Jerome | BPJ 2 | BPJ 2.1 | ||
---|---|---|---|---|
36 | Utter | a | ` | u |
39 | Foot | U | u * | V * |
36.2 | Up | u | V * | C * |
naming-dot | / | ^ | ` * | |
half-Ye | ^ | C * | ^ |
The main innovation in my proposal lies in the mapping of vowel letters, where I've tried to provide memory cues which work in terms of the inner structure of Quikscript itself: I wanted my mapping to make sense primarily in terms of the observed systematic relations between Quikscript letter shapes and values as mapped by Read, and secondarily by recurring relations between Quikscript and Traditional Orthography spellings. This mapping is meant to be a transcription of Quikscript rather than any phonemic or phonetic transcription system, which may be 'external' to Quikscript. When using this mapping one will have to think in terms of Quikscript letter shapes and letter relations. This is meant to be an advantage in that one will 'think Quikscript' when typing in this transcription, which will hopefully be of help both to those who are used to write (as opposed to type) in Quikscript in learning to type with this mapping, and to those who are going to learn to write and type Quikscript simultaneously.
(mappings with a following * differ from Jerome kmap, and the half-letters with an * after their name are not found in the Quikscript Manual.)
QS # | (Jerome) | Name | BPJ 1 |
---|---|---|---|
#1 | p (p) |
Pea | p |
#2 | b (b) |
Bay | b |
#3 | t (t) |
Tea | t |
#4 | d (d) |
Day | d |
#5 | k (k) |
Key | k |
#6 | g (g) |
Gay | g |
#7 | T (T) |
Thaw | T |
#8 | H (H) |
They | D * |
#9 | f (f) |
Fee | f |
#10 | v (v) |
Vie | v |
#11 | s (s) |
See | s |
#12 | z (z) |
Zoo | z |
#13 | S (S) |
She | S |
#14 | Z (Z) |
J'ai | Z |
#15 | c (c) |
Cheer | c |
#16 | J (J) |
Jay | J |
#17 | j (j) |
Ye | j |
#18 | w (w) |
Way | w |
#19 | h (h) |
He | h |
#20 | ( ) |
Why | W * |
#21 | N (N) |
Ing | G * |
#22 | m (m) |
May | m |
#23 | n (n) |
No | n |
#24 | l (l) |
Low | l |
#25 | r (r) |
Roe | r |
#26 | i (i) |
It | i |
#27 | I (I) |
Eat | I |
#28 | e (e) |
Et | e |
#29 | E (E) |
Eight | E |
#30 | A (A) |
At | a * |
#31 | F (F) |
I | A * |
#32 | y (y) |
Ah | q * |
#33 | Y (Y) |
Awe | Q * |
#34 | o (o) |
Ox | o |
#35 | q (q) |
Oy | O * |
#36 | a (a) |
Utter | y * |
#37 | Q (Q) |
Out | Y * |
#38 | O (O) |
Owe | B * |
#39 | U (U) |
Foot | u * |
#40 | M (M) |
Ooze | U * |
#41 | ç (ç) |
Loch | K * |
#42 | æ (æ) |
Llan | L * |
#43 | è (è) |
Excite | x * |
#44 | é (é) |
Exam | X * |
#23.2 | G (G) |
Noon | N * |
#36.2 | u (u) |
Up | V * |
- | / (/) |
naming dot | ` * |
- | left bracket | ( | |
- | right bracket | ) | |
#1.5 | ] (]) |
half-Pea | P * |
#3.5 | [ ([) |
half-Tea | F * |
#4.5 | B (B) |
half-Day | R * |
#12.5 | L (L) |
half-Zoo | C * |
#17.5 | ^ (^) |
half-Ye * | ^ |
#18.5 | @ (@) |
half-Way | M * |
#19.5 | \ (\) |
half-He | H * |
#20.5 | ~ (~) |
half-Why * | ~ |
Of the fifteen Quikscript vowel letters fourteen form pairs of related shapes, and Read clearly endeavored to map related sounds to the letters within each pair. The six 'lax' or 'checked' vowel1 phonemes of Read's conservative RP English2 were each mapped a letter within one of these pairs — the one without an extra loop in in the four cases where the difference between the paired letters consists of such a loop —, while the other letter of each pair has been mapped to a similar-sounding 'tense' vowel or diphthong. Five of the six RP 'lax' vowel phonemes are readily associated with five of the six vowel letters of Traditional Orthography, and thus the Quikscript letters for those phonemes have been mapped to the lowercase forms of their associated Traditional Orthography letters: #26 = {i}
, #28 = {e}
, #30 = {a}
, #34 = {o}
, #39 = {u}
, and the sixth RP 'lax' vowel phoneme represented by Quikscript letter #36 — which is paired with a 'looped' letter #37 — has been mapped to the sixth Traditional Orthography vowel letter {y}
.
