TRANSLATIONS
 
next page previous page up home

My sudden insight that the henua type of glyph (GD37) probably has as its origin the fact that trees all over the world marked time not only by illustrating the seasons through their visible changes (flowering, leaves budding, fruits ripening, leaves being dropped etc), but also - more subtly - by some species being picked out as seasonal markers regularly located over the year (e.g. 13 kinds of trees each 'governing' 28 days), may explain the occurences of e.g. these types of glyphs:

Ab6-42 Ab4-23 Aa7-74
Qb6-128 Kb3-15 Eb3-19

Metoro normally said henua when referring to GD37 glyphs. That word does not mean 'tree' or 'wood', but is an expression which describes what the GD37 glyph symbolized, viz. the 'land' (which changes in its cycles over the year):

Henua

Land, ground, country; te tagata noho i ruga i te henua the people living on the earth. Placenta: henua o te poki. Vanaga.

1. Land, country, region; henua tumu, native land. 2. Uterus. 3. Pupuhi henua, volley. Churchill.

M.: Whenua, the Earth; the whole earth: I pouri tonu te rangi me te whenua i mua. 2. A country or district: A e tupu tonu mai nei ano i te pari o taua whenua. Tangata-whenua, natives of a particular locality: Ko nga tangata-whenua ake ano o tenei motu. Cf. ewe, the land of one's birth. 3. The afterbirth, or placenta: Ka taka te whenua o te tamaiti ki te moana. Cf. ewe, the placenta. 4. The ground, the soil: Na takoto ana i raro i te whenua, kua mate. 5. The land, as opposed to the water: Kia ngaro te tuapae whenua; a, ngaro rawa, ka tahi ka tukua te punga. Text Centre.

We have here an example showing that Metoro might have 'translated' the rongorongo glyphs not by explaining the picture but by explaining what the picture meant. We must remember that.

At the start of the X-area we do not find henua (Gd37) but niu (GD18), although much points at these niu glyphs are standing at the exact beginning of the new year:

Aa1-13 Pa5-67

How may that be explained? Is a picture of a tree too trivial? The halo around the head of a holy 'person' would be difficult to picture in a rongorongo glyph (because such glyphs show only outlines). Radiating light in all directions would be easier drawn (as in niu).

The picture is from Wikipedia: 'Madonna and Child' by Ambrogio Lorenzetti. I searched for 'swaddling', with the idea that I would find some picture of how Polynesians swaddled their newborn babies. However I did not find any such information.

Newborn babies do not show their arms and legs, because they have been swaddled. That could explain the shape of Pa5-70 (and similar glyphs):

"The child has hardly left the mother's womb, it has hardly begun to move and stretch its limbs, when it is given new bonds. It is wrapped in swaddling bands, laid down with its head fixed, its legs stretched out, and its arms by its sides; it is wound round with linen and bandages of all sorts so that it cannot move …

Whence comes this unreasonable custom? From an unnatural practice. Since mothers despise their primary duty and do not wish to nurse their own children, they have had to entrust them to mercenary women. These women thus become mothers to a stranger's children, who by nature mean so little to them that they seek only to spare themselves trouble. A child unswaddled would need constant watching; well swaddled it is cast into a corner and its cries are ignored … 

It is claimed that infants left free would assume faulty positions and make movements which might injure the proper development of their limbs. This is one of the vain rationalizations of our false wisdom which experience has never confirmed. Out of the multitude of children who grow up with the full use of their limbs among nations wiser than ourselves, you never find one who hurts himself or maims himself; their movements are too feeble to be dangerous, and when they assume an injurious position, pain warns them to change it." (Wikipedia citing Jean Jacques Rousseau - Emile: Or, On Education, 1762.)

Barthel suggested that 'no-legs-visible' might indicate 'a god' (gods do not need to move):

"Für das Zeichen 290, [the type of glyph which I have here exemplified by Hb8-136 - maybe 'a swaddled person' with winglike arms]

das den Götternamen 'mata viri' bzw. 'ruanuku' vorangeht, wurde die Lesart 'atua' noch in weiteren Textzusammenhängen geprüft."

"Im Tafeltext Gr5 folgen die Zeichen 20 [he means the right part of Ga5-10] und 79 [Ga5-11] als 'mata - viri' aufeinander. Es liegt nahe, das vorangehende Zeichen 290 [the left part of Ga5-10] als 'atua' zu deuten. Zeichen 290, mit seinem massigen Leib und den fehlenden Beinen, könnte möglicherweise ein Idol aus Stein oder Holz darstellen und den Begriff der Gottheit symolisierien."

-
Ga5-10 Ga5-11 Ga5-12 Ga5-13 Ga5-14 Ga5-15 Ga5-16
...
Kb1-12 Kb1-13 Kb1-14 Kb1-15 Kb1-16

As to ruanuku he refers to Ab6-82:

Ab6-76 Ab6-77 Ab6-78 Ab6-79 Ab6-80
Ab6-81 Ab6-82 Ab6-83 Ab6-84 Ab6-85

Neither of these two examples convinces me that we should read atua (mata viri respectively ruanuku). Ga5-10--16 (and the parallel text in K) has as its subject matter the period when sun turns around at midsummer and the right part of Ab6-82 may symbolize the 'double faces' as seen in Ab6-47 and Ab6-54 (possibly Mars respectively Venus).

Though of course Barthel may still be right - given that there are gods representing midsummer respectively the double 'faces' of Mars / Venus.

'No legs' may also be a way of making threatening 'persons' harmless. In ancient times they used to bury (dead) people with legs and arms immobilized by ropes. Which reminds me of an image of 'Oro (a war god):

"The war god Oro. Oro was worshipped all over Polynesia, but in Tahiti not represented by a carved human figure. Instead he was a bundle tied up with cords over a wooden core; limbs and facial features can, however, be recognized. Pleaded sinnet cloth on wood. Tahiti." (Larousse)

"Vermutlich können in der Osterinselschrift unter den Vorkommen von Zeichen 290 noch weitere Götternamen ausfindig gemacht werden. Auch das Zeichen 1 [i.e. henua, GD37] in seiner begrifflich verwandten Rolle kann hierfür als nützlich gelten. Mit 'toko' wurden auf Neu-Seeland die holzgeschnitxten Götterstäbe, auf Tahiti pfostenartige Figuren der groβen Gottheiten bezeichnet. An diese übertragene Bedeutung sollte man bei der Beurteilung des Stabzeichens denken. In den Tafeltexten scheint jedenfalls das Zeichen 1 an einigen Stellen direkt für das Zeichen 290 eintreten zu können, d.h. 'toko' für 'atua' gesetzt worden zu sein."

He gives no examples, though, of where in the rongorongo texts his number 1 is equivalent to his number 290, and I cannot remember any parallel texts by means of which this conclusion may be drawn.