SKY Journal of Linguistics 20 (2007), 367–397
The Basic Colour Terms of Finnish1
This article describes a study of Finnish colour terms the aim of which was to establish
an inventory of basic colour terms, and to compare the results to the list of basic terms
suggested by Mauno Koski (1983). Basic colour term in this study is understood as
Brent Berlin and Paul Kay defined it in 1969. The data for the study was collected using
the field method of Ian Davies and Greville Corbett (1994). Sixty-eight native speakers
of Finnish, aged 10 to 75, performed two tasks: a colour-term list task (name as many
colours as you know) and a colour naming task (where the subjects were asked to name
65 representative colour tiles). The list task was complemented by the cognitive salience
index designed by Sutrop (2001). An analysis of the results shows that there are 10
basic colour terms in Finnish—punainen ‘red,’ sininen ‘blue,’ vihreä ‘green,’ keltainen
‘yellow,’ musta ‘black,’ valkoinen ‘white,’ oranssi ‘orange,’ ruskea ‘brown,’ harmaa
‘grey,’ and vaaleanpunainen ‘pink’. These results contrast with Mauno Koski’s claim
that there are only 8 basic colour terms in Finnish. However, both studies agree that
Finnish does not possess a basic colour term for purple.
Basic colour terms are a relatively well studied area of vocabulary and
studies on them cover many languages of the world. Research on colour
terms became particularly intense after the publication of Berlin and Kay’s
(1969) inspiring and much debated monograph.
Berlin and Kay argued that basic colour terms in all languages are
drawn from a universal inventory of just 11 colour categories (see Figure
1). According to their theory, every language has between 2 and 11 basic
colour terms, and they present a hierarchy which specifies a limited number
of evolutionary paths that a language can take when adding new colour
categories. Languages start with two basic colour terms: BLACK and
1 This study was supported by the Estonian Science Foundation grant no. 6744. I am
very grateful to Prof. Urmas Sutrop for his extensive help and guidance and for his
useful remarks on this article, and to two anonymous reviewers for their comments.
WHITE (capitals denote the hypothetically universal colour categories); the
third term to be aquired is RED; the fourth term is either GREEN or
YELLOW; the fifth term is whichever of GREEN or YELLOW is missing;
the sixth term is BLUE, and so on. If a language has a particular basic
colour term, then it should also already entail all the earlier basic colour
terms of the hierarchy.
[white] [green] → [yellow]
→ [red] → → [blue] → [brown] →
[black] [yellow] → [green]
I II III IV V VI VII
Figure 1. Temporal-evolutionary ordering of basic colour terms after Berlin and Kay
(1969). The Roman numbers indicate the corresponding evolutionary stage.
This hierarchy has been modified since Berlin and Kay’s original study,
concerning precisely the earlier stages of development (see Kay 1975, Kay
and McDaniel 1978, Kay et al. 1991, Kay et al. 1997 etc.).
Basic colour term was defined by Berlin and Kay as follows (1969:
5–7) and will be used in this article accordingly (some examples from
Finnish are added):
1) It is monolexemic; that is, its meaning is not predictable from the
meaning of its parts, e.g. the colour name harmaa ‘grey’ in Finnish;
2) Its signification is not included in that of any other colour term, e.g.
ruusunpunainen ‘rose red’ and viininpunainen ‘wine red,’ which are
two kinds of red for most speakers of Finnish;
3) Its application is not included in that of any other colour term, e.g.
kastanja ‘chestnut,’ which may be predicated only for hair;
4) It must be psychologically salient for informants. Indices of
psychological salience include, among others, a) a tendency to occur at
the beginning of elicited lists of colour terms, b) stability of reference
across informants and occasions of use, c) occurrence within the
idiolects of all informants. Examples of this criterion may be punainen
‘red,’ keltainen ‘yellow,’ etc.
The above list consists of what are considered the primary criteria. When
the colour name is still in doubt, the following subsidiary criteria should
also be considered:
5) The dubious form should have the same distributional potential as the
previously established basic terms;
THE BASIC COLOUR TERMS OF FINNISH
6) Colour terms that are also names of an object are suspect. This criterion
would exclude orange in English, if it were a dubious case on the basic
criteria (1)–(4). Some Finnish examples are persikka ‘peach’ and luumu
7) Recent foreign loanwords may be suspect, e.g kretliini ‘violet’.
8) In cases where lexemic status is difficult to access [criterion (1)],
morphological complexity is given some weight as a secondary
criterion. For example, the Finnish term sinivihreä ‘blue green’ might
be eliminated by this criterion.
Berlin and Kay studied 98 languages in total, and they also collected
primary experimental data for 20 languages. The only Finno-Ugric (Uralic)
language represented in their study was Hungarian, which they erroneously
classified as Altaic, and they suggested that it has, exceptionally, 12 basic
colour terms, including two reds—piros and vörös (which they glossed to
English as light red and dark red, respectively) (Berlin and Kay 1969: 35–
36). However, as was often the trouble with their field work, the number of
subjects was insufficient (in fact they only had one Hungarian subject).
Uusküla and Sutrop have subsequently carried out a field study with 125
Hungarian native speakers, the results of which have allowed them to argue
that in fact there are 11, and not 12 basic colour terms in Hungarian
(Uusküla and Sutrop 2007, Bogatkin-Uusküla and Sutrop 2005a).
In 1983, Mauno Koski discussed Berlin and Kay’s theory with
reference to the Finnish language. His seminal monograph explored all
Finnic languages, including Finnish, Estonian, South Estonian, Livonian,
Karelian, Ludic, Veps, Votic and Ingrian. Koski made the first attempt to
investigate the inventory of basic colour terms in Finnic languages, also
categorising wide etymologies of colour names and colour verbs. However,
he only worked with dictionaries and dialect collections, without
employing any field methods.
The first empirical study of a Finno-Ugric language, carried out with a
sufficient number of subjects (80), using precise colour stimuli and set
within the theoretical framework provided by Berlin and Kay, was a study
of the Estonian language undertaken by Urmas Sutrop (1995, 2000a, 2002).
According to Sutrop, Estonian possesses exactly 11 basic colour terms—
must ‘black,’ valge ‘white,’ punane ‘red,’ kollane ‘yellow,’ roheline
‘green,’ sinine ‘blue,’ pruun ‘brown,’ hall ‘grey,’ lilla ‘purple,’ roosa
‘pink’ and oranž ‘orange’. Koski, however, had proposed that there are
only 10 basic colour terms in Estonian, excluding the term oranž ‘orange’
from his list of basic terms (Koski 1983). Recently, as mentioned above,
Hungarian has also been studied empirically (Bogatkin-Uusküla and Sutrop
2005a, Uusküla and Sutrop 2007). Some examples of Finno-Ugric colour
term studies carried out with other methods include: 1) a minor empirical
study of Mansi (Vogul), a Ugric language (Sipőcz 1994), which is mainly
based on literature, although 50 non-standard colour circles were shown to
three female native speakers who were asked to name the colour; 2) a
linguistic study on the colour terminology of Estonian dialects (Oja 2001);
3) a linguistic study on Moksha Mordvin (Turunen 2002).
To date, Finnish colour terms have not been investigated with field
methods. Since the method for establishing basic colour terms used by
Berlin and Kay is very complicated and time-consuming when used on a
large, albeit sufficient, number of subjects (Berlin and Kay 1969: 5–7), Ian
Davies and Greville Corbett have proposed a new field method based on
Berlin and Kay’s original procedure (Davies and Corbett 1994, 1995). This
makes the interviews easier to conduct and limits them to approximately
20–40 minutes each, depending on the language. Many European
languages, like Russian (Davies and Corbett 1994), English (Davies and
Corbett 1995), Estonian (Sutrop 1995, 2001, 2002), Hungarian (Bogatkin-
Uusküla and Sutrop 2005a, Uusküla and Sutrop 2007), Turkish (Özgen and
Davies 1998), and Catalan (Davies et al. 1995) as well as many exotic
languages (e.g. Davies et al. 1992, Davies et al. 1994) have been studied
with this field method.
