Part 9
The dictation check
This scene tests whether its lexical markers survive careful spoken transmission. It does not test exact recovery of every period. Sentence-end is the one boundary Phi deliberately leaves to a silent mark.
01The question arrives
siora comes from the neighbouring house with sulae's exact words:
ne siora kau womu shua. ne siora haolu.
NAME siora ALL home come. NAME siora speak.
ne sulae sha wa lo thia kau lo mia so shua sho to thilou.
NAME sulae QUOT.COMP Q PL 2SG ALL PL 1SG FUT come QUOT.COMP.CLOSE PST inquire.
The direct question begins with wa inside sha ... sho. Its lo mia is still sulae's "us," because an exact quotation preserves the original speaker's reference. Motion toward that group uses allative kau. Outside the frame, ne sulae identifies the asker and to thilou reports the past inquiry.
Both mentions of siora take ne because the example uses portable neutral Phi. Established conversation could omit the second marker once the referent was clear, but physical presence would not be the reason.
02The reply goes back
The parent gives siora an exact reply to carry:
no wei ne sulae sha kona ne sulae. lia. lo mia kau lo thia so shua. su theula towe nai sho haolu.
IMP DAT NAME sulae QUOT.COMP VOC NAME sulae. yes. PL 1SG ALL PL 2SG FUT come. OPT UNIV well be QUOT.COMP.CLOSE speak.
The outer directive has a recipient, wei ne sulae, followed by the exact words and final haolu. Inside the quotation, kona ne sulae. calls the recipient, lia. answers, the future clause states movement toward the addressee, and su opens a wish. The several periods inside the quotation are visible divisions among its utterances; the lexical forms do not pretend to replace them.
03What to compare
A transcription can be checked at two levels. First compare every word. Pay particular attention to frame boundaries and names. Then compare the periods separately. A difference in sentence division matters for the written text, but it does not disprove the narrower claim that these grammatical distinctions were spoken.
The same limit appears more sharply with source material. An identifier such as DOI:10.1000/182 sits outside Phi syntax. Exact oral recovery requires an agreed spelling, citation, or display convention appropriate to that artifact.
Practicerun the check
With a partner, have one person read the two framed messages while the other transcribes them. Use a clear natural pace. Compare the lexical forms first and period placement second. Record each difference without treating it as a character test. Then exchange roles with a new message that contains:
- A direct yes-or-no question with
wa - An exact Phi quotation in
sha ... sho - A name marked with
ne - A separate
konaaddress - A wish opened by
su - An interjection chosen for its own meaning
- A discourse marker opening a related sentence
Working alone, record a message and return after its exact wording is no longer fresh. Transcribe it without looking at the source, then make the same two comparisons. An outside identifier can be tried separately with an agreed oral convention; it is a useful reminder that no grammar turns arbitrary notation into speech by itself.
05The far room
A third round tests the claim where it earns its keep: under rough conditions. Have the reader stand in the next room, or face away across a courtyard, and read the message once at a natural pace. The transcriber writes what arrives. The interesting result is not the score; it is which distinctions survive the distance. A frame boundary, a ne before a name, an opening wa: each was designed to be carried by a word rather than a mark, and the far room is where that design either holds or does not. Record what was lost as evidence about the conditions, not about the language or the listener, and bring the message closer until it survives whole.