The only language historically attested for Easter Island is Rapanui, which belongs to the Eastern Polynesian language family. This makes it highly likely that the rongorongo inscriptions represent a pre-
The signs for the plain vowels /a/, /e/, /i/, /o/, and /u/ are also used for the syllables starting with consonant /h/ or the glottal stop /'/ – alternatively it could be said that /h/ and the glottal stop were not written. Vowel length is not marked which means that, for example, the ra-
Words: single element glyphs
With a few possible exceptions, such as ŋu, koru, ti, and ve, basic glyphs are also used independently as word signs. These include most of the grammatical markers that are present in the inscriptions. Some of the disyllabic signs seem to function only in this capacity (e.g., nui, tea). In the following enumeration, meanings that are not found in Rapanui vocabularies or grammars are marked with an asterisk: 'a, 'of' (possessive alienable); *ā, 'to surround' (1); 'ā, postverbal (continuous) marker; ana, 'if' (irrealis); 'ana, identity marker; e, exhortative; agent marker; hē, question marker; i, accusative marker; perfective marker; (')i, 'in', 'at' (preposition); ī, 'to corrupt', 'to rot'; ina, 'to dislike', 'to hate' (2); o, 'of' (possessive inalienable); 'o, 'because of'; hū, 'to burn', 'to be(come) angry'; ŋa, plural marker, ŋa (or ŋā), 'exhaustion', 'to exhaust' (3); ŋi'i, 'glare of the sun', 'to glare'; ka, aspect marker, kā, 'to ignite', 'to be/make angry'; kai, 'food', 'to eat'; kē, 'other', 'different'; ki, 'to' (preposition), 'when' (preverbal temporal marker), 'in order that' (preverbal purpose marker); kī, 'to speak'; ko, prominence marker; 'overthere' (locative); ma, 'for' (benefactive alienable); ma'a, 'to know', 'to understand'; mai, 'from'; 'hither' (directional); mo, 'for' (benefactive inalienable); nā, demonstrative (medial); na'a, 'to hide'; nei, proximal marker; nō, 'just'; nui, 'big', 'great', 'large size'; pa (or pā), 'obstruction', 'to obstruct'; pe, pehe, 'like'; pihi, 'to end'; pō, 'to obscure', 'darkness'; pū, 'hole', 'to go into a hole'; rā, intensifier; ra'ā, 'sun'; rehe, 'weak', 'to weaken'; riva, 'good', 'healthy' (4); rō, emphatic marker; ru (or rū), 'to tremble'; ta (or tā), 'colour', 'to colour'; tea, 'white'; tehe, 'to flow', 'to menstruate', 'to dissolve'; to'o, 'to take', 'to accept', 'to remove'; tohu, 'curse(d)'; 'to curse'; *tu (or tū), 'to deform' (5); vaha, 'crevice'; vae, 'to choose'; va'e, 'foot', 'leg'.
Words: multiple element glyphs
Singular glyphs, i.e., independent syllabic or disyllabic signs, account for roughly 30 percent of the signs in the inscriptions. The remainder of the texts consists of 'fused' glyphs, i.e., signs that are composed of two or more basic elements. The process of fusing elements produces both words and strings of words. The latter may consist of a nominal or verbal nucleus, preceded and/or followed by one or more grammatical markers, occasionally also including adverbs or adjectives. Although the rules for the fusion process appear to be set by word formation and grammatical structure, this does not necessarily mean that multi-
The two most common methods of sign composition are the 'stacking' of two components on top of each other and the "gluing together" of two adjacent signs either directly or through the use of an 'arm', 'wing', or some other part. Some researchers have advocated a fixed bottom up reading order for stacked glyphs as this seems to be indicated by some parallel text fragments. However, as there are also examples that suggest the contrary, nothing has been proven for the inscriptions in general. More importantly, if there had been such a thing as a fixed reading order, it would have severely limited the scribes’ range of combinatory possibilities.
The scribes' options were further expanded by the existence of allographs and the possibility of using only parts of signs. In particular, the signs for i ('i, hi), no, ra, and ta are often reduced to 'fingers' (fig. 3a), 'thorns' (fig. 3b), 'hairs' (fig. 3c), and 'head' (fig. 3d) or 'body' (fig. 3e), respectively, thus facilitating the creation of compound signs. One of the advantages of this system is that (near) homonyms such as toa, 'sugarcane' (fig. 3f), and to'a, 'enemy', 'hostile' (fig. 3g), and tau, 'beautiful' (fig. 3h), and tau, 'period' (fig. 3i), can easily be distinguished.
