The colonisation of the colour pink: variation and change in Māori’s colour lexicon

3.1 Stage I–IV categories (white, black, red, green, and yellow)

PEP ‘white’. PEP exhibited at least two words for WHITE: *tea ‘white’ (< Proto-Central Pacific *tea) and *muka ‘white, light in colour’ (< Proto-Polynesian *muka). In at least three Eastern Polynesian languages, PPn *qali ‘bare, clearly visible, transparent (of water)’ has undergone semantic extension and is used as a colour term: Hawaiian ali ‘crystal clear, white’, Māori ari ‘white, visible’, and Tuamotu ari ‘visible, be white’ (POLLEX). Since all three languages are independent primary branches of PEP (Walworth 2014), it is possible that this semantic extension had been completed in PEP.

PEP ‘black’. No fewer than eight words for BLACK are reconstructable to PEP. Among these words, *pago and *siwa are only attested in Walworth’s Central Eastern Polynesian group, and not in Rapa Nui or any other Polynesian languages. This suggests that Central Eastern Polynesian languages have productively invented new words for this colour category. Four other words are inherited from Proto-Polynesian (PPn) – *kiwa ‘dark hue, black’, *kalaa ‘hard, black, volcanic stone’, *quli ‘black, dark in colour’, and *qusi ‘dark-coloured, including dark blues and greens’. Two other words, *kele ‘dark, black’ and *magu ‘black’, can be traced back to PEP.

PEP ‘red’. PEP had at least three terms for RED. *mea ‘reddish’, the oldest term of all, was inherited from Proto-Malayo-Polynesian *ma-qiReq ‘black’ (ACD). The words *ula ‘glow, red’ and *kula ‘red’ are both inherited from PPn. Three other PPn words – *kefu, *melo ‘red/brown’, and *sega – also deserve a note: all three are reconstructed with RED-related semantics, and the meaning of their reflexes in individual language vary among the concepts we call red, yellow, and brown in English: *kefu ‘light-coloured (blond, brown, reddish) especially of hair’, *melo ‘red/brown’, and *sega ‘yellow or yellowish-red; orange’. The ambiguity suggests that in the PEP stage, the colours called red and brown in English are within a single RED colour category (cf. [1], [2]).

PEP ‘green’. At least four words are reconstructable: *namu ‘green/blue’, (Branstetter 1977: evidenced by Tahitian ninamu, Māori pounamu ‘greenstone, green’, and Tumotu (ni)namu ‘blue’), *mata ‘raw, green, unripe’ (Branstetter 1977), *qui ‘blue, green, type of coconut’, and *qusi ‘dark-coloured, including dark blues and green’.^[3] Except for *mata and *namu, all these words are reconstructable to PPn. This suggests that the English colour categories green and blue were not distinguished in either PPn or PEP, with the boundary between the GRUE category and the BLACK category in PEP possibly cutting through the middle of the green and blue categories in English.

PEP ‘yellow’. At least two unambiguous words for YELLOW, *felo ‘yellow’ and *koo-rega ‘plant sp.; yellowish’ are reconstructable. This provides evidence that speakers of PEP already identified yellow as a distinct colour category. Two other words can also be traced back to PEP: *rega-rega ‘yellow’ and *sega ‘yellow or yellowish-red; orange’.

3.2 Absence of other categories (blue, brown, grey, purple, pink, and orange)

Categories that fall under later stages of Berlin and Kay’s hierarchy appear not to have been established in PEP.

PEP ‘blue’. Comparative data from POLLEX shows that various Polynesian languages do not distinguish between green and blue, as evidenced by the PPn words *qui ‘blue, green; type of coconut’ and *qusi ‘dark-coloured, including dark blues and greens’. These words are likely to refer to GRUE in PEP.