Next we map the Quikscript letters which are paired with the Quikscript letters for the 'lax' vowels to the corresponding Traditional Orthography uppercase vowel letters. Some of these mappings make no 'phonetic' sense; rather one has to remember the mapping of lowercase Traditional Orthography vowel letters to Quikscript letters, and then the rule that the Quikscript letters paired with these letters are mapped to the corresponding Traditional Orthography uppercase letters: #27 = {I}
, #29 = {E}
, #31 = {I}
, #35 = {O}
, #37 = {Y}
, #40 = {U}
.
We then come to the seventh Quikscript vowel letter pair #32-#33. They are problematic, since there are no Traditional Orthography vowel letters readily available for them. I choose to map them to the one Traditional Orthography upper-lowercase letter pair not readily associated with any English phoneme, namely {Q}
-{q}
. This does not mean that there are no memory cues available for these mappings: #33 can be seen as a shape variation of #34, and the Traditional Orthography uppercase Q is of course visually similar to a Traditional Orthography uppercase O, and, likewise #32 can be seen as a shape variation of #30, and Traditional Orthography lowercase q is — admittedly vaguely — similar to a Traditional Orthography lowercase script a.
Last we come the unpaired Quikscript vowel letter #38. It's sound is usually associated with the Traditional Orthography letter o, or one of the digraphs oa, ou or ow which all contain o, but due to my choosen system for maping the 'paired' Quikscript vowel letters to ASCII letters {o}
and {O}
are already occupied by Quikscript letters #34-#35. #38 must therefore be mapped to the otherwise unused Traditional Orthography uppercase letter B, which at least allows us to use the words "bowl" and "bow" (the noun), which recall the shape of the letter in the Quikscript Manual, as memory cues. I readily admit, however that this is the weakest point of my proposal: if one finds the system for maping the other vowels an advantage in favor of my proposal one will simply have to take this weakness in the bargain! It was tempting to map #38 to the ASCII @
symbol, but I preferred to map all the Quikscript letters which are sanctioned by the Quikscript Manual to Traditional Orthography upper- or lowercase letters leaving the ASCII symbol characters alone, and the namer dot to the 'backtick' {\
}` which is superfluous in normal typing, pace TeX.
There are few innovations among the primary consonants, since the majority of Quikscript consonant letters have an obvious one-to-one counterpart in Traditional Orthography. Most of the remaining Quikscript consonant letters have been mapped them to the uppercase version of the letters for the most similar sounds already mapped to a lowercase letter, and I think few will argue against the mappings #7 = {T}
, #8 = {D}
, #13 = {S}
, #14 = {Z}
, nor with the mapping of #15 to lowercase {c}
, where the added complication of having to press the shift key seemed unwarranted, despite the apparent regularity of uppercase corresponding to an added h in Traditional Orthography. One might have preferred mapping #13 to ASCII uppercase {C}
and #14 to ASCII uppercase {J}
, but the system of vowel mappings has already preempted the pair {y}
-{Y}
for use as vowels, so that ASCII lowercase {j}
had to be mapped to #17, making it necessary to map #16 to ASCII uppercase {J}
, in spite of this being a step away from native speakers' and writer's intuitions about Traditional Orthography letter-to-sound correspondances. At least it agrees with the old Quikscript font mapping.