Composite signs are often constructed with an anthropo-
Several methods were employed to save precious writing space. Occasionally, signs are miniaturized to fit in open spaces (fig. 5a-
A specific method of abbreviation could be used for the many reduplicated words of the type ABAB: one of the syllabic components was left unwritten, which means that the phonetic value of the remaining component has to be repeated by the reader (ABA(B)) (fig. 5e). At least in one case this strategy was also employed for a word of the ABA type: the very frequently occurring glyph for ‘man’ or 'person', taŋata, is written with only a single ta-
Fig. 6 shows examples of fused signs, including the most common words in the inscriptions. They are arranged according the order of the syllabic grid. As can be seen, they are mostly composed of two elements with no fixed reading order other than left to right (asterisks indicate hypothetical terms based on cognates in related languages): aha, 'what', 'which'; 'ana, 'cave'; 'a'ano, 'wide', 'width'; ara, 'path', 'road'; atu, 'away' (directional); haŋa, 'to want', 'to accept', 'to like'; 'aŋahē, 'when?'; haŋai, 'to feed'; haha, 'to feel', 'to grope (in the dark)'; hara-
The shape of the constituting components of a word seems to have played a role in whether it was written as a fused sign. Apparently, if words required extra elements to fuse their parts together they could also be left unconnected. Examples of this are words such as hehe (fig. 7a), heka (fig. 7b), and oho (fig. 7c). In addition, there is a group of words that – likely for the same reason – were never written as fused signs, such as haha, 'mouth' (fig. 7e), toa, 'sugarcane' (fig. 3f), to'a, 'enemy' (fig. 3g), and veve, 'poor', 'wretched', 'haste', 'to hurry' (fig. 7f).
(1) ā: the meaning is possibly connected to RAP 'to drive', 'to herd', as this may involve encirclement. The term is, for example, interpreted as such in Barthel 1974:114.
(2) ina: the phonetic value of the sign has been assumed to derive from mahina, 'moon', hypothesizing that it is a representation of the crescent moon. If this proves to be valid, it would suggest that the breast ornament rei is also a representation of the moon. Possible cognates are Tuamotuan *ina, 'to be enraged', ina-
(3) ŋa or ŋā: this meaning is only attested in a reduplicated form: gaga, 'to faint' (Churchill 1912, 175); ŋaŋa, 'estar exhausto, agotado, sin fuerza, desfallecer' (Englert 1978:111).
(4) riva: possibly, this is not a basic sign as it may be a composite of the "wing glyph" ri and the "boathouse glyph" va.
(5) tu or tū: the precise meaning has not been determinded. As the verb seems to be involved in the changing of shape, size, or colour of objects, the provisional term 'to deform' is used. Possibly, the word is connected to haka-
(6) taŋata: if the hypothesis of syllabic reduplication is correct, the sign could also be read as ŋaŋata, the plural form of taŋata. Interestingly, the inscriptions appear to use the sign mostly for 'people'. As Rapanui uses both terms in a plural sense, it will be difficult to determine the correct reading.
(8) hehe: cf. CE: feefee, 'a boil or abscess'; HAW: heehee, 'a boil, running sore'; MQA: heehee, furoncle, clou, tumeur, inflammation'; MAO: wheewhee, 'boil, abscess'; TAH: feefee, 'furoncle'; TUA: henga, 'a lump, swelling'; TVL: fete, 'swollen, to swell' (POLLEX).
(9) mā or ha'amā: if the first value is correct, this could be a case of marking the long vowel or of overspelling. The latter may have been necessary as the "body" glyph could not be written independently, i.e., it required a "head". The independent ma-
(10) As the reduplication mama is represented by a wide open beak, the glyph's value probably derives from mama, 'to gape', 'to yawn'.
(11) nohu: the combination usually appears in a more contracted form, somewhat resembling a spined fish. This suggests that it may have been assembled with certain fish with poisonous spines, in particular the scorpion fish (nohu), in mind.
(12) raŋirua,: the sign appears to be a combination of two other composite signs, raŋi and rua. The term does not occur in Rapanui vocabularies but may be related to MAO rangirua, 'ambiguous', 'having two aspects', 'in doubt', 'uncertain', 'irregular' (Williams 1957:324).
Fig. 1 Syllabic grid
Fig. 6 Fused signs
Fig. 2 Disyllabic signs
ina koru kai mai mama nei nui
riva tea to(h)u tua va(')e rapa (?) hare (?)
a b c d
e f g h i
a b c d
a b c
d e f
a b c
d e f