PEP ‘brown’. It is unclear whether PEP speakers treated brown as a specific colour category. The ambiguous meaning of *maatuke ‘dark-brown’ and *melo ‘red/brown’, and *kefu ‘light-coloured (blond, brown, reddish) especially of hair’ suggests that the English category of brown is incorporated within one or two larger PEP categories, RED and/or BLACK.^[4]

PEP ‘grey’. In some Polynesian languages, the reflex of Proto-Oceanic *sina ‘grey or white-haired’ refers to the colour category WHITE (e.g., East Uvea, Niue, Nuguria). Only in one language (Moriori) does it refer to the colour grey. There is thus insufficient evidence to reconstruct this colour to PEP. It is likely that *sina refers to the specific instance of grey hair rather than being generalised to refer to the colour itself.

3.3 Interim conclusion

There is strong comparative evidence for five colour categories (BLACK, WHITE, RED, YELLOW, GRUE), meaning that PEP, the immediate ancestor of Māori, was at Stage IV of Berlin and Kay’s hierarchy.

A central question for reconstructing colour categories in a protolanguage that is no longer spoken is whether there was likely to be a basic colour term attached to the category – that is, a word that refers to the colour itself in general rather than to any particular type of coloured object. If so, there is a further challenge of determining which of the attested words is the word that would have been used for the colour category and which were subsidiary words. For example, in English, the words green, emerald, chartreuse, and jade all represent greens but only one of those words is used for the overall category, with the others representing particular shades of green and/or particular green-colour objects. For PEP, we are so far removed in time that it is not possible to make this determination with any certainty.

In summary, PEP was mostly likely a language at Stage IV of the Berlin and Kay hierarchy with BLACK, WHITE, RED, YELLOW, and GRUE colour categories. The English blue was not a colour category. English blues were in either the GRUE or BLACK category. The comparative data from POLLEX also shows that PCEP, the hypothetical daughter language of PEP excluding Rapa Nui, had a colour system identical with that of PEP in having a Stage IV system. Thus we can assume that both immediate anscestors of Māori were Stage IV languages.

4.1 Stage I colours (white and black)

Māori in the 19th century, like every language, has words for BLACK and WHITE. At least seven Stage I colour terms are retained from PEP; five others (i.e., *muka and *ari for WHITE; *kalaa, *qusi, and *kiwa for BLACK) are not reported and may have been lost. Nine native Stage I colour terms are documented in Tregear (1891), which have no clear etymology linked to PEP. It is unclear whether these terms were simply not included in Williams’s dictionary or emerged after the publication of Williams (1871). At the very least, we may conclude that all these novel terms were developed within 500 years – after the Māori left the Society Islands and settled New Zealand (13th century) and before English started to impact the Māori language (19th century). It is also possible that these additional terms documented by Tregear (1891) (6) already existed in the early 1800s but were not documented by Williams. We remain agnostic about the chronology of these terms.

4.2 Stage II colour (red)

The category of RED also accrued a series of new words. Three PEP terms had been lost; five terms reported in Tregear (1891) (waiarangi, mataura, reperepe, matahanahana, tututupo) are not attested in Williams’s first and third editions. Tregear (1891) also contains a wide range of terms that contain the basic terms listed in (5) as the root. For example: towhero ‘red’, kurawhero ‘red’, puwherowhero ‘reddish-brown’.

4.3 Stage III/IV colours (green and yellow)

Green. Surprisingly, none of the four PEP words for GRUE leaves a reflex in Williams (1844) or Tregear (1891). Instead, three lexical innovations (8) are attested:

As with the discussion of blue and grey (4.4 and 4.5), pounamu is a situation where the colour of a specific object is sometimes used to describe the colour of other things of a similar shade. Note that pounamu can also mean blue, which would indicate that Māori had not yet reached a point at which blue and green were distinguished.

Kākāriki and matomato both are clearly defined as being the colour green. Matomato could possible derive from PEP *moto but kākāriki appears to have no PEP precursor. Kākāriki, in addition to being a word for green, describes four different objects that are green. The word literally means little parrot (kākā ‘parrot’, riki, ‘little’). The parrot in question is endemic to New Zealand so Māori would have not seen it prior to settling that land. Given this, it is possible that the name was first given to the parrot, then later attached to the colour category, and finally the colour word became attached to the lizard, shrub, and melon.