The remapping — in fact swapping of the mappings — of {G}
and {N}
may be a cause for concern for those used to 'traditional' Quikscript fonts, but I think it is warranted: the Traditional Orthography spelling ng at least contains a g, while words like now, knack, knife, need — the the Quikscript Manual examples of initial alternate #23 — don't, making transcriptions like {loGiG, NY, Nak, NAf, NId}
more readibly recognizable than {loNiN, GY, Gak, GAf, GId}
. I'm prepared to back down on this point, however.
I have been considering a different arrangement which would allow the mapping of #17 to ASCII lowercase {y}
and other differences which would allow memory cues which are perhaps more intuitive in terms of native speakers' and writers' expectations of Traditional Orthography sound-to-letter correspondences, including the well-nigh universal inclination to map #36 to lowercase {u}
.
These changes do in fact amount to a different proposal, since it no longer includes the 'Quikscript-internal' 'paired' vowel letter mappings. People may find it attractive in spite of the fact that it will constitute a compromise between an ASCII transliteration of Quikscript and yet another quasi-phonemic ASCII transcription of English.
QS # | (Jerome) | Name | BPJ 1 | BPJ 2.1 |
---|---|---|---|---|
#14 | Z (Z) |
J'ai | Z | J * |
#16 | J (J) |
Jay | J | j * |
#17 | j (j) |
Ye | j | y * |
#31 | F (F) |
I | A * | Y * |
#32 | y (y) |
Ah | q * | A * |
#35 | q (q) |
Oy | O * | q |
#36 | a (a) |
Utter | y * | u * |
#37 | Q (Q) |
Out | Y * | B * |
#39 | U (U) |
Foot | u * | V * |
#38 | O (O) |
Owe | B * | O |
#36.2 | u (u) |
Up | V * | C * |
#12.5 | L (L) |
half-Zoo | C * | Z * |
Any mapping of Quikscript letters to ASCII letters will in effect be a transliteration, and as such it must be easy to read even if displayed in a normal ASCII font rather than a Quikscript font, because there will be times, more or less frequent, when we will be reading transcribed text without having the time or opportunity to view it in a Quikscript font.
The more I think of it the more convinced I get that in an ASCII mapping of Quikscript the ASCII lowercase letters should as far as possible be mapped to their regular Traditional Orthography sounds - say their sounds as they would say when teaching spelling in school - while the ASCII uppercase letters should be used to signal their own otherness.
For that reason the five lowercase vowel letters i, e, a, o, u should be mapped to #26, #28, #30, #34, #36, j and y to #16 and #17 and so on for all the Traditional Orthography single letters which have a default phoneme correspondence.
Once the system of marking Quikscript paired vowels with capitalization has been abandoned by the remapping of {u}
and {y}
there is no point in keeping ASCII {A, O, Y}
for #31, #35, #37; indeed these mappings become more misleading than the old uppercase {F}
for #31, since people will expect them to be mapped to vowels, but these sounds are never written with these Traditional Orthography letters, unlike I and E which at least map to the sounds of #27 and #29 in recent french loans like machine and ballet. It becomes better to use uppercase {Y}
for #31, since this letter and sound are associated in Traditional Orthography (my, rye, edify, hyphen). It is also obviously better to map ASCII uppercase {A}
to #32, since this sound is always written a in accents like RP which maintain the three-way contrast between #32, #33, #34. In the same vein ASCII uppercase {O}
can be conveniently mapped to #38, which is 'the long sound of o' in Traditional Orthography. For #33 we will map uppercase {Q}
also in this mapping, if onlu because we will need to map Q to some Quikscript letter, and at least uppercase {Q}
looks like uppercase {O}
with a squiggle, just like #33 looks like #34 with a squiggle!