The present study was carried out to establish the basic colour terms
of Finnish with a particular interest in whether there are 11 basic colour
terms or only 8 basic terms—valkoinen (valkea) ‘white,’ musta ‘black,’
punainen ‘red,’ vihreä ‘green,’ keltainen ‘yellow,’ sininen ‘blue,’ ruskea
‘brown,’ harmaa ‘grey’—as Koski has argued (1983). No experimental
study has yet been carried out to examine which of the two terms valkoinen
and valkea is the basic term for white, or whether the colour terms violetti
‘purple’ and oranssi ‘orange’ are basic colour terms, or whether there is a
basic term to indicate pink colour in Finnish. The method of Davies and
Corbett was complemented with the cognitive salience index, which for a
colour term is calculated from its frequency and mean position, and which
can also be used to discriminate basic terms from non-basic ones (Sutrop
2001, 2002: 35).
THE BASIC COLOUR TERMS OF FINNISH
2. Case study: Finnish colour terms
Language: Finnish, Finnic, Finno-Ugric, Uralic.
Regions where the data have been collected: Helsinki, Turku,
Lempäälä, Espoo and Tuusula, all in Finland.
Dates: 1–7 August 2005 (Helsinki, Turku and Lempäälä); 7–14
September 2006 (Helsinki, Espoo and Tuusula).
Subjects: There were 68 subjects in total2, 42 female and 26 male,
whose age ranged from 11 to 75, with a mean of 39.5 years. The age of
men ranged from 10 to 75 years with a mean of 41.4 years, and the age of
women ranged from 11 to 70 years with a mean of 38.4 years.
In 2005 I interviewed 29 subjects, 19 of which were female and 10
male, their age ranging from 11 to 75, with a mean of 42.7 years. The age
of men ranged from 16 to 75 with a mean of 44.9 years, and the age of
women ranged from 11 to 68 with a mean of 41.5 years.
In 2006 I interviewed 39 subjects in total, 23 female and 16 male,
whose age ranged from 10 to 70 with a mean of 37.2 years. The women’s
age ranged from 17 to 70 with a mean of 35.8 years and the men’s age
ranged from 10 to 70 with a mean of 39.2 years.
The subjects were from the following locations (in alphabetical order):
Espoo, Helsinki, Hämeenlinna, Impilahti, Jyväskylä, Lohja, Kaarina,
Kiiminki, Kirkkonummi, Korppilahti, Kouvola, Lempäälä, Mikkeli,
Mouhijärvi, Orimattila, Oulainen, Oulu, Pieksämäki, Pori, Riihimäki,
Rovaniemi, Salo, Savo, Savonlinna, Sipoo, Sääksjärvi, Tampere, Turku,
Tyrvää, Urjala, Uusikaupunki, Vaasa, Vammala, Vantaa, Viipuri, and
All the subjects were native speakers of Finnish, with different dialect
backgrounds; some of the subjects were unable to name their dialect, but
aware that they spoke “somewhat dialectally”. One subject used Finnish
Sign Language to communicate with her parents. All the subjects
completed the list task first and then the colour naming task. The subjects
were not informed, until the beginning of the test, that the questions would
refer to colours and colour terms.
2 In fact there were 69 subjects in total. One subject did not have a normal colour vision,
which was tested by using The City University Color Vision Test (Fletcher 1980).
Throughout this study only the responses of subjects with normal colour-seeing ability
Colour vision: All the subjects had normal colour-seeing ability. All
subjects were tested by using The City University Color Vision Test
(Fletcher 1980), where the test was conducted after the list task and before
the colour naming task. In the colour vision test, the subject is shown ten
black tiles, in the middle of which is a dot of a certain tone of colour
surrounded by four dots of colour of differently coloured dots. The
interviewee has to say which dot is the most similar to the central one:
above, down, right or left. The test makes it possible to diagnose almost all
the anomalies of colour vision like deuteronopia, protonopia, tritanopia,
The language of the interview: Finnish.
The field method. The field method proposed by Davies and Corbett (1994,
1995) is used: an interview comprises two parts, a list task and a colour
The list task. The subjects were asked to name as many colours as they
knew. All the terms were written down in the order in which the subjects
listed them. The experimenter wrote down exactly what the subjects said.
After this, the subjects were thanked and went on with the Colour Vision
Test described above, and following this, the colour naming task.
The colour naming task. The subjects were shown the 65 coloursquares
(tiles), one square at a time, in random sequence. The order was
different for each subject and the colours were shown in sufficient daylight
on a grey base. The experimenter asked, indicating each colour tile, the
unvaried question: ‘What colour is that?’ in Finnish. All the answers were
written down as said.
Stimuli. In the colour naming task, 65 standard tiles were used as
stimuli. Each tile was a 5 x 5 cm sized wooden square covered with
coloured paper. These colours had been chosen from the Color Aid
Corporation range of colour papers using the Ostwald colour system
(Ostwald 1939). The rationale for the 65 colour sample selection can be
found in Davies et al. (1992).
The Ostwald colour system. In the Ostwald colour system, the main
features of colour are colour tone i.e. hue, content of white i.e. tint and
content of black or blackness i.e. shade. The brightness of the grey scale is
also divided into eight grades according to their white and black content.
Color Aid uses a modification of the Ostwald colour system, where there
THE BASIC COLOUR TERMS OF FINNISH
are 24 chromatic colours—6 basic colours: Y–yellow, O–orange, R–red,
V–violet, B–blue, G–green and their transition tones e.g. YO–yelloworange,
YOY–yellow-orange-yellow. Every colour tone breaks down into
four light variants T1–T4, in which the share of tint (share of white)
increases pro rata, and into three dark variants S1–S3, where the shade
(share of black) increases. In addition, some extra-system colours have
been used, such as Sienna and Rose Red. Co-ordinates CIE3 of the colour
tiles used in the experiment (lightness, content of red and content of green)
are available in the study by Davies and Corbett (1994: 70–71).
4. Finnish colour terms: results
In this section the results of the Finnish colour terms are presented. First,
the list task and the colour naming task will be analysed separately, and
then the results of both tasks will be combined.
The subjects named 5876 Finnish colour terms, among which 1014
were different. All the compound names, of different types of connection,
provided by the subjects were referred to by different names (e.g. vaaleavihreä
‘light green’ and vaalean-vihreä 4 ‘light green,’ kirkas-punainen
‘bright red’ and kirkkaan-punainen ‘brightish-red,’ lehden-vihreä ‘leafish
green’ and lehti-vihreä ‘leaf green’). The phonetic variants like liila and
lila ‘lilac’ or ‘purple,’ beige and beessi5 ‘beige’ are also treated separately.
4.1 The list task
In the list task the subjects named 1506 colour terms in all, among them
332 different ones. The average list of named colours contained 13.47
entries. The lowest numbers of colour names which came to the subjects’
minds were 8 and 9, offered respectively by a 75-year-old pensioner and a
52-year-old consultant (both were men). The most colour terms, 53, were
offered by a 68-year-old retired woman, who had taught Finnish at
3 Known as the CIE 1931 color space or the CIE XYZ color space created by the
International Commission of Illumination (CIE).