Yellow. In the English-Māori section, Tregear (1891) gives three words: pungapunga, kowhai, para. In the Māori-English section, we find that all three are plausible words for the colour category and that there is a fourth word, renga, which is related to words in other Polynesian languages.

Of these options, the modern educational resources all agree that kōwhai is today the accepted word for yellow. However, it is renga that is a direct descendent of the PEP term for YELLOW, *rega-rega. What is important here is the status of the word renga in Tregear’s dictionary. While Tregear is clear that renga is a word for four specific yellow-coloured things, he is doubtful that renga represents the colour yellow. The implication is that the brightly coloured kōwhai flower was such a compelling exemplar of the colour yellow that its name was adopted for the colour itself, supplanting the PEP-derived colour term renga. Māori is the only descendent of PEP to do this. All other languages that trace back to PEP have a derivative from PEP *rega-rega; Tregear explicitly notes that Samoan lega, Tahitian rea, Hawaiian lena, Tongan ega, Rarotongan renga, and Mangarevan rega all mean yellow. The transition from PEP *rega-rega to Māori kōwhai demonstrates a typical case of replacement innovation.

4.4 Stage V colour (blue)

Although blue is considered an early colour term in Berlin and Kay’s theory, there is no entry for blue in Williams (1871) or Tregear (1891). This indicates that, at the time of developing these dictionaries, Māori did not have a universally agreed word to describe what would be the English category of blue.

Finally, it is noteworthy that there are several compounds for the blue sky in the Māori-English part of Tregear (1891): kikorangi ‘the blue sky’; te kahu-o-te-rangi ‘the blue sky’; kahuraki ‘blue sky’. Given that Māori rangi is a regular reflex of Proto-Austronesian *langiC ‘sky’, kiko and kahu could be colour-related adjectives used in compound words such as ‘blue sky’. However, Rangi-nui is the name of the sky god in Māori mythology and the modern meanings of kiko (skin, flesh) and kahu (clothing, cloak) give these compound words the descriptive translations ‘the flesh of the sky god’ and ‘the cloak of the sky god’, both of which suggest that these words are describing the blue sky and not the more abstract concept of the colour blue.

Tregear (1891) lists three other words that can be translated blue but these refer only to dark blue and put the English concept of dark blue in the Māori BLACK colour category:

We thus see that Tregear’s words for the English blue either are referring to the Māori GRUE or BLACK category or are words for a specific blue-coloured object (the sky). The fact that Williams (1871) does not have any entry for blue supports this conclusion. There is thus no evidence that Māori had reached Stage V.

4.5 Beyond Stage V (brown, purple, orange, grey, pink)

Williams (1844) reports no terms for brown. The third edition of the dictionary (Williams 1871) and Tregear (1891) both list two words: ura and pākā, deriving from the homophonous word ‘dry’ listed in Williams (1844).

It is important to note that both words are ambiguous between the English categories of brown and red (POLLEX), as seen in (5). We can thus infer that these two colour categories were not distinct in pre-colonisation Māori, whereby the English brown is part of the Māori RED. We can thus discount Māori as having developed a separate colour category for the English brown.

No entries for orange, pink, or purple are attested in Williams or Tregear, except for paapura ‘purple’ (Williams 1844), which is clearly an English loanword. The word tawa – modern Māori purple (Moorfield 2011) – is present purely as the name of a tree (Williams 1844), as is karaka – modern Māori orange (Tregear 1891).

The modern Māori words for pink – māwhero and kuratea – also do not appear in Williams and Tregear. Hina, a reflex of Proto-Oceanic *sina ‘grey hair’, is given as the translation for grey in Tregear’s English-Māori section. However, in the Māori-English section, hina is given as ‘grey or white-haired’ rather than the colour grey. Williams (1844) also translated the word as ‘grey-headed’ rather than a colour term. The evidence is therefore that Māori, like Proto-Oceanic and PEP, had no word for grey per se, rather there is perhaps the hint that a word used to mean a specific coloured thing (grey hair) could have begun to be adopted to mean any thing of that colour.

