies 2011. Basic colour terms of Arabic. In Carole P. Biggam,
Carole A. Hough, Christian J. Kay & David R. Simmons (eds.), New directions in colour
studies. Amsterdam: Benjamins. 53–58.
Anastasiadis-Simeonidis, Anna 1987. On
omata dhlwtik
a cr
wmatoV thV Koin
hV
Neoellhnik
hV.Sumbol
h lexikologik
h[Nominals denoting colour in Common
Modern Greek. A lexicological approach]. Studies in Greek Linguistics 8: 391–413.
Androulaki, Anna, Natalia Go
ˆmez-Pestan
˜a, Christos Mitsakis, Julio Lillo Jover, Kenneth
Coventry & Ian Davies 2006. Basic colour terms in Modern Greek: Twelve terms including
two blues. Journal of Greek Linguistics 7: 3–47.
Babiniotis, Georgios 1998. Le xik o
´t hV N
eaV Ellhn ik
hV G l
wssaV [Dictionary of
Modern Greek Language/DMGL]. Athens: Kedro Leksikologias.
Berlin, Brent & Paul Kay 1969. Basic color terms: Their universality and evolution. Berkeley, CA:
University of California Press.
Biggam, Carole P. 2004. Prototypes and foci in the encoding of colour. In Christian J. Kay &
Jeremy J. Smith (eds.), Categorization in the history of English. Amsterdam: Benjamins. 19–40.
C O L O U R V E R B S I N M O D E R N G R E E K :A C O G N I T I V E A P P R O A C H 149
ProofsProofs
Biggam, Carole P. 2012. The semantics of colour: A historical approach. Cambridge, UK:
Cambridge University Press.
Biggam, Carole P., Carole A. Hough, Christian J. Kay & David R. Simmons (eds.) 2011. New
directions in colour studies. Amsterdam: Benjamins.
Borg, Alexander 2011. Towards a diachrony of Maltese basic colour terms. In Carole P. Biggam,
Carole A. Hough, Christian J. Kay & David R. Simmons (eds.), New directions in colour
studies. Amsterdam: Benjamins. 73–90.
Casson, Ronald W. 1997. Color shift: Evolution of English color terms from brightness to hue.
In C. L. Hardin & Luisa Maffi (eds.), Color categories in thought and language. Cambridge:
Cambridge University Press. 225–239.
Charitonidis, Chariton 2005. Verb derivation in Modern Greek: Alternation classes, conceptual
structures, semantic fields. Doctoral dissertation, University of Cologne. In Europa
¨ische
Hochschulschriften, Reihe 21: Linguistik, Band 284. Frankfurt: Peter Lang.
Charitonidis, Chariton 2007. Greek -ı
´zo derivatives: A conceptual analysis. Linguistik Online
30: 1/2007.
Charitonidis, Chariton 2011. Making verbs happen: Interviews on Greek verb endings
[=Languages of the World 43]. Mu
¨nchen: Lincom Europa.
Charitonidis, Chariton 2012a. The coordinative structure of the Greek -(i)a
´zo verbs.
Linguistische Berichte 231: 299–315.
Charitonidis, Chariton 2012b. The interaction between affixes and bases: The case of the
Greek verb suffixes. In Zoe Gavriilidou, Angeliki Efthymiou, Evangelia Thomadaki
& Penelope Kambakis-Vougiouklis (eds.), Selected Papers of the 10th International
Conference on Greek Linguistics. Komotini, Greece: Democritus University of Thrace.
202–212.
Charitonidis, Chariton 2013. On the denotational and socio-expressive properties of the
Greek verbal derivatives in apo-,ek(s)-, and kse-.Bucharest Working Papers in Linguistics
15(1): 79–112.
Charitonidis, Chariton (to appear). Colour verbs in Modern Greek and English. In Proceedings
of the 21st International Symposium on Theoretical and Applied Linguistics. Aristotle University
of Thessaloniki.
Clark, Eve V. & Herbert H. Clark 1979. When nouns surface as verbs. Language 55: 767–811.