This will leave ASCII {q}
as the odd man out among the ASCII lowercase letters; I have mapped it to #35, not only because it is the old mapping, but also because lowercase q does have some similarity, to the Quikscript letter itself — a loop with a downwards tail to the right — as well as to the Traditional Orthography digraph oy.
With lowercase {u}
mapped to #36 some other ASCII letter must be found for #39, and after some consideration I have come to the conclusion that uppercase {V}
, as proposed by Ted Larson freeman, is indeed the best choice: it looks similar to the Quikscript letter, and historically the Traditional Orthography letters U and V3 were variant forms of the same letter.
ASCII uppercase {B}
will still be mapped to a vowel, and now memory-cued to the word "bout", and probably those who are familiar with the ȣ ligature form of the ou digraph will be aided by its visual similarity to uppercase B.
I choose to map ASCII uppercase {C}
to alternate #36, since Ted Larson Freeman brought the graphic similarity between these letters of the two alphabets to attention, and with {V}
mapped to #39 no other of the available uppercase letter fits better. That uppercase {C}
is more readable than lowercase in this role, especially next to {k}
, because capitalization serves to signal that C does not have any of the usual sound values of Traditional Orthography c, nor, indeed, marks a vowel is easily seen by comparing the following:
{C} =#36, {c} =#15 |
{C} =#15, {c} =#36 |
|
---|---|---|
{skClpcur} |
{skclpCur} |
sculpture |
{akCmpliS} |
{akcmpliS} |
accomplish |
{surkCs} |
{surkcs} |
circus |
{rukCs} |
{rukcs} |
ruckus |
{kClcurCl} |
{kclCurcl} |
cultural |
{kCstCm}/{kCsFCm} |
{kcstcm}/{kcsFcm} |
custom |
There are, not counting derivations, some 30 words where #5 and #15 occur next to each other4, but I deem these to be of far less text frequency than in particular the combinations #36+#24 and #36+#22, espescially if one considers that several of these words have alternative pronunciations with #3+#17 instead of #15, and thus should perhaps be spelled accordingly.
The remaining Quikscript letters must and should be mapped to uppercase ASCII letters. The Traditional Orthography equivalents of these Quikscript letters will normally be digraphs, and as a general principle the uppercase ASCII character choosen to represent them should correspond to one of the letters of these digraph. There will, even must, be deviations from this principle, but these should be well grounded and provided with good memory-cues. As it turns out only T, S, W for #7, #13, #20 will more or less unproblematically follow this principle. Analogy then suggests that D, C, Z, N be mapped to #8, #14, #15, #21, but again it turns out that only {D}
#8 will stand relatively unaffected by other considerations. Traditional Orthography never writes the sound of #8 with a spelling containing the letter d, but the analogy to T for #7 is relatively flawless here. Not so with the sound of #14: spellings with z corresponding to this sound are virtually confined to the use of zh in transcribed Russian names. The Traditional Orthography consonant letter most regularly associated with this sound is actually s, but there is another Traditional Orthography letter which is fairly regularly associated with the sound, namely j in recent French loan words. This suggests uppercase {J}
as the best mapping for #14, as does the fact that words like garage and mirage have variant pronunciations with #16. There is also the fact to consider that uppercase ASCII letters which don't get mapped to basic or alternate letters will almost inevitably end up mapped to half-letters, whether we include those in our 'official' proposals or not, and it seems much better to map the ASCII uppercase {Z}
to the frequently used half-z and ASCII uppercase {J}
to #14 than the other way around!
With #21 the problem is somewhat different. What causes problems here is that ASCII uppercase {N}
is clearly the best, not to say the only reasonable, mapping for the alternate #23, and thus ASCII uppercase {G}
emerges as the best mapping for #21 as discussed previously.
Similar considerations apply to #15. From a practical point of view ASCII uppercase and lowercase {C}
-{c}
are equally suitable for #15 once it is understood that #5 is ASCII {k}
and #11 is ASCII {s}
, but since any ASCII letter not mapped to a basic Quikscript letter will have to be mapped to a sound not normally associated with with the Traditional Orthography letter c it is better to let the signal value of capitalization come into play with this other mapping, since lowercase c is at least part of the default Traditional Orthography spelling for #15.