4.6 Interim conclusion: changes in colour terms between Phases A and B

A comparison of the colour systems in Proto-Eastern-Polynesian and pre-colonisation Māori reveals a few major generalisations, (11a–c).

The evidence above indicates that pre-colonisation Māori had words for the colour categories: BLACK, WHITE, RED, YELLOW, and GRUE. It is not completely clear whether kōwhai had made the jump from being the name of a particular colour of flower to being a general colour term: the fact that whero could mean yellow could indicate that Māori was in a transition. It is thus remotely possible that Māori had reverted^[7] to four colour categories, with words for only BLACK, WHITE, RED, and GRUE, but the existence of the historic renga indicates that it is far more likely that kōwhai had supplanted renga. It is also not clear whether kākāriki had made the jump to being a generic term for the colour, but the existence of four different things all named kākāriki would indicate that it is likely that it had. It is fairly clear that there was, at the time, no generally accepted word for the English colour term blue. None of the other English basic colour categories has an unambiguous (or indeed any) equivalent in Māori.

On balance, the evidence is that the Māori of the mid-19th century had five basic colour terms. This makes it a Stage IV language in Berlin and Kay’s (1969) classification.

5.1 Māori colour terms as of 1950s (Phase Ca)

The table below (12) lists all of the colour terms reported in Williams (1957), which we assume to reflect Māori’s colour system as of the 1950s.^[8] This system features two major developments: the drastic expansion of Māori’s colour categories from five to 11 and the emergence of compound words for colours belonging to the less basic colour categories. It can first be seen that all Stage I–IV categories obtain new words. Surprisingly, between the 1890s and 1950s there was no further innovation for blue despite extensive contact with English. This suggests that there was a straightforward generalisation from the words for ‘blue sky’ to the colour concept ‘blue’.

A second major observation is that Māori had developed words to describe all 11 English colour categories, accompanied by the emergence of a series of English loans – parāone ‘brown’, karei ‘grey’, pāpura ‘purple’, ārangi ‘orange’ – and innovated compound words. Specifically, the category of pink is finally distinguished through the use of compounding strategy, combining existing colour terms for naming a less basic colour that is nonnative to Māori: kuratea as kura ‘red’ + tea ‘white’; māwhero as mā ‘white’ + whero ‘red’. This category, which does not exist in most East Polynesian languages, therefore demonstrates an interesting case: its establishment is clearly due to English influence, but not formed through an English loan.

It is important to note that many colour terms remain that refer to the pre-colonisation colour categories. This is seen in the apparent multi-functionality of several colour terms. For example, ura can refer to both English red and English brown; whero can refer to English red, yellow, or orange. Similarly, uri remains as a root for maka-uri ‘black’ and is also the form for green. This reveals that the language’s colour system had not yet conformed to English’s.

The increasing number of compound words and reduplications at this stage is also noteworthy. Many of the emerging terms for less basic colour categories, such as brown, grey, pink and in compound forms could be seen as evidence of these terms representing more basic categories, such as in tūā ‘somewhat’ + kura ‘red/brown’ for English brown, and mā ‘white’ + whero ‘red’ for English pink. See Section 6.2 for a discussion of various types of compound strategies attested in Phase C and later stages.

5.2 Māori colour terms as of 1990s (Phase Cb)

Language materials published around the 1990s reveal the beginning of levelling in Māori’s colour terminology. Ryan’s dictionary (first published in 1995), for example, provides evidence of how colour terms in Māori had developed in the half-century since Williams (1957). While there is as wide a range of colour words as in Williams, Ryan chooses to list selected colour names in a separate table at the start of the book (Ryan 2012: 16), which we assume, from context, indicates his understanding that these are the commonly used basic colour terms at the time of compiling the dictionary. Here, for example, in contrast to Williams (1957), whero is forced into the role of the English red, with other Māori words being given for the English concepts of brown and orange.

