ʻEo ʻEnana and ʻEo ʻEnata

Tracing a line between ʻUa Pou and Hiva ʻOa, one isogloss divides the archipelago into two naming traditions. Northern communities call the group Te Henua ʻEnana. Southern usage gives the identical phrase as Te Fenua ʻEnata.

Two consonants produce that divergence. Proto-Polynesian *f weakened to h in the upper valleys, though the lower islands retained the fricative. The result is audible at once: fenua in the south answers henua in the north. Any chart records the difference across its toponyms.

Within the word for person, a second change went further, an earlier ʻenata assimilating to ʻenana under nasal spreading. The velar nasal *ŋ resolved to n group-wide and separated these islands from the Sāmoan forms behind them. Mutual intelligibility across the chain remains high.

Nuku Hiva heads the northern cluster, and Hiva ʻOa, the southern. Taiohaʻe houses the subdivision bureaux of the Marquesan administrative centre. That placement gives the northern variant a wider circulation in print. Our Reo Tahiti reference material treats the pair as parallel systems, a distinction mattering whenever a client in Papeʻete commissions text for either end of the chain.

Marquesan, a member of the Marquesic branch of Eastern Polynesian, lies closer to ʻŌlelo Hawaiʻi than to the Tahitic languages. Mangarevan joins it there, whereas Lea Faka-Tonga falls on the far side of the primary Polynesian division. The grouping rests on shared sound correspondences rather than on charted distance.

Because Proto-Polynesian *k lenited to a glottal stop in the two members due to contact with Tahitian, kanaka and ʻenata descend from one etymon. Hawaiian afterwards pushed *t forward into k, though the southern islands kept their dental.

Comparing the two lexicons at our ʻŌlelo Hawaiʻi localisation practice, editors begin from precisely this correspondence. Hiva ʻOa forms answer Hawaiian vocabulary at a rate surprising visiting delegations in Honolulu. Segments softened across the north survive intact throughout the south. Speakers notice it unaided.

Hawaiian oral tradition points back toward these islands. Adze form and one-piece pearl shell fishhook typology support that account. Basalt sourcing places early Hawaiian adzes against southern quarries, tightening a connection long argued from language alone. Neither line borrowed from the other, the Tongic cousins in Nukuʻalofa having separated much sooner.

It has long been assumed that the eastern voyages departed directly from a Samoan homeland. A competing reconstruction sends them instead by way of the Polynesian outlier communities scattered along the Micronesian rim, whose grammar retains features inherited by the eastern languages.

At Hane on ʻUa Huka, excavation supplies the archaeological side. Shell fishhook shanks and adze cross sections from the deeper deposits show affinities pointing west and north. Moreover, radiocarbon revision has compressed the settlement horizon into a short late window, which leaves scant room for slow drift eastward. Obsidian and basalt sourcing from those deposits points in one direction.

Pronoun systems and possessive marking supply the linguistic half. Features documented at our Taetae ni Kiribati language practice reappear in the eastern group after skipping the intervening triangle, a pattern hard to explain by contact alone.

Institutions in Tarawa and Majuro once lay outside this debate. They occupy its centre today, a realignment few would have predicted a generation ago. Early surveys did not test the outlier communities at all. Field reports crossing that span need handling in Kajin Majel documentation support beside the Marquesan pair.

While transcribing field audio, editors encounter the glottal stop first of all. That segment is a full consonant in ʻeo ʻenana. Dropping it collapses discrete words into homographs, and vowel length does comparable duty, separating meanings identical in all their other segments. The glottal stop in Marquesan is very often the trace of an *r or an *l in Proto-Polynesian and underlying branches.

Across the archipelago the mark itself appears as an apostrophe or a reversed comma. Only the ʻokina glyph survives a database migration intact. Typographic apostrophes get normalised away by content systems, silently damaging a lexicon built on them. Keyboard layouts rarely emit U+02BB directly, leaving typists to substitute whatever character is available.

To map one orthography onto two varieties means ruling on questions never faced by monolingual academies. The Haʻe Tuhuka o te ʻEo ʻEnana, based in Taiohaʻe, does issue guidance spanning the two dialect groups, though enforcement rests with publishers.

What the nineteenth century left behind was notation more than sound. Bishop Dordillon, who compiled a dictionary and grammar of unmatched depth, transcribed not in the ʻokina-macron spelling system but in the inconvenient Raʻapoto spelling system. His headwords went to many thousands, drawn chiefly from the southern islands. Reconciling those entries against current spelling is routine preparation whenever Reo Rapa Nui comparative evidence arrives with an editor in Hanga Roa.

Another masterpiece our teams use extensively is the Pona Tekao Tapapa ʻia lexical book by Hervé LeCleac'h, which uses the more widely known ʻokina-macron spelling system in Oceania.