4 For the purposes of easier reading and convenience, composite colour names are
written here with a hyphen [-], in contrast to the Finnish literary norm, and in order to
show the concrete meaning of the composite parts of the compound (sometimes the
hyphen is omitted, i.e. the enclitic variant is used).
5 The forms lila ‘lilac; purple’ and beessi ‘beige’ differ from the Finnish literary norm,
but are used in spoken language.
university. The second highest number, 50 colour names, was offered by a
69-year-old cameraman. Women offered more colour names than men, and
people with a higher level of education named more colour names than
those with a lower level of education.
Table 1 presents all the colour terms that the subjects named first in
the list. The colour term most frequently mentioned first for by both men
and women was punainen ‘red’ (altogether 27 times). It was followed by
musta ‘black’ (10 times) and sininen ‘blue’ (10 times), valkoinen ‘white’ (9
times), and vihreä ‘green’ (4 times). All the colour names named only once
are considered odd.
Term English gloss Women (42) Men (26) Total (68)
punainen red 18 9 27
musta black 5 5 10
sininen blue 5 5 10
valkoinen white 6 3 9
vihreä green 2 2 4
keltainen yellow 3 0 3
liila purple 2 0 2
oranssi orange 1 0 1
sini6 blue 0 1 1
violetti purple 0 1 1
Table 1. The first offered colour terms in the list task.
Table 2 shows the naming frequency, mean position, salience index, and
their corresponding rank orders for colour terms offered by five or more
subjects in the list task. The list task characterises every named colour term
by two parameters—the frequency of the word, i.e. how many subjects
named each colour term, and the mean position, i.e. in which position in the
sequence the colour term was named on average.
Term Gloss Frequency Rank Mean
Rank Salience Rank
punainen red 66 2 2.44 1 0.398 1
sininen blue 63 5 3.13 2 0.296 2
vihreä green 60 9 4.85 3 0.182 3
keltainen yellow 67 1 5.52 4 0.178 4
musta black 65 4 5.98 5 0.160 5
valkoinen white 66 2 7.26 6 0.134 6
oranssi orange 62 7 9.65 9 0.095 7
6 Here, only the first part of a colour name sininen ‘blue’ is used. It was named by a
person who started his list with colour names without nen-endings, such as sini instead
of sininen ‘blue’ and puna instead of punainen ‘red’.
THE BASIC COLOUR TERMS OF FINNISH
ruskea brown 62 7 10.27 12 0.089 8
harmaa grey 63 5 11.64 14 0.080 9
51 10 9.69 10 0.077 10
violetti purple 42 11 9.12 8 0.068 11
40 12 11.83 16 0.050 12
liila purple 29 13 9.03 7 0.047 13
turkoosi turquoise 26 14 12.00 17 0.032 14
23 15 13.00 21 0.026 15
pinkki pink 16 20 12.50 20 0.019 16
beessi beige 17 19 13.94 23 0.018 17
19 16 16.00 30 0.017 18
beige beige 16 20 13.63 22 0.017 19
lila purple 9 31 9.67 11 0.014 20
hopea silver 18 17 19.78 43 0.013 21
kulta gold 18 17 20.33 45 0.013 22
13 24 14.92 28 0.013 23
10 28 12.10 18 0.012 24
14 22 17.50 34 0.012 25
14 22 17.71 35 0.011 26
okra ochre 10 28 14.20 26 0.010 27
roosa pink 10 28 14.30 27 0.010 28
11 25 16.73 33 0.010 29
11 25 18.45 38 0.009 30
purppura purple 7 34 12.14 19 0.008 31
11 25 19.55 42 0.008 32
8 32 18.13 36 0.006 33
aniliini aniline red 5 38 11.60 13 0.006 34
5 38 11.80 15 0.006 35
5 38 14.00 24 0.005 36
5 38 14.00 26 0.005 37
sini-harmaa blue grey 8 32 23.38 47 0.005 38
6 36 18.67 39 0.005 39
puna-ruskea reddish brown 5 38 15.80 29 0.005 40
5 38 16.20 31 0.005 41
5 38 16.40 32 0.004 42
6 36 19.83 44 0.004 43
kupari copper 7 34 23.71 48 0.004 44
5 38 18.40 37 0.004 45
sini-vihreä blue green 5 38 19.00 40 0.004 46
5 38 19.40 41 0.004 47
5 38 21.20 46 0.003 48
5 38 25.00 49 0.003 49
pronssi bronze 5 38 28.40 50 0.003 50
Table 2. Frequency, mean position, salience index, and the corresponding rank orders
for colour terms mentioned by five or more subjects in the list task ranged by the rank
of the cognitive salience index.
It can be seen that the two parameters—naming frequency and mean
position—provide different colour words as candidates for basic colour
term status. Urmas Sutrop (2001, 2002) has offered a cognitive salience
index to join these parameters. The cognitive salience index is described in
detail in Sutrop (2001). According to Sutrop, this index is preferable to
other list task (free-list) indices (such as Smith 2003, Smith and Borgatti
1997), because it is free from the effects that depend on the length of
individual lists (Sutrop 2001: 272). In addition, Sutrop’s cognitive salience
index also works with a small samples or small numbers of subjects.
The formula is calculated as follows: S = F / (N x mP). The dividend
considers the frequency (F) with which a term is named in the list task. The
divisor N mP consideres the weight of the mean position (mP) in which the
term is named, and N is the number of subjects. If all subjects have named
a term (F = N) and the mean position of that term is 1, then the salience (S)
is also 1 for that term. The cognitive salience index is normed to vary
between 1 and 0. The basic terms in every domain are the most salient. The
salient index of the most ideally salient term has the figure 1. Terms that
tend to be named last and with a low frequency have a value declining
towards 0. The term that is not mentioned at all has the salience 0. The
cognitive salience index gives comparable results between different
investigations, as it does not depend on the length of the individual lists
(Sutrop 2001: 267).
THE BASIC COLOUR TERMS OF FINNISH
Frequency, mean position, and the integral cognitive salience index
are all good criteria for discriminating basic terms from non-basic ones.
Sometimes the discrimination must also be made between more and less
basic terms. According to Sutrop, in such cases, certain linguistic criteria
can well be applied.
Sutrop states that his index is not only good for distinguishing basic
colour terms from non-basic ones. Far more, with the cognitive salience
index all list task interviews are analysable. Under the term “list task,”
Sutrop means written or oral interviews in anthropology, linguistics,
psychology, or other social sciences. The format of the list task is, “Please
list all X-s that you know” (Sutrop 2001: 263). According to this format,
the researcher or interviewer can ask his/her subjects to name as many
animals as they know, or as many fruits as they know, or all the colours
they know, etc. The question could also be: “Please name everything that
you can sense with your nose”. Sutrop’s cognitive salience index has also
been used to study emotion words in Estonian and Finnish (Vainik 2002,
2006, Tuovila 2005).
In this study, Sutrop’s index has been used because it combines the
tendency of a basic term to occur at the beginning of the elicited lists (high
mean position) with its occurrence in the idiolects of all the subjects (high
term frequency). These two parameters correspond to the criteria of
psychological salience in the definition of the basic colour term by Berlin
and Kay (1969: 6) (presented in Section 1). In addition, the cognitive
salience index helps us separate possible basic colour terms from the nonbasic
The most frequently named colour term is keltainen ‘yellow,’ named
by 67 subjects. This term is followed by punainen ‘red’ and valkoinen
‘white’ (both named 66 times). Punainen ‘red,’ however, occurs at the
beginning of the elicited lists (mean position rank 1), while the mean
position rank of valkoinen ‘white’ is only 6. Only 12 terms were named by
at least half of the subjects (Fr ≥ 34): punainen ‘red,’ sininen ‘blue,’ vihreä
‘green,’ keltainen ‘yellow,’ musta ‘black,’ valkoinen ‘white,’ oranssi
‘orange,’ ruskea ‘brown,’ harmaa ‘grey,’ vaalean-punainen ‘pink,’ violetti
‘purple’ and vaalean-sininen ‘light blue’.