Of these, kirı̄ni, purū, parāone, kerei, and ārani are all loan words from English. Ryan uses the loan word kirı̄ni for green rather than kākāriki, although the latter is in his dictionary with the same range of meanings as in Tregear. That could indicate one of two things: that his sources are not confident that kākāriki represents all greens or that his sources felt that kākāriki represented something broader than the English concept of green, stretching into the English concept of blue (i.e., kākāriki was GRUE), thereby needing a different word, kirı̄ni, that is limited to the English concept.

Bauer (1997) gives a slightly different list in her contemporaneous work (14), where the 11 English basic colour terms are mapped to Māori words. Here we see that she has assigned Māori terms for all colours up to Stage V, then loan words or compounds for brown, grey, orange, pink, and purple. This again indicates support for Māori incorporating loan words for concepts that were not present in historic Māori culture.

5.3 Māori colour terms after 2000 (Phase D)

The early 1970s onwards saw a bottom-up community attempt to revitalise the Māori language, after decades of severe decline (Kerttula 2018). After a slow beginning, this movement gained substantial traction in the 1980s and ’90s, with the development of kohanga reo (lit. language nests, Māori-language pre-schools) and then primary schools where Māori was the primary language (May and Hill 2018; Smith 2000).

The table below outlines colour terms reported in Biggs (2012), Moorfield (2011), and Tauroa (2006), and the Te Aka Māori Dictionary (https://maoridictionary.co.nz/), an ongoing dictionary project that complements the series of four textbooks and related resources in the Te Whanake series for learning Māori.

The sheer number of variants reported for more primary colour categories deserves a note. According to available descriptions in the dictionary, some of these terms denote a specific subtype of the category. For example, karera and horepara refer to light green, and uri refers to dark green, whereas kaakanapa is translated as ‘green, as deep water’; the differences among several variants of, for example, white and brown are also documented: ahoaho ‘pure white’, mā ‘white, pale, clean, faded’, hāura ‘brown’, hiwera ‘brown or red, as if burnt’, kākaka ‘brown, rusty coloured’, and mangaeka ‘pale brown.’ However, the exact difference between some variants – especially those reported in earlier dictionaries – remains unclear and awaits future investigation.

A key question crucial for understanding the development of Māori’s colour lexicon following the national revitalisation efforts is how colours are taught in the Māori language, as this will influence how future speakers of Māori think and speak about colour. We thus compiled colour names from 16 different educational resources provided for pre-school and primary school use. These are tabulated in (16). The official resources from the New Zealand Ministry of Education (entry 15 in the table) have ten colour categories that match all the English categories other than grey.

The first point to note is the rejection of English loans. Following the revitalisation efforts starting in the 1970s, there has been a strong push in the community to adopt words that are historically attested, though those words were not historically used for the colours. These words are taken from the names of plants or animals that are distinctively of that colour.

As seen above, for most basic colour categories, there is general consensus about what word should be taught to children. The major exceptions are brown and purple. For purple, the educational resources generally agree on poroporo or waiporoporo (variations on the same root) but disagree with the Ministry of Education’s tawa. Poroporo and tawa are plants with purple flowers and purple fruit respectively. For brown, we have disagreement between the loan word parāone, the compound parauri (lit. dark yellow), and pākākā (scorched, reddish brown).

The other notable point is that there is general agreement that there are two equally valid words for black, pango and mangu, and two equally valid words for blue, kahurangi and kikorangi. This could be considered parallel to the French use of two words, brun and marron, for one particular English colour category, ‘brown’ (Forbes 1979). We note that two of the resources use the existence of two words for blue to distinguish between dark and light blue, as in Russian, but we can find no independent scholarly source to justify this distinction.