(Again mappings with a following * differ from Jerome kmap, and the half-letters with an * after their name are not found in the Quikscript Manual.)
QS # | (Jerome) | Name | BPJ 2 |
---|---|---|---|
#1 | p (p) |
Pea | p |
#2 | b (b) |
Bay | b |
#3 | t (t) |
Tea | t |
#4 | d (d) |
Day | d |
#5 | k (k) |
Key | k |
#6 | g (g) |
Gay | g |
#7 | T (T) |
Thaw | T |
#8 | H (H) |
They | D * |
#9 | f (f) |
Fee | f |
#10 | v (v) |
Vie | v |
#11 | s (s) |
See | s |
#12 | z (z) |
Zoo | z |
#13 | S (S) |
She | S |
#14 | Z (Z) |
J'ai | J * |
#15 | c (c) |
Cheer | c |
#16 | J (J) |
Jay | j * |
#17 | j (j) |
Ye | y * |
#18 | w (w) |
Way | w |
#19 | h (h) |
He | h |
#20 | ( ) |
Why | W * |
#21 | N (N) |
Ing | G * |
#22 | m (m) |
May | m |
#23 | n (n) |
No | n |
#24 | l (l) |
Low | l |
#25 | r (r) |
Roe | r |
#26 | i (i) |
It | i |
#27 | I (I) |
Eat | I |
#28 | e (e) |
Et | e |
#29 | E (E) |
Eight | E |
#30 | A (A) |
At | a * |
#31 | F (F) |
I | Y * |
#32 | y (y) |
Ah | A * |
#33 | Y (Y) |
Awe | Q * |
#34 | o (o) |
Ox | o |
#35 | q (q) |
Oy | q * |
#36 | a (a) |
Utter | u * |
#37 | Q (Q) |
Out | B * |
#38 | O (O) |
Owe | O |
#39 | U (U) |
Foot | V * |
#40 | M (M) |
Ooze | U * |
#41 | ç (ç) |
Loch | K * |
#42 | æ (æ) |
Llan | L * |
#43 | è (è) |
Excite | x * |
#44 | é (é) |
Exam | X * |
#23.2 | G (G) |
Noon | N * |
#36.2 | u (u) |
Up | C * |
- | / (/) |
naming dot | ` * |
- | left bracket | ( | |
- | right bracket | ) | |
#1.5 | ] (]) |
half-Pea | P * |
#3.5 | [ ([) |
half-Tea | F * |
#4.5 | B (B) |
half-Day | R * |
#12.5 | L (L) |
half-Zoo | Z * |
#17.5 | ^ (^) |
half-Ye * | ^ |
#18.5 | @ (@) |
half-Way | M * |
#19.5 | \ (\) |
half-He | H * |
#20.5 | ~ (~) |
half-Why * | ~ |
19.12.2007 Benct Philip Jonsson bpj@melroch.se
In phonetics and phonology, checked vowels are those that usually must be followed by a consonant in a stressed syllable, while free vowels are those that may stand in a stressed open syllable with no following consonant.
Received Pronunciation (RP) is a form of pronunciation of the English language which has been long perceived as uniquely prestigious amongst British accents.
Originally V—u constituted an upper-lowercase pair. The split into two letters was invented in the 16th century and completed in the 18th.
actual, actuary, actuate, architecture, beachcomber, beechcraft, chukchi, conjecture, contractual, effectuate, factual, fluctuate, fracture, hutchcraft, inchcape, ineffectual, infrastructure, intellectual, juncture, latchkey, lecture, manufacture, picture, prefecture, punctuate, puncture, sanctuary, stricture, structure, tincture, witchcraft.
Source: http://www.speech.cs.cmu.edu/cgi-bin/cmudict
See http://en.wikipedia.org/wiki/CMU_Pronouncing_Dictionary ↩