Conklin, Harold C. 1955 [1964]. Hanuno
´o color categories. In Dell Hymes (ed.), Language
in culture and society: A reader in linguistics and anthropology New York: Harper & Row.
189–192. (From Southwestern Journal of Anthropology 1955, 11(4): 339–344).
Corbett, Greville G. & Gerry Morgan 1988. Colour terms in Russian: Reflections of typological
constraints in a single language. Journal of Linguistics 24: 31–64.
Davies, Ian & Greville Corbett 1994. The basic color terms of Russian. Linguistics 32: 65–89.
Efthymiou, Angeliki 2011. The semantics of verb forming suffixes in Modern Greek.
In Proceedings of the 19th International Symposium of Theoretical and Applied Linguistics,
3–5 April 2009, School of English, Aristotle University of Thessaloniki. 174–184.
Efthymiou, Angeliki, Georgia Fragaki & Angelos Markos 2012. Productivity of verb-forming
suffixes in Modern Greek: a corpus-based study. Morphology (online) DOI doi: 10.1007/
s11525-012-9202-4.
Embick, David & Rolf Noyer 2006. Distributed morphology and the syntax-morphology
interface. In Gillian Ramchand & Charles Reiss (eds.), The Oxford handbook of linguistic
interfaces. Oxford: Oxford University Press. 289–324.
150 C H A R I T O N C H A R I T O N I D I S
ProofsProofs
Filopoulos, Vassilis 1994. Sumbol
hsth dhmiourg
ia ellhnik
hV orolog
iaV crwm
atwn
[Towards the creation of a Greek colour terminology]. Terminologie et Traduction, 2,
Comissio
´n Europe
´enne. 179–202.
Haspelmath, Martin & Andrea D. Sims 2010. Understanding morphology. London: Hodder.
Holton, David, Peter Mackridge & Irene Philippaki-Warburton 1997. Greek: A comprehensive
grammar of the modern language. London: Routledge.
Jackendoff, Ray 1983. Semantics and cognition. Cambridge, MA: MIT Press.
Jackendoff, Ray 1990. Semantic structures. Cambridge, MA: MIT Press.
Jameson, Kimberly & Roy G. D’Andrade. 1997. It’s not really red, green, yellow, blue: An
inquiry into perceptual color space. In C. L. Hardin & Luisa Maffi (eds.), Color categories
in thought and language. Cambridge: Cambridge University Press. 295–319.
Kay, Paul & Chad K. McDaniel 1978. The linguistic significance of the meanings of basic color
terms. Language 54(3): 610–646.
Kay, Paul & Luisa Maffi 1999. Color appearance and the emergence and evolution of basic color
lexicons. American Anthropologist 101(4): 743–760.
Kay, Paul & Luisa Maffi 2005. Colour terms. In Martin Haspelmath, Matthew S. Dryer, David
Gil & Bernard Comrie (ed.), The world atlas of language structures. Oxford: Oxford University
Press. 534–545.
Lexik
othV Koin
hV Neoellhnik
hV [Dictionary of Common Modern Greek/dcmg] 1999.
Thesaloniki: Institute of Modern Greek Studies.
Lieber, Rochelle 2004. Morphology and lexical semantics. Cambridge, UK: Cambridge University
Press.
Lieber, Rochelle 2007. The category of roots and the roots of categories: What we learn from
selection in derivation. Morphology 16(2): 247–272.
Lo
¨bner, Sebastian 2002. Understanding semantics. London: Arnold.
Lucy, John A. 1991. How to wreak havoc with a category: The neglect of linguistic form in the
analysis of categories. Paper presented to the Second Bi-Annual Meetings of the Society for
Psychological Anthropology.
Lucy, John A. 1992. Language diversity and thought: A reformulation of the linguistic relativity
hypothesis. Cambridge: Cambridge University Press.
Lucy, John A. 1997. The linguistics of ‘color’. In C. L. Hardin & Luisa Maffi (eds.), Color
categories in thought and language. Cambridge: Cambridge University Press. 320–346.