6.1 The evolution of Māori’s colour system: how does Māori correspond to Berlin and Kay’s theory?

Examining Māori’s colour system over time (Sections 3, 4, and 5, see also a comprehensive list of colour terms reported in each phase in the Appendix) reveals two overarching themes in how colour naming developed. The first is the lack of progression on Berlin and Kay’s hierarchy during the first 500 years of settlement of New Zealand. Although there was a wide range of replacement innovations of existing PEP colour terms, Māori remained in Stage IV. This is potentially due to the absence of linguistic and cultural stimulation following the settlement of New Zealand, which had no prior human habitation, that is, there was no pressure to add new categories. Some of the replacements are clear cases of semantic extension, many of which derive from floral and faunal terms native to New Zealand. The table below outlines a few representative cases.

It is noteworthy how fluid the colour terms are from PEP to pre-colonisation to modern use. Over the same period of time in which the most basic colour terms of English merely changed spelling (hwit/white, blæc/black, réod/red, geolu/yellow, groene/green), Māori basic colour names underwent considerable change. Consider these four examples:

1. PEP *rega-rega ‘yellow’ clearly led to Māori renga ‘yellow’ but, by the 19th century, renga was no longer the generally accepted term for the colour. We have already discussed the idea that kōwhai could have replaced renga because the kōwhai flower was such a fine exemplar of the colour and would, therefore, have been useful in teaching children the colour.

2. Similarly, the endemic little green parrot kākāriki gave its name to the colour green supplanting whatever PEP word had been in use prior to the settlement of New Zealand.

3. PEP *kula ‘red’ clearly led to Māori kura. The two early Māori dictionaries both have kura and whero as red. It is not clear how and when whero ‘red’ became universally accepted as the word for RED, but the evidence is that the transition had already begun in the early 19th century and was definitely completed by the late 20th century.

4. PEP *tea is identical to Māori tea, which meant both white and pale in pre-colonisation Māori. However, today the agreed word for white is mā, with tea retaining its use as an alternative for white, but also meaning light in colour, and being a suffix to indicate lightness, for example in ōrangitea (light blue).

The second overarching theme concerns changes following the extensive interaction with English and Māori’s recent grass-roots revitalisation. This has caused the language to move from Stage IV pre-colonisation to a position where it has repurposed existing Māori words to represent those English colour categories that did not previously exist in Māori’s native categories. The evidence from Ryan (2012) and Bauer (1997) is that this repurposing is a very recent phenomenon. It is possible that the introduction of kohanga reo (Māori language pre-schools) led to a desire for teaching resources for young children; this led to a direct translation of English language training materials, with their inherent assumption about basic colour terms and the boundaries between different colour categories. This, combined with there being no adult monolingual speakers of Māori, could mean that learners of Māori adopt English’s colour categories. Further research would be needed to ascertain whether there remains, amongst speakers of Māori, the vestiges of the pre-colonisation colour categories.

The educational resources indicate the following colour terms (18) are the current consensus in the educational community for each of the English colour categories. This indicates that the Māori colour categories now match English’s categories, if kikorangi and kahurangi both represent English blue. An expected outcome in the near future is that children taught with these educational resources would not be aware of the significantly different nuances in the meanings of traditional Māori colour terminology.

Three of the modern terms for the colours are based on historic usage, six are based on the colours of particular natural objects, with two of those having been adopted prior to colonisation (kowhai ‘yellow’ [a flower] and kākāriki ‘green’ [a bird]). Two categories (black, blue) each have two equally valid Māori words, two categories (brown, purple) are not settled, each with three variants included. Māwhero ‘pink’ is one of two terms that is not monolexemic, something we discuss in Section 6.2. Parauri ‘brown’, the other term that is not monolexemic, is literally dark yellow (para-uri). This is intriguing in that one might have expected the word for brown to be more closely related to red than yellow. It also appears that this is a colour where the community has not yet settled on the term that will end up dominating the space, with various sources selecting different terms: parāone, parauri, pākākā. Parauri appears more often in the educational resources than the other two options. The situation for brown differs from the situation that pertains with mangu/pango ‘black’ or kikorangi/kahurangi ‘blue’. In those two cases, most sources have both competing names as equally valid, whereas all but one source states only a single word for brown, they just do not agree on what that single word should be.

