The now deprecated scheme of March 2006

In March of 2006 I revised an earlier scheme whereby we could us just 16 letters (or symbols) so that each could represent a complete CV syllable. When thinking about this, I recalled;

1. an idea Dirk Elzinga had suggested way back in May 1999. I quote a snippet from that mail:
   '... where each syllable has the same consonant but varies in the vowel quality. The vowel qualities are "resting" (central unrounded), rounded, and fronted.
   p [p|, pu, pi] b [pa, po, pe] ("|" is "barred-i")
   t [t|, tu, ti] d [ta, to, te]'
2. an idea that R. Srikanth had proposed at about the same time
   '.... It struck me that there is an equally simple system to exhaust all phonotactically allowed open syllables by requiring pairs of letters to jointly determine the two syllables they will represent (for example):
   sk /suki/
   sg /segi/
   tk /tuka/
   tg /tega/
   etc.'
3. J.R.R. Tolkien's Fëanorian letters where "the primary letters are arranged in four témar (series), each of which had six tyellar (grades)."

I still have a full copy of Dirk's proposal, but I do not have details of how R. Srikanth's system worked. However, the system I have come up with is described below.

 

The Phonology

Vowels There are four, thus:   Front (unrounded) Back (rounded) High /i/ /u/ Low /ɛ/ /ɔ/

Consonants There are eight, including the 'zero consonant', thus:   series #0 series #1 series #2 series #3 sonorant /ɰ/ /l/ /n/ /m/ obstruent /k/ /s/ /t/ /p/

Note:

1. The low vowels may be any vowel from mid to low; thus /ɛ/ may be pronounced as any vowel between [e] and [æ] inclusive, and similarly /ɔ/ may be pronounced as any vowel between [o] and [ɒ] inclusive.
2. /ɰ/ may a velar approximant, as the symbol suggests, or it may be realized as a glide onset [j] before front vowels or [w] before back vowels; or it be entirely silent. Appropriately, it is in series #0 (the reason for numbering from 0 rather than from 1 becomes apparent in the last section below).
3. The obstruents are voicelss when initial, but may become voiced between vowels.
4. The phoneme /l/ may be realized as any dental or alveolar approximant, whether lateral or not, or as a dental/alveolar flap.

The Orthography

The vowels are not written with separate letters. The correct pronunciation is determined by the consonant letters. There are 16 consonant letters, of four grades for each of the four consonant series, thus:

	series #0	series #1	series #2	series #3
grade #0	y	l	n	m	sonorant	followed by a high vowel
grade #1	k	s	t	p	obstruent
grade #2	h	r	ɴ	м	sonorant	followed by a low vowel
grade #3	g	z	d	b	obstruent

Note:

1. Grade #2, series #2: this is the IPA'Latin letter small capital n' (U+0274)*; if this symbol is not available, N may be used instead.
2. Grade #2, series #3: this is the 'Cyrillic small letter em' (U+043C)*; if this symbol is not available, M may be used instead.
3. The alphabet is ordered gradewise (like Tolkien's Tengwar), thus: y l n m k s t p h r ɴ м g z d b

* The language will probably not use separate upper and lower case; but if it does, N and M will be the upper case forms of ɴ and м; the upper case forms of m abd n will be larger versions of the lower case forms, as in the Irish alphabet.
 

Which high or low vowel occurs after the consonant depends upon the vowel harmony of the word (and affixes); all the vowels must be either back vowels throughout or front vowels throught. This is determined by the vowel of first syllable, which is determined by the first two consonants thus:

First consonant	Second consonant	Vowel in first syllable
followed by high vowel	followed by high vowel	/i/
followed by high vowel	followed by low vowel	/u/
followed by low vowel	followed by high vowel	/ɛ/
followed by low vowel	followed by low vowel	/ɔ/

Examples (syllables should be evenly stressed with slight emphasis on the last syllable):

sdrp /sutɔlɔpu/

dspr /tɛsipilɛ/

spdr/sipilɛtɛ/

drps /tɔlɔpusu/

 
 

Mapping the symbols to bits

Jeff Prothero chose an alphabet of 16 letters because he wished that each letter should map uniquely to a four-bit pattern. I also have only 16 letters, so they too can be mapped to four bits if desired. Indeed, the system I proposed in September 2005 did that; but that system started with the bit-patterns and attempted to map them to humanly pronounceable syllables. I now think this is the wrong approach and that one should start with a 'human-friendly' system and then map that to bits (which, of course, is what we did with ASCII and now do with Unicode). I have, therefore, now adopted a different mapping.

• The two most significant bits denote the grade;
• The two least significant bits indicate the series.

The table below shows this (hexadecimal values are also given).

	series #0	series #1	series #2	series #3
grade #0	0000 0	0001 1	0010 2	0011 3	sonorant	followed by a high vowel
grade #1	0100 4	0101 5	0110 6	0111 7	obstruent
grade #2	1000 8	1001 9	1010 A	1011 B	sonorant	followed by a low vowel
grade #3	1100 C	1101 D	1110 E	1111 F	obstruent

Note that:

1. The bit order is exactly the same as the gradewise alphabetic order given above.
2. The most significant bit (MSB) is 0 if the consonant is followed by a high vowel, 1 if followed by a low vowel.
3. The next most significant bit is 0 if the consonant is a sonorant, 1 if it is an obstruent.

Also, it may be noted, the vowel of the initial syllable may be considered as being determined by MSB of the the first two consonants, as shown in the table on the right.

Front Back High 00 01 Low 10 11