Lyons, John 1999. The vocabulary of color with particular reference to Ancient Greek and
Classical Latin. In Alexander Borg (ed.), The language of color in the Mediterranean: An
anthology on linguistic and eth-nographic aspects of color terms. Stockholm: Almqvist & Wiksell.
38–75.
Onysko, Alexander & Sascha Michel 2010. Unravelling the cognitive in word formation.
In Alexander Onysko & Sascha Michel (eds.), Cognitive perspectives on word formation. Berlin:
de Gruyter. 1–25.
O
¨zgen, Emre & Ian R. L. Davies 1998. Turkish color terms: Tests of Berlin and Kay’s theory of
color universals and linguistic relativity. Linguistics 36(5): 919–956.
Paramei, Galina V. 2007. Russian ‘blues’: Controversies of basicness. In Robert E. MacLaury,
Galina V. Paramei & Don Dedrick (eds.), Anthropology of color: Interdisciplinary multilevel
modeling. Amsterdam: Benjamins.
Pearsall, Judy (ed.) 1999. The New Oxford Dictionary of English [OED]. Oxford: Oxford
University Press. Online version: www.oed.com.
C O L O U R V E R B S I N M O D E R N G R E E K :A C O G N I T I V E A P P R O A C H 151
ProofsProofs
Plag, Ingo 1999. Morphological productivity. Structural constraints in English derivation. Berlin:
Mouton de Gruyter.
Plag, Ingo 2003. Word-formation in English. Cambridge Textbooks in Linguistics. Cambridge:
Cambridge University Press.
Ralli, Angeliki 2005. Morjolog
ia [Morphology]. Athens: Patakis.
Ra
¨tsep, Kaidi 2011. Preliminary research on Turkish basic colour terms with an emphasis on
blue. In Carole P. Biggam, Carole A. Hough, Christian J. Kay & David R. Simmons (eds.),
New directions in colour studies. Amsterdam: Benjamins. 133–145.
Regier, Terry, Paul Kay & Richard S. Cook 2005. Focal colors are universal after all. In
Proceedings of the National Academy of Sciences 102(23): 8386–8391.
Sandford, Jodi L. 2011. Warm, cool, light, dark, or afterimage: Dimensions and connotations
of conceptual color metaphor/metonym. In Carole P. Biggam, Carole A. Hough, Christian J.
Kay & David R. Simmons (eds.), New directions in colour studies. Amsterdam: Benjamins.
205–218.
Smitherman, Thomas 2013. Color terms. In Georgios K. Giannakis (ed.), Encyclopedia of
Ancient Greek language and linguistics. BrillOnline (www.brillonline.nl).
Stanulewicz, Danuta & Adam Pawłowski 2011. Kashubian colour vocabulary. In Carole P.
Biggam, Carole A. Hough, Christian J. Kay & David R. Simmons (eds.), New directions in
colour studies. Amsterdam: Benjamins. 105–119.
Stavropoulos, D. N. 1988. Oxford Greek-English learner’s dictionary [OGELD]. Oxford: Oxford
University Press.
Steinvall, Anders 2002. English colour terms in context. Skrifter fra
˚n moderna spra
˚k 3. Umea
˚
Universitet (Sweden).
Sutrop, Urmas 2011. Towards a semiotic theory of basic colour terms and the semiotics of Juri
Lotman. In Carole P. Biggam, Carole A. Hough, Christian J. Kay & David R. Simmons
(eds.), New directions in colour studies. Amsterdam: Benjamins. 39–48.
Wierzbicka, Anna 1990. The meaning of color terms: Semantics, culture, and cognition.
Cognitive Linguistics 1(1): 99–150.
Wierzbicka, Anna 1996. Semantics: Primes and universals. Oxford: Oxford University Press.
Author’s address: (Chariton Charitonidis)
Allgemeine Sprachwissenschaft
Institut fu
¨r Linguistik
Universita
¨t zu Ko
¨ln
D-50923 Ko
¨ln
Germany
E-mail: charitonidis@uni-koeln.de
152 C H A R I T O N C H A R I T O N I D I S