The evolution of Māori’s colour system thus demonstrates how colonisation with and without language contact may impact the basic vocabulary of a language differently – both in the Māori settling a new land (no prior human habitation, with lexical innovations but no expansion in colour categories) and in their subsequent contact with English (extensive language contact, resulting in drastic expansion of colour categories). This pattern of progression in colour categories under colonisation-induced contact demonstrates a strikingly similar case with Bulu (Bantu), the colour system of which has expanded from three to six with lexical borrowings from French, the colonial language (Grimm 2012). See also Huisman et al. (2022) for a similar case of categorial expansion in colour lexicon under sociocultural pressure observed in the Japonic family. Notably, New Zealand Sign Language has also undergone a similar developmental pathway in line with Berlin and Kay’s hierarchy, expanding its colour system from Stage I (the colour black and a generic colour sign) to Stage VII (11 colours) following extensive contact with Australian Signed Language via Australasian Signed English in deaf education from 1979 and ongoing language contact between the two countries (McKee 2016).

6.2 Borrowability, basicness, and word-formation strategies used for forming less basic colour terms

An important observation from the development of Māori’s colour lexicon is that none of its five native categories – BLACK, WHITE, RED, YELLOW, and GRUE (Section 5) – adopted any loans throughout Māori’s contact history with English. This is in contrast to the other colour categories that rank later in Berlin and Kay’s hierarchy, all of which have accrued various English loans, except pink (19). Note that there are loan words for blue purū and green kirı̄ni, which indicates that the Māori category GRUE did not correspond neatly to either of those two English categories. All the words for the category GRUE became attached to the new category green, with the word for blue sky repurposed for the colour category blue. Similarly, the Māori words in the category RED incorporated English concepts of red, brown, and orange, but it is the brown and orange that accrued loans and new words following the contact, whereas red did not.

This observation follows from the well-established generalisation that basic vocabulary are more resistant to borrowing (Thomason and Kaufman 2001: inter alia). The basicness of these terms is not only suggested by Berlin and Kay’s hierarchy, but also by the fact that all five are included in the Swadesh 100 list (Swadesh 1971), ranked from 87 to 91 (in the order of red – green – yellow – white – black). Notably, colour terms under these five categories also demonstrate high derivational productivity, which has been argued to be a major criterion for basicness (Kerttula 2007).

It is also noteworthy that the extensive loans attested in Stages VI and VII follow consistently from Thomason and Kaufman (2001)’s generalisation that borrowing of “nonbasic” vocabulary usually commences when a language reaches Stage 3 of the five-point scale of intensity of contact (20), by which Māori has undoubtedly passed Stage 3 following European colonisation.

Beyond cases of English loans, the word-formation strategies attested with terms associated with categories nonnative to Māori also deserve a note. Anishchanka (2007) discusses seven cross-linguistically common strategies for forming non-basic colour terms (21):

Four of the seven strategies are attested in Māori, alongside the frequent use of reduplication to form variants under the same colour category (e.g., teatea ‘white’, hiwahiwa ‘black’, pungapunga ‘yellow’, kiwikiwi ‘grey’. Some examples are presented in the table in (22).

The emergence of various types of compound words following intensive contact with English, such as māwhero ‘pink’ and parauri ‘brown’, clearly demonstrate that these colour categories have been grafted into Māori culture rather than developing naturally. Among these cases, the words for pink deserve a special note: while pink is definitely a basic colour term in English, the word māwhero should clearly not be a basic colour term under Berlin and Kay’s rules, because it is not monolexemic. There is, indeed, no historically attested word for pink reconstructable to any proto-level within the Austronesian family (ACD, POLLEX). Further evidence for the colonisation of the concept of the colour category pink comes from the list of words translated pink in Te Aka dictionary: kuratea (red + white), māwhero (white + red), mākurakura (white + red + reduplication), māwherotea (white + red + white), pūwhero (intensifier + red), pūwhero mā (intensifier + red + white), pūwherowhero (intensifier + red + reduplication), all of which are a compound of two or three of white/pale (mā, tea), red (whero, kura), and an intensifier (pū).