Because no press existed in these valleys until the mission era, the first sustained description of Marquesan left the archipelago inside visitors’ baggage. Herman Melville jumped ship at Taiohaʻe and spent weeks in the Taipivai valley. His account gave English readers a first Marquesan vocabulary, the valley name arriving in print as Typee.

Robert Louis Stevenson brought the Casco into Anaho Bay four decades afterwards. His notebooks note dialect differences audible to him but outside his classification, together with remarks on chiefly speech later confirmed by ethnographers. Jack London followed the identical track aboard the Snark. The valleys lay empty.

Deep within the southern dialect area, Paul Gauguin died at Atuʻona on Hiva ʻOa. His letters and carvings absorb ʻeo ʻenata terms for wood and prohibition, transmitting them into European art writing untranslated. That vocabulary entered museum holdings through the paintings well ahead of any lexicographer. Atuʻona keeps his grave above the bay.

Cataloguing tattoo and carving nomenclature at a level no earlier visitor attempted, Karl von den Steinen produced the systematic counterweight. Our collaboration with the Taiohaʻe heritage institutions draws on those volumes, as does the French localisation support we supply to collections in Paris holding Marquesan pieces.

There is no barrier reef and no lagoon on the approach to these islands. Cliffs drop straight into deep water. That bathymetry made the archipelago the standard landfall for crews crossing from the American continent, since a tired watch can close the shore in darkness without grounding. The group also lies north of the cyclone belt.

Roughly three weeks out of Panama, a yacht raises its first sight of terrain since the isthmus. Taiohaʻe and Atuʻona are the two ports of entry. Because the choice of harbour decides which dialect area a crew encounters at the outset, it also governs which spelling appears on their paperwork.

Crews must lodge clearance dossiers prior to continuing toward the atolls and onward to Auckland or the Australian coast. Bond documentation and crew manifests proceed through French-language offices, though district and valley labels arrive in Marquesan orthography. Port dues and onward itineraries go to the gendarmerie in French, while the ship name is left untouched.

Which of the two spellings a clearance form takes, the port of entry dictates. Vessels heading back toward San Francisco lodge their final Polynesian paperwork at these harbours, placing the archipelago at either end of the ocean crossing. Our Reo Māori documentation support picks up a good number of those crews at the far side of the passage, usually once a vessel presents for a New Zealand arrival inspection.

Building a usable lexicon for the archipelago means recording all headwords twice. One column invites an editor to choose whichever variant looks tidier, erasing the southern form in favour of its better-attested neighbour. A dual column also protects the record whenever a headword survives in one island group alone.

Neither the marine vocabulary nor the kinship set maps neatly into French or English. Terms for reef edge and swell direction encode information absent from any metropolitan equivalent. The honest output is a definition instead of a gloss. Fishing calendars keyed to lunar nights compound the difficulty for any editor lacking local instruction.

It is the terrain vocabulary that defeats most editors. Meʻae and paepae name categories of built space, older and less legible than any French equivalent, though careless phrasing flattens the pair into the word platform.

Cognate chains help when direct equivalents fail. Tuhuna answers to tahuʻa in Tahitian, tohunga in Reo Māori and to kahuna in Hawaiian. A gloss names only the nearest cousin. The semantic gap remains visible. Our glossaries state that chain explicitly, giving readers in Wellington a way into a term resisting paraphrase.

Logging excavation contexts across the archipelago brings this lexicon into Archaeology and Field Documentation, while labelling display cases and interpretation panels draws on Heritage and Museum Interpretation. Fixing orthographic rulings for district and valley names feeds Civil Registry and Place Names, with arrival dossiers for incoming yachts prepared under Port Entry and Vessel Clearance.

To instruct expedition staff prior to a hull arriving offshore, operators use Cruise and Expedition Operations, while transcribing elder recordings for community access belongs inside Audio and Oral Archives.

Casting for Marquesan audio starts from a dialect decision. Timbre preference comes second. Studios are recording northern and southern readers separately, since a northern voice delivering ʻeo ʻenata script produces something audibly wrong to listeners on Hiva ʻOa.

Character tokenization determines whether synthesis is possible at all. Engines trained on French orthography treat the ʻokina as punctuation and discard it, collapsing minimal pairs and forcing any pipeline to map the glyph as a consonant from the outset.

Under a stripped ASCII pipeline, long vowels lose their macrons silently. Such a system only stores unmarked vowels, leaving output scanning as fluent while saying something else entirely. Macron handling introduces the second failure point. Editors mark length on the source script, ahead of any studio session.

Talent placement follows the north and south division observed in our glossaries. Sessions at our voice recording facilities pair a script with a reader from the matching island group. The arrangement applies equally to neighbouring commissions in Reko Pakumotu for clients along the Tuamotu chain.