According to the mean position, the candidates for basic colour term
status are punainen ‘red,’ sininen ‘blue,’ vihreä ‘green,’ keltainen ‘yellow,’
musta ‘black,’ valkoinen ‘white,’ liila ‘purple’ (rank 7), violetti ‘purple,’
oranssi ‘orange,’ vaaleanpunainen ‘pink’ and lila ‘purple’ (rank 11),
followed by ruskea ‘brown’ (rank 12) and aniliini ‘aniline red7’ (rank 13).
The colour term harmaa ‘grey’ remains in rank 14. This denotes the fact
that if the 3 non-basic colour terms liila ‘purple,’ lila ‘purple’ and aniliini
‘aniline red’ are named at all, this is done in the middle of the lists.
The most salient terms according to Sutrop’s cognitive salience index
in Finnish are punainen ‘red,’ sininen ‘blue,’ vihreä ‘green,’ keltainen
‘yellow,’ musta ‘black,’ valkoinen ‘white,’ oranssi ‘orange,’ ruskea
‘brown,’ harmaa ‘grey,’ vaaleanpunainen ‘pink’ and violetti ‘purple’.
Cognitive salience is repeated in figure 2.
Figure 2. The most salient Finnish colour terms according to the cognitive salience
4.2 The colour naming task
In the colour naming task, subjects named the 65 colour squares in 4370
ways, among these there were 855 different terms. Some subjects said that
they did not know the name for some given tile on 50 occasions: eleven
subjects did not know how to name the colour tile with the Color Aid code
ORO T3 (mostly vaalea-oranssi ‘light orange’), eight subjects did not
name the colour tile YOY S2 (mostly beessi ‘beige’), etc. On average, 24.3
7 Also known as magenta in English.
THE BASIC COLOUR TERMS OF FINNISH
different names were given for each tile. In general, men had more
difficulty with naming the colour tiles than women, which combines very
well with earlier studies about English colour terms (such as Rich 1977,
Swaringen et al. 1978, Simpson and Tarrant 1991). Men’s colour
vocabulary was usually smaller than women’s. Although they did not use
the colour terms inaccurately, they tended to name colour tiles with basic
colour terms, i.e. they did not specify whether the colour was light or dark,
rich or pale, etc. When women said that one tile was vaaleansininen ‘light
blue,’ men just mentioned it was sininen ‘blue,’ because the lightness or
darkness of the tile seemed irrelevant to them. The male subjects also rarely
used colour terms connected with fashion trends, like persikka ‘peach,’
luumu ‘plum,’ lime ‘lime,’ pinkki ‘pink,’ roosa ‘pink,’ etc. These colour
names are probably familiar to women from mail order fashion catalogues,
where many new colour terms are invented.
Table 3 shows the distribution of the most frequent terms given to
each tile with the number of subjects who used each term in the Ostwald
THE BASIC COLOUR TERMS OF FINNISH
Table 3. Distribution of the most frequent terms and their corresponding frequencies in the tile naming task. Fr–frequency
THE BASIC COLOUR TERMS OF FINNISH
Table 4 shows the most frequent terms used in the tile naming task, their
total frequency, the number of tiles for which they were dominant, the
number of tiles for which they were named at least once, and the
frequency/tile ratio. The number of tiles for which a term was used at least
once shows specificity and the extension of the colour terms in the colour
space. The final column frequency/tile ratio shows the degree of consensus
among the subjects. It can be seen that the most frequent terms have greater
consensus than the rest.
According to the frequency measure (Fr > 130) 8 , there are 8
candidates for basic term status: vihreä ‘green,’ sininen ‘blue,’ ruskea
‘brown,’ vaaleanpunainen ‘pink,’ punainen ‘red,’ violetti ‘purple,’ oranssi
‘orange,’ and harmaa ‘grey’. Three other candidates for basic status musta
‘black,’ valkoinen ‘white’ and keltainen ‘yellow’ fail to achieve high levels
of frequency. This may be explained by the fact that there were only two
colour tiles in the whole colour naming task that subjects could possibly
name as yellow (Y and YOY). Moreover, only two tiles could be named
black (besides the colour tile BLACK also GRAY 8), and only one was
white, but it was not a pure white for Finnish subjects. That is why this tile
was not only named by the colour word valkoinen ‘white,’ but was also
described by other colour terms, such as maalarinvalkoinen ‘house
painter’s white’ (5 times), luonnonvalkoinen ‘nature white’ (3 times),
murrettu valkoinen ‘broken white,’ likantunut valkoinen ‘dirty white,’
luunvalkoinen ‘bone white,’ vaalea harmaa ‘light grey’ etc. The rationale
for the 65 colour sample selection can be found in Davies et al. (1992).
8 Used for 2 colour tiles on average (2 x 65 = 130).
No. of tiles
vihreä green 193 82 14 13.79
sininen blue 185 74 11 16.82
ruskea brown 178 71 8 22.25
vaalean-punainen pink 158 39 10 15.80
punainen red 143 86 8 17.88
violetti purple 141 – 12 11.75
oranssi orange 133 90 6 22.17
harmaa grey 130 88 7 18.57
liila purple 99 – 12 8.25
keltainen yellow 97 88 6 16.17
musta black 97 95 4 24.25
vaalean-vihreä light green 92 – 9 10.22
turkoosi turquoise 84 – 9 9.33
vaalean-sininen light blue 79 43 8 9.88
valkoinen white 77 49 4 19.25
tumman-vihreä dark green 58 – 6 9.67
vaalea-vihreä light green 50 – 6 8.33
pinkki pink 48 – 12 4.00
tumman-ruskea dark brown 46 – 4 11.50
vaalean-harmaa light grey 42 – 6 7.00
tumman-sininen dark blue 40 – 6 6.67
vaalea-liila light purple 39 – 7 5.57
sini-vihreä blue green 38 – 7 5.43
lila purple 33 – 13 2.54
vaalea-harmaa light grey 32 – 5 6.40
vaalean-keltainen light yellow 30 – 3 10.00
roosa pink 29 – 7 4.14
vaalea-violetti light purple 28 – 9 3.11
taivaan-sininen sky blue 26 – 5 5.20
vaalean-liila light purple 25 – 7 3.57
kirkkaan-sininen bright blue 23 – 5 4.60
tumma-liila dark purple 23 – 9 2.56
Table 4. The most frequent terms in the tile naming task, their total frequency, their
dominance frequency, the number of tiles for which they were named at least once, and
the frequency/tile ratio.
The final column, indicating the frequency/tile ratio, shows the consensus
of use. The higher the ratio, the greater the consensus among the subjects.
It can be seen that the most frequent terms have greater consensus than the
THE BASIC COLOUR TERMS OF FINNISH
others. According to the frequency/tile ratio measure (< 15), there are 9
candidates for basic status: musta ‘black,’ ruskea ‘brown,’ oranssi
‘orange,’ valkoinen ‘white,’ harmaa ‘grey,’ punainen ‘red,’ sininen ‘blue,’
keltainen ‘yellow,’ and vaaleanpunainen ‘pink’. After these 9, there is a
gap and only then two candidates for basic status vihreä ‘green’ and violetti
‘purple’ will follow. The reason for that may be the large number of green
and purple colour tiles in this task. As can be seen in Table 4, the colour
name vihreä ‘green’ was used to name 14 tiles and the colour name violetti
‘purple’ was elicited by 12 colour tiles.
