Transcription: Difference between revisions

From The Toaq Wiki
(formatting)
(why eqti when u can emti)
 
(7 intermediate revisions by 3 users not shown)
Line 1: Line 1:
Below is a list of techniques one can use to transcribe foreign words into Toaq's phonology.
'''Transcription''' is the act of representing the sound of a foreign name in Toaq, given its restrictive rules of what constitutes a valid syllable or not. (Name phrases headed with the name marker {{t|mı}} or the quote marker(s) {{t|shu}} and {{t|mo}} should themselves contain valid Toaq.) For ad-hoc, on-the-fly, one-use borrowings, more meticulous-sounding transcriptions may be more appropriate (see first section); for new coinings/borrowings, however, you should take artistic licence in collapsing the foreign word into a more Toaq-friendly form (see second section).


* Render approximate onsets /w/ as {{T|ba}}.
Examples on this page use English as the source language, but the general process is language-agnostic.
* Render approximate onsets /j/ as {{T|j}}.
 
* Render "n + consonant" pairings as a syllable-final {{T|q}} on the previous word: "pans" -> {{T|paqsy}}.
== Conforming to Toaq phonotactics ==
* Render "t + s" pairings as {{T|c}} and "d + z" pairings as {{T|z}}: "rights" -> {{T|raıcy}} and "rides" -> {{T||raızy}}. (This may seem like an obvious rule, but it can be easy to forget that these particular "consonant clusters" are allowed in Toaq's morphology.)
There are a few important restrictions which need to be sidestepped when transcribing:
* There may be at most one consonant in front of any vowel, and at most a {{t|q}} following any vowel.
With this in mind, the primary way of “inflating” a foreign word so that it conforms to Toaq’s phonotactics is by breaking up consonant clusters with a vowel that is the same as a neighbouring one and nasals to either {{t|-m}} or {{t|-q}}
{|
| ''past'' || → || {{t|pasata}}
|-
| ''rumble'' || → || {{t|raqbala}}
|-
| ''cardamom'' || → || {{t|karadamoq}}
|-
| ''dragon'' || → || {{t|daragoq}}
|}
 
== Reducing consonant clusters ==
Depending on the situation, it may be appropriate to remove a sound or two from the word you’re transcribing to avoid long strings of repetitive-sounding syllables (''scrumptious'' → {{t|sakaraqpachasa}} anybody??). This approach involved identifying the less important sound out of a two in a consonant cluster; usual suspects include sibilants like {{t|s}} and liquids like {{t|l r}}:
{|
| ''stooge'' || → || {{t|tuju}} || (/st/ → /t/)
|-
| ''transpose'' || → || {{t|raqposo}} || (/tɹ/ → /ɹ/ → {{t|r}}; /nsp/ → /np/ → {{t|-qp-}})
|-
| ''smile'' || → || {{t|maılı}} || (/sm/ → /m/)
|-
| ''cardamon'' || → || {{t|kadamom}} || (/ɹd/ → /d/)
|-
| ''scrumptious'' || → || {{t|saraqchasa}}
|}
Also pay attention to soundalikes and actual pronunciations of words rather than their spellings, which may aid in producing a nicer-sounding form:
{|
| ''rumble'' || → || {{t|raqbo}} || (final syllabic /l̩/ sound often realized as [o] or something similar in some dialects of English)
|-
| ''empty'' || → || {{t|emtı}} || (/mpt/ → /mt/ reduction already present in spoken English)
|}
There are two things to keep in mind when following this approach:
* You are very likely to run into existing Toaq words. Such a connotation may sometimes be undesirable. (Note that if you’re attempting [[wikipedia:phono-semantic matching|phono-semantic matching]], then this technique may give you bountiful results.)
* Intelligibility suffers. If you’re trying to coin a word via a borrowing, then this is perhaps the appropriate thing to do, but if you’re invoking the name of a person or thing on the fly, you may not be understood. Exercise due taste.
 
For consonant clusters ending in /s/ or /z/ specifically, it is often enough to turn them into {{t|s}} or {{t|c}} or {{t|z}}:
{|
| ''capsaicin'' || → || {{t|kaseısıq, kaceısıq}} || (/ps/ → /s, ts/)
|-
| ''axolotl'' || → || {{t|asoloty, acoloty}} || (/ks/ → /s, ts/)
|}
Similar treatment could be applied to consonant clusters ''beginning'' in /s/ or /z/, such as /st/, /sk/ (''past'' → {{t|?paca}}); however, this sort of transformation is highly dependent on your sense of aesthetics and doesn’t read very clearly. It may be acceptable in borrowings and coinages.
 