In addition to the naming frequency and frequency/tile ratio the
dominance frequency is also presented. A term is considered dominant if at
least half of the subjects use the same name for a given tile, which means
that the so-called dominance index has to be DI ≥1/2. This is the reason
why some of the terms do not have a dominance frequency at all. The
dominance index is counted to calculate the specificity index (SI). The
specificity index is the dominant frequency/total frequency ratio at the
same level. If the specificity index was 1, all the subjects used the same
term only as the dominant term and there was absolute consensus among
the subjects (see Davies and Corbett 1994: 79). The specificity index
together with dominant colour terms on different consensus levels for
Finnish colour terms is shown in Table 5.
It is possible to consider dominance and specificity indices on
different levels of consensus. In this article the following limits for
dominant indices are used (numbers are rounded where necessary):
DI 1/10 1/4 1/3 1/2 2/3 3/4 1
Frequency pro tile ≥ 7 17 23 34 45 51 68
Table 5 shows the dominant colour terms on different consensus levels
together with their specificity indices. There is no dominant colour term for
any tile at the absolute consensus level in Finnish. The highest is the
consensus for the colour term musta ‘black’ (SI = 0.98). Surprisingly,
exactly the same holds true for Hungarian colour names where the
consensus 0.98 is the highest for the colour term fekete ‘black’ (Uusküla
and Sutrop 2007). It is interesting that, when one compares those results
with the studies of other languages in the geographic area, which have been
carried out using the same method, then for Russian the specificity index is
the highest for the colour term belyj ‘white’ (SI = 1, which indicates the
absolute consensus among the subjects) (Davies and Corbett 1994: 79). In
Estonian the specificity index is also the highest for the colour term for
white, valge (SI = 0.99) (Sutrop 1995: 806, 2000a: 160, 2002: 84–85). As
can be seen from the Table 5, in Finnish the lowest consensus holds for the
colour term vaaleanpunainen ‘pink’ (SI = 0.25). At the same time the
colour name violetti ‘purple’ does not have a specificity index at all,
because there was no dominant colour tile in the task. The tile the most
frequently named as violetti was tile VRV—named 26 times, i. e. 38% of
all responses. Looking at the SI indices one can see that vaaleansininen
‘light blue’ also has a value in this column (SI = 0.54). This can be
explained by the fact that the colour tile BGB T3 was named as
vaaleansininen by 43 subjects (63% of responses). In-depth discussion of
the status of this colour name among others will follow below.
Term Gloss SI DI
musta black 0.98 2 2 2 2 1 1
keltainen yellow 0.91 2 1 1 1 0 0
harmaa grey 0.68 4 3 2 2 0 0
oranssi orange 0.68 4 3 3 2 0 0
valkoinen white 0.64 9 4 3 2 0 0
punainen red 0.60 5 3 2 1 1 0
vaalean-sininen light blue 0.54 5 3 3 1 1 0
vihreä green 0.42 7 5 1 1 0 0
sininen blue 0.40 7 4 4 1 0 0
ruskea brown 0.40 9 5 4 1 0 0
vaalean-punainen pink 0.25 5 2 2 1 0 0
violetti purple 0.00 5 4 2 0 0 0
liila purple 0.00 4 0 0 0 0 0
vaalean-violetti light purple 0.00 4 2 1 0 0 0
turkoosi turquoise 0.00 4 2 1 0 0 0
tumman-vihreä dark green 0.00 4 2 1 0 0 0
vaalea-vihreä light green 0.00 3 2 1 0 0 0
pinkki pink 0.00 3 2 2 0 0 0
tumman-ruskea dark brown 0.00 3 2 0 0 0 0
vaalean-harmaa light grey 0.00 3 1 1 0 0 0
tumman-sininen dark blue 0.00 3 0 0 0 0 0
vaalea-liila light purple 0.00 2 0 0 0 0 0
Table 5. Dominant colour terms in the tile naming task. SI–specificity index, DI–
Looking at the lowest consensus level (threshold DI 1/10) it can be seen
that 64 tiles have a dominant colour term. The only tile which does not
have a dominant term is YOY S2, while both beessi ‘beige’ and beige
THE BASIC COLOUR TERMS OF FINNISH
‘beige’ are named 4 times (not 7). On the 25% consensus level (DI 1/4) one
can find 44 tiles with 17 dominant names. On the 50 % consensus level (DI
1/2) there are only 19 tiles with 11 dominant colour names, which are the
most probable candidates for basic term status: musta ‘black,’ keltainen
‘yellow,’ harmaa ‘grey,’ oranssi ‘orange,’ valkoinen ‘white,’ punainen
‘red,’ vaaleansininen ‘light blue,’ vihreä ‘green,’ sininen ‘blue,’ ruskea
‘brown,’ and vaaleanpunainen ‘pink’. As has been argued before, the 50 %
threshold is not exceeded by the colour name violetti ‘purple’. Violetti is
the most probable candidate for purple category, while Finnish also has
other colour names to fill in this gap, like liila and lila ‘purple,’ retliini and
retuliini ‘purple,’ and sinipunainen ‘bluish red’ (means also purple) (see
Section 4.3 for this issue). On the next, 67 % consensus level (DI 2/3) there
is only 1 tile—BLACK with the colour name musta ‘black’.
If the specificity index at the 50 % consensus level is taken, there
would be 10 candidates for basic status; the threshold SI 1/2 > 0.30.
4.3 Combined results and discussion
In the list and the colour naming task the subjects named 1014 different
colour terms. Of the 332 terms listed in the first list task, 167 were never
used in the colour naming task (including hopea ‘silver,’ kulta ‘gold’ and
pronssi ‘bronze’). On the other hand, in the colour naming task the subjects
used 669 new different colour names not listed in the first task.
Morphologically, there were 2711 monolexemic terms (101 different) and
3165 compound terms of which 913 were different (including
vaaleanpunainen ‘pink’), named in the two tasks.
As a preliminary result, 12 candidates fulfilled at least one criterion
according to the different tasks and measures. These account for 85 % of
the total responses (2304) in the list and the colour naming task (including
violetti ‘purple,’ vaaleanpunainen ‘pink’ and vaaleansininen ‘light blue’).
There are 11 standard terms: musta ‘black,’ valkoinen ‘white,’ punainen
‘red,’ keltainen ‘yellow,’ vihreä ‘green,’ sininen ‘blue,’ ruskea ‘brown,’
oranssi ‘orange,’ harmaa ‘grey,’ vaaleanpunainen ‘pink,’ violetti ‘purple,’
and one complex term vaaleansininen ‘light blue’. All the other terms
suspected to have a (nearly) basic term status, like sinipunainen, kretliini
and valkea (Koski (1983) treats them as respective synonyms for purple
and white in his monograph), pinkki ‘pink,’ roosa ‘pink,’ turkoosi
‘turquoise’ and beige ‘beige’ (which are discussed often by Koski (1983)),
liila and lila ‘lilac, purple’ (a relatively high frequency in the list task), and
tummanvihreä ‘dark green,’ vaaleanvihreä ‘light green,’ tummansininen
‘dark blue’ (a relatively high frequency in the colour naming task), do not
meet any of the criteria established above (see Table 7).
All previous results for establishing basic colour terms in Finnish have
been combined and the established terms, ordered according to their level
of basicness, are presented in Table 7. The current analysis of the combined
results was carried out with the same methods as in the case of basic colour
terms of Estonian (Sutrop 1995, 2000a, 2002). In the list task, the naming
frequency (Fr ≥ 40) and mean position (mp < 8), and in the colour naming
task, the naming frequency (Fr ≥ 130), dominance index (DI 1/2 ≥ 1), and
specificity index (SI > 0.20) are considered, measured against given
numerical values as thresholds which have to be superseded. The salience
index is not included here because it depends on the frequency and mean
position of the term in the list task.
The last column of Table 7 shows the sums of these criteria, the value
of which for one colour term could be from 0 to 5. The higher this number,
the more certain the status of the colour term as basic. In other words, it
shows the terms’ level of basicness.
List task Colour naming task
Term Gloss Fr > 40 Mp < 8 Fr ≥ 130 DI 1/2≥1 SI > 0.2
punainen red + + + + + 5
vihreä green + + + + + 5
sininen blue + + + + + 5
musta black + + – + + 4
valkoinen white + + – + + 4
keltainen yellow + + – + + 4
ruskea brown + – + + + 4
oranssi orange + – + + + 4
harmaa grey + – + + + 4
vaaleanpunainen pink + – + + + 4
vaaleansininen light blue + – – + + 3
violetti purple + – + – – 2
liila purple – – – – – 0
turkoosi turquoise – – – – – 0
beige beige – – – – – 0
Table 7. Summary of the results where colour terms are ranged according to the level of
basicness. Fr–naming frequency, mp–mean position, DI–dominance index, SI–
The most salient colour terms in Finnish are punainen ‘red,’ sininen ‘blue’
and vihreä ‘green’ (those terms superseded all 5 thresholds). 4 thresholds
are superseded by the colour terms musta ‘black,’ valkoinen ‘white,’
THE BASIC COLOUR TERMS OF FINNISH
keltainen ‘yellow,’ ruskea ‘brown,’ harmaa ‘grey,’ vaaleanpunainen
‘pink,’ and oranssi ‘orange’.
At this point it is relevant to note that oranssi has been adapted to
modern Finnish as a basic colour term, in contrast to what Koski has
proposed (1983: 265). In Finnish, the colour name oranssi does not mean
orange (fruit) as it does in English9. This colour term has been displaced by
another term, appelsiini ‘orange’. There is only one possible reason why
colour term oranssi could not be a basic term according to the criteria of
Berlin and Kay: namely, it is a late loanword borrowed into Finnish in the
early 20th century through Swedish. However, that criterion could be
circumvented, while it is considered only as subsidiary (Berlin and Kay
Before making a final decision about the colour name
vaaleanpunainen ‘pink,’ literally ‘light red’10 Berlin and Kay’s criteria for
basicness should be reconsidered (1969: 5–7). The first criterion by Berlin
and Kay prescribes that a basic term must be monolexemic; that is, its
meaning should not be predictable from the meaning of its parts. Thus,
Finnish vaaleanpunainen (vaalea ‘light’ and punainen ‘red’) being a
compound word, cannot be a basic colour term according to this linguistic
criterion. Nevertheless, its meaning is, indeed, independent from the
meanings of its parts, because there were no subjects who named light red
with this term. Instead, other terms to indicate light red were used, like
heleä punainen ‘light red’. Colour term vaaleanpunainen ‘pink’ also
occurred in such compounds as tumma-vaaleanpunainen ‘dark pink’ and
vaalea-vaaleanpunainen ‘light pink’ which clearly shows that it is thought
as a new concept—pink colour. According to the dictionaries such as
Alanne (1982) or Hurme and Pesonen (1986), vaaleanpunainen in
translated to English only as pink (and vice versa), never as light red,
although the meaning of vaalean- is given as light or pale.
Berlin and Kay’s original characteristics have been critizised,
rearranged and given different weights by T. D. Crawford (1982), A. E.
Moss (1989), Jerome Smith et al. (1995) and others. They have argued that
it may be useful to distinguish between psychological—or perceptual—
9 In English the word orange means both fruit and colour, although it is not clear for
every native speaker which of the two meanings is the first. This is the only basic colour
term in English considered transparent, while the other basic colour terms are opaque.
10 Etymologically vaalea(n-) is derived from valkea ‘white, light,’ thus its meaning
might be ‘white red’ which corresponds to pink colour.
basicness and linguistic basicness. For instance, Crawford, in his revision
of the original criteria has rejected all linguistic criteria.
Thus, Crawford (1982: 342) argues: “A basic colour term occurs in
the idiolects of all informants. It has stability of reference across informants
and across occasions of use. Its signification is not included in that of any
other colour term. Its application is not restricted to a narrow class of
By those criteria vaaleanpunainen is, indeed, a basic colour term: it
belongs to the idiolects of all informants, it has stability of reference across
informants and across occasions of use (in the colour naming task more
than 50 % of the subjects named colour tile R T4 with that name); its use is
not restricted to a narrow class of objects and its signification is not
included in that of any other colour term, and it is cognitively salient as is
shown by the salience index (see Table 2). Hence, it could be argued that in
Finnish vaaleanpunainen ‘pink’ is a basic term psychologically, like
narancssárga ‘orange’ (literally ‘orange-yellow,’ from narancs ‘orange’
and sárga ‘yellow’) in Hungarian. 11 All other colour terms (inter alia
monolexemic ones) that could be used to refer to the category of pink in
Finnish, like roosa ‘pink’ and pinkki ‘pink,’ do not meet any criteria set up
for the basic term.
Three thresholds in Table 7 (naming frequency in the list task,
dominance and specificity indices in the colour naming task) are
superseded by a colour term vaaleansininen ‘light blue’. It is possible that
this term strives for basic status, because it forms a symmetrical pair with
colour name vaaleanpunainen ‘pink’. The same phenomenon was found in
Hungarian, where the status of the basic colour term sárga ‘yellow’ is
extremely weak, while another colour name citromsárga ‘lemon-yellow’
tends to replace it in order to make a symmetrical pair with the basic colour
term narancssárga ‘orange, literally ‘orange-yellow’ (Bogatkin-Uusküla
and Sutrop 2005b: 97). However, in Finnish vaaleansininen ‘light blue’
does not appear to substitute for the basic colour term sininen ‘blue,’ but it
lays claim to the 11th place among basic colour terms, which position
should, according to Berlin and Kay’s theory, rather be occupied by a term
for purple, i.e. one of the following: violetti, liila (or lila), kretliini
11 For more about Hungarian colour names see Uusküla and Sutrop 2007.
12 The reason why there are so many terms for purple in Finnish might be explained by
the fact that none of these terms have a strong basic status. There are also meaning
THE BASIC COLOUR TERMS OF FINNISH
It is also possible that the meaning of colour name sininen ‘blue’ will
change in time, so that it refers to a darker colour than presently, and thus a
need will arise for an additional basic colour term—vaaleansininen ‘light
blue,’ which would then refer to a lighter region of the blue spectrum .13
There are many languages in which two (basic) colour terms are used
for blue, amongst them Russian, which has sinij ‘(dark) blue’ and goluboj
‘light (cold) blue’ for blue, and therefore 12 basic colour terms in total14
(Davies and Corbett 1994); Turkish has two terms for blue: lacivert ‘dark
blue’ and mavi ‘blue’ (Özgen and Davies 1998); Italian has three terms for
blue: blu ‘blue,’ azzurro ‘azure’ and celeste ‘sky blue’ (Kristol 1979,
1980), of which, according to Philip, two are basic colour terms: blu ‘dark
blue’ and azzurro ‘light blue’ (Philip 2006: 61–62).