== Handling foreign sounds ==
* /w/ and /j/ can be turned into {{t|ꝡ}}. In past versions, they were turned either into {{t|b}} and {{t|j}} respectively or dropped.
* /ts/ and /dz/ should always be {{t|cy, zy}}, never {{t|tysy, dyzy}} (e.g., ''rights'', ''rides'' → {{t|raıcy, raızy}} not {{t|*raıtysy, *raıdyzy}}). /z/ can stay as {{t|z}}, even though the two are not the same.
* /θ/ and /ð/ (as in English ''thought'', ''thigh'') may be transformed into {{t|s/z, t/d, f/–, p/b}}. There is no consensus about which transcription is the superior one, but some may be more appropriate for some source languages than others (e.g., Spanish [θ] is really a variation of /s/, so transcribe it as {{t|s}}).

Latest revision as of 06:43, 28 December 2022

Transcription is the act of representing the sound of a foreign name in Toaq, given its restrictive rules of what constitutes a valid syllable or not. (Name phrases headed with the name marker or the quote marker(s) shu and mo should themselves contain valid Toaq.) For ad-hoc, on-the-fly, one-use borrowings, more meticulous-sounding transcriptions may be more appropriate (see first section); for new coinings/borrowings, however, you should take artistic licence in collapsing the foreign word into a more Toaq-friendly form (see second section).

Examples on this page use English as the source language, but the general process is language-agnostic.

Conforming to Toaq phonotactics

There are a few important restrictions which need to be sidestepped when transcribing:

  • There may be at most one consonant in front of any vowel, and at most a q following any vowel.

With this in mind, the primary way of “inflating” a foreign word so that it conforms to Toaq’s phonotactics is by breaking up consonant clusters with a vowel that is the same as a neighbouring one and nasals to either -m or -q

past pasata
rumble raqbala
cardamom karadamoq
dragon daragoq

Reducing consonant clusters

Depending on the situation, it may be appropriate to remove a sound or two from the word you’re transcribing to avoid long strings of repetitive-sounding syllables (scrumptioussakaraqpachasa anybody??). This approach involved identifying the less important sound out of a two in a consonant cluster; usual suspects include sibilants like s and liquids like l r:

stooge tuju (/st/ → /t/)
transpose raqposo (/tɹ/ → /ɹ/ → r; /nsp/ → /np/ → -qp-)
smile maılı (/sm/ → /m/)
cardamon kadamom (/ɹd/ → /d/)
scrumptious saraqchasa

Also pay attention to soundalikes and actual pronunciations of words rather than their spellings, which may aid in producing a nicer-sounding form:

rumble raqbo (final syllabic /l̩/ sound often realized as [o] or something similar in some dialects of English)
empty emtı (/mpt/ → /mt/ reduction already present in spoken English)

There are two things to keep in mind when following this approach:

  • You are very likely to run into existing Toaq words. Such a connotation may sometimes be undesirable. (Note that if you’re attempting phono-semantic matching, then this technique may give you bountiful results.)
  • Intelligibility suffers. If you’re trying to coin a word via a borrowing, then this is perhaps the appropriate thing to do, but if you’re invoking the name of a person or thing on the fly, you may not be understood. Exercise due taste.

For consonant clusters ending in /s/ or /z/ specifically, it is often enough to turn them into s or c or z:

capsaicin kaseısıq, kaceısıq (/ps/ → /s, ts/)
axolotl asoloty, acoloty (/ks/ → /s, ts/)

Similar treatment could be applied to consonant clusters beginning in /s/ or /z/, such as /st/, /sk/ (past?paca); however, this sort of transformation is highly dependent on your sense of aesthetics and doesn’t read very clearly. It may be acceptable in borrowings and coinages.

Handling foreign sounds

  • /w/ and /j/ can be turned into . In past versions, they were turned either into b and j respectively or dropped.
  • /ts/ and /dz/ should always be cy, zy, never tysy, dyzy (e.g., rights, ridesraıcy, raızy not
  • raıtysy, *raıdyzy). /z/ can stay as z, even though the two are not the same.
  • /θ/ and /ð/ (as in English thought, thigh) may be transformed into s/z, t/d, f/–, p/b. There is no consensus about which transcription is the superior one, but some may be more appropriate for some source languages than others (e.g., Spanish [θ] is really a variation of /s/, so transcribe it as s).