As far as vaaleansininen ‘light blue’ in Finnish is concerned, basic
status is far from certain—the cognitive salience index places it 12th after a
big gap. Unlike vaaleanpunainen ‘pink,’ the meaning of which is
cognitively not predicted from its parts (linguistically it means light + red,
but it is not used in this sense at all), vaaleansininen still clearly denotes
light blue, because it is used only of light blue tiles in the colour naming
task. As was already argued above, linguistic complexity of a term does not
have to possess heavy weight if it can be considered as a basic term
psychologically (for more on basic terms, see Sutrop 2000b).
differences. According to “Uusi suomi-englanti suursanakirja” 1984 [New Finnish-
English Dictionary] (Hurme et al.) liila translates to English as lilac (also in the colour
naming task subjects named light purple tiles mostly with that name or modifying it
with vaalea(n)- ‘light’). According to the same dictionary colour names violetti,
(k)retliini and sinipunainen should all be translated into English as violet, not purple.
The only name that could be translated as purple among these is sinipunainen. In this
article, all terms are translated into English as purple in order to emphasise their equal
chance to establish themselves as the basic term.
13 Hereby it should be stated that two closely related languages Finnish and Estonian
(both studied with the same method) have a different prototypical blue. While for
Finnish it lays in colour tile BVB (56 % of total answers), for Estonian it is colour tile B
(64 % of total answers). At this point I find it relevant to explain the difference between
Uusküla (2006) and the present study where, according to the answers of 29 subjects I
have suggested that prototypical blue in Finnish is a colour tile B. It clearly shows how
different number of subjects can give different results.
14 Russian also has many terms for purple, e. g. fioletovyj ‘purple,’ sirenevyj ‘mauve,’
lilovyj ‘lilac’. Among these fioletovyj ‘purple’ is considered as basic (see Davies and
Corbett 1994: 80–86).
Vaaleansininen ‘light blue’ should therefore not be regarded as a basic
colour term in Finnish and should rather be regarded as an anomaly.
Only two thresholds (naming frequency in the list task and frequency
in colour naming task) in Table 7 are superseded by the colour term
violetti, which has been glossed as purple throughout this article, although
its better meaning in English would be violet as given in Hurme et al.
(1984). The basic status of this colour name seems to be rather
questionable. When considering the cognitive salience index (see Table 2),
violetti might be a basic colour term, as we find it on the 11th place. While
there is also another colour term for purple in Finnish—liila whose
preferable meaning in English would be ‘lilac’15—these colour terms were
put to a simple test.
In the present article, two phonetic variants of liila—liila and lila
‘lilac’—are counted as different words in order to show the exact responses
of the subjects. If these two variants were counted as one word—under the
colour name liila—somewhat different results would be obtained.
According to that, naming frequency of liila in the list task would be 38
(liila has been offered 29 times and lila 9 times) and the salience index
would score much higher—0.062, instead of the present 0.050 (see Table 2
thereinbefore). One can see that this calculation would raise liila to the 12th
place instead of the present 13th, but not higher, so that violetti would still
be the most salient term for the category of purple.
In the colour naming task there was no consensus about which colour
tile could be referred to with the colour name violetti. The tile most
frequently named with that term was the tile VRV (violet-red-violet) (26
times out of 68) and the next one was the tile VBV (violet-blue-violet) (25
times out of 68). Dominance (which is calculated so that at least 50 % of
the subjects must name one colour tile with a certain colour term) is one of
the most important criteria to establish for basic colour term. While the
colour name violetti is not a dominant colour name for any of the colour
tiles, it does not have a specificity index either. Hence, violetti should not
be counted as a basic colour term, indicating that there is no basic colour
word for the category of purple in Finnish at all. These considerations arise
from the data analysed in this study: clearly, more fieldwork might need be
15 As referred to before, languages with more than one (basic) colour term for blue
usually also contain more than one colour name for purple (Corbett and Morgan 1988,
Moss 1989, Morgan and Corbett 1989, Davies et al. 1995). Problems appear when
attempts are made to gloss those terms amongst languages (see Davies and Corbett
1994, Davies et al. 1995).
THE BASIC COLOUR TERMS OF FINNISH
carried out in the future, as a greater number of interviews may provide
somewhat different results.
The findings from the list and the colour naming task converge to
suggest that Finnish has 10 basic colour terms and it does not have a basic
term for purple, thus differing from Berlin and Kay’s list of 11 universal
terms (see Table 8). According to Mauno Koski, whose only sources where
dictionaries and dialect collections, the Finnish colour term inventory
contains 8 basic terms—he excludes violetti ‘purple,’ oranssi ‘orange’ and
vaaleanpunainen ‘pink’ (Koski 1983: 265). According to the present study,
however, the last two colour names—oranssi ‘orange’ and
vaaleanpunainen ‘pink’—are in fact basic terms. The basic colour term for
white in Finnish is valkoinen ‘white,’ not valkea ‘light,’ which, to some
extent, have been treated synonymically by Koski and Todorova (Koski
1983, Todorova 1991). Actually, the meaning of this word could also be
glossed to English as ‘light, lightness; fire’ (see Tuomi 1966 about its
etymology). In spite of that, there are some expressions where valkea
functions as a colour name, like valkeissa vaatteissa ‘dressed in white’ etc.
(although valkoisissa vaatteissa ‘dressed in white’ is more usual). None of
the subjects in the present study used this word as a simple word either in
the list or in the colour naming task. It only occurred 4 times in compounds
lumi-valkea ‘snow white,’ maidon-valkea ‘milk white,’ puhtaan-valkea
‘pure white,’ and hopea-valkea ‘silver white’.
English gloss Mauno Koski (1983) Present study
white valkoinen valkoinen
black musta musta
red punainen punainen
green vihreä vihreä
yellow keltainen keltainen
blue sininen sininen
brown ruskea ruskea
orange (oranssi) oranssi
grey harmaa harmaa
pink – vaaleanpunainen
purple (violetti) –
Table 8. Basic colour terms in Finnish ranged by Berlin and Kay’s original basic colour
5. Summary and conclusion
In the list and colour naming tasks 68 subjects named 5876 colour terms of
which 1014 were different. In the list task there were 1506 terms offered in
total, out of which 332 were different. In the colour naming task the
subjects gave 4370 colour names to 65 colour squares. Among these there
were 855 different names.
There are 10 basic colour terms in Finnish. Ranged top-down by
cognitive salience index they are the following: punainen ‘red,’ sininen
‘blue,’ vihreä ‘green,’ keltainen ‘yellow,’ musta ‘black,’ valkoinen ‘white,’
oranssi ‘orange,’ ruskea ‘brown,’ harmaa ‘grey’ and vaaleanpunainen
‘pink’. The colour name oranssi ‘orange’ is adjusted to modern Finnish.
The basic colour term for category of pink in Finnish is vaaleanpunainen
‘pink,’ while the colour term violetti ‘purple’ is not a basic colour term.
Thus the Finnish language corresponds to the last, the seventh stage of
Berlin and Kay’s scheme, but being still on a way to getting basic term for
purple. The most probable candidate to fill in that gap in the future is
colour term violetti ‘purple’. Differences between the present study and the
basic colour term inventory suggested by Mauno Koski are the following:
1) the colour term oranssi ‘orange’ is a basic colour term, which either
means that Koski was mistaken, or that the Finnish language has changed
over the past 25 years so that it now possesses a basic term for orange; 2)
according to the present study Finnish also has a basic term for pink.
Alanne, Vieno Severi (1982) Suomalais-englantilainen suursanakirja. Porvoo/Helsinki/
Berlin, Brent & Kay, Paul (1969) Basic color terms: their universality and evolution.
Berkeley: University of California Press.
Bogatkin-Uusküla, Mari & Sutrop, Urmas (2005a) Tänapäeva ungari keele
põhivärvinimed. Keel ja Kirjandus 7: 558–570.
—— (2005b) Kas ungari keeles on kaks punase värvi põhinime piros ja vörös?
Emakeele Seltsi Aastaraamat 50: 93–110.
Corbett, Greville & Morgan, Gerry (1988) Colour terms in Russian: reflections of
typological constraints in a single language. Journal of Linguistics 24: 31–64.
Crawford, T. D. (1982) Defining ‘basic color term’. Anthropological Linguistics 24:
Davies, Ian, MacDremid, Catriona, Corbett, Greville, McGurk, Harry, Jerrett, David,
Jerrett, Tiny & Sowden, Paul (1992) Color terms in Setswana: a linguistic and
perceptual approach. Linguistics 30: 1065–1103.
THE BASIC COLOUR TERMS OF FINNISH
Davies, Ian, Davies, Christine & Corbett, Greville (1994) The basic colour terms of
Ndebele. African Languages and Cultures 7, 1: 36–48.
Davies, Ian & Corbett, Greville (1994) The basic color terms of Russian. Linguistics 32:
—— (1995) A practical field method for identifying basic colour terms. Languages of
the World 9, 1: 25–36.
Davies, Ian, Corbett, Greville & Margalef, José Bayo (1995) Colour terms in Catalan:
an investigation of eighty informants, concentrating on the purple and blue
regions. Transactions of the Philological Society, 93, 1: 17–49.
Fletcher, Robert (1980) The City University colour vision test. 2nd ed. London: Keeler.
Hurme, Raija, Malin, Riitta-Leena & Syväoja, Olli (1984) Uusi suomi-englanti
suursanakirja. Porvoo/Helsinki/Juva: WSOY.
Hurme, Raija & Pesonen, Maritta (1986) Englantilais-suomalainen suursanakirja.
Kay, Paul (1975) Synchronic variability and diachronic change in basic color terms.
Language and Society 4: 257–270.
Kay, Paul & McDaniel, Chad, K. (1978) The linguistic significance of the meanings of
basic color terms. Language 54: 610–646.
Kay, Paul, Berlin, Brent & Merrifield, William (1991) Biocultural implications of
systems of color naming. Journal of Linguistic Anthropology 1: 12–25.
Kay, Paul, Berlin, Brent, Maffi, Luisa & Merrifield, William (1997) Color naming
across languages. In C. L. Hardin and Luisa Maffi (eds.), Color Categories in
Thought and Language. Cambridge: Cambridge University Press.
Koski, Mauno (1983) Värien nimitykset suomessa ja lähisukukielissä. Suomalaisen
Kirjallisuuden Seuran toimituksia, 391. Helsinki: Suomalaisen Kirjallisuuden
Kristol, Andres (1979) Il colore azzurro nei dialetti italiani. Vox Romanica 38: 85–99.
—— (1980) Color systems in Southern Italy: A case of regression. Language 56, 1:
Morgan, Gerry & Corbett, Greville (1989) Russian colour term salience. Russian
Linguistics 13: 125–141.
Moss, Anthony E. (1989) Does Russian have a basic term for purple? Linguistics 27:
Oja, Vilja (2001) Linguistic studies of Estonian colour terminology. Dissertationes
Philologiae Estonicae Universitatis Tartuensis, 9. Tartu: Tartu University Press.
Ostwald, Wilhelm (1939) Die kleine Farbmeßtafel nach Wilhelm Ostwald. Bearbeitet
von Gerhard Streller und Grete Ostwald. Text von Gerhard Steller. Göttingen:
Philip, Gill (2006) Connotative meaning in English and Italian colour-word metaphors.
metaphorik.de 10: 59–93. (www.metaphorik.de)
Rich, Elaine (1977) Sex-related differences in colour vocabulary. Language and Speech
20, 4: 404–409.
Simpson, Jean & Tarrant, Arthur W. S. (1991) Sex- and age-related differences in
colour vocabulary. Language and Speech 34, 1: 57–62.
Sipőcz, Katalin (1994) A vogul nyelv színnevei. Studia Uralo-Altaica Supplementum, 3.
Smith, J. Jerome (1993) Using ANTHROPAC 3.5 and a spreadsheet to compute a freelist
salience index. Cultural Anthropology Methods 9, 3: 1–3.
Smith, J. Jerome, Furebee, Louanna, Maynard, Kelly, Quick, Sarah & Ross, Larry
(1995) Salience counts: A domain analysis of English colour terms. Journal of
Linguistic Anthropology 5: 203–216.
Smith, J. Jerome & Borgatti, Stephen P. (1997) Salience counts–and so does accuracy:
Correcting and updating a measure for free-list-item salience. Journal of
Linguistic Anthropology 7, 2: 208–209.
Sutrop, Urmas (1995) Eesti keele põhivärvinimed. Keel ja Kirjandus 12: 797–808.
—— (2000a) The basic colour terms of Estonian. Trames 4, 1: 143–168.
—— (2000b) Basic terms and basic vocabulary. In Mati Erelt (ed.), Estonian:
Typological Studies, pp. 118–145. Tartu: Tartu Ülikooli eesti keele õppetooli
—— (2001) List task and a cognitive salience index. Field Methods 13, 3: 263–276.
—— (2002) The vocabulary of sense perception in Estonian: structure and history.
Opuscula Fenno-Ugrica Gottingensia, 8. Frankfurt am Main: Peter Lang.
Swaringen, Sandra, Layman, Stephanie & Wilson, Alan (1978) Sex differences in color
naming. Perceptual and Motor Skills 47: 440–442.
Todorova, Elena (1991) Tsvetut na bulgarskata i finskata ezikova sistema. In Milena
Tsaneva, Petur Pashov and Boian Vulchev (eds.), Bulgaristichni izsledvaniia:
Treti bulgaro-skandinavski simpozium (20–26 septembri 1985 g). Sofia: Univ.
izdatelstvo ‘Sv. Kliment Okhridski.’
Tuomi, Tuomo (1966) Tuli ja valkea. Virittäjä: 99–107.
Tuovila, Seija (2005). Kun on tunteet. Suomen kielen tunnesanojen semantiikkaa.
Acta Universitatis Ouluensis, B Humaniora 65. Oulu: Oulun Yliopisto.
Turunen, Rigina (2002) Die Farbbezeichnungen im Mokscha-Mordwinischen. Finnisch-
Ugrische Forschungen 57, 1–3: 167–194.
Uusküla, Mari (2006) Distribution of colour terms in Ostwald’s colour space in
Estonian, Finnish, Hungarian, Russian and English. Trames, 10, 2: 152–168.
Uusküla, Mari and Sutrop, Urmas (2007) Preliminary study of basic colour terms in
modern Hungarian. Linguistica Uralica 43, 2: 102–123.
Vainik, Ene (2002) Emotions, emotion terms and emotion concepts in an Estonian folk
model. Trames, 6, 4: 322–341.
—— (2006) Intracultural variation of semantic and episodic emotion knowledge in
Estonian. Trames, 10, 2: 169–189.
Özgen, Emre and Davies, Ian (1998) Turkish color terms: tests of Berlin and Kay’s
theory of color universals and linguistic relativity. Linguistics 35–36: 919–956.
THE BASIC COLOUR TERMS OF FINNISH
Institute of the Estonian Language, Tallinn
Tel.: +372 56636380
colour finishPosted: 13 Agustus 2010 in colour language
SKY Journal of Linguistics 20 (2007), 367–397