Hello and thank you for your interest in this project. Any comments or feedback would be appreciated.

What is it, what does it aim to be? I'll answer generally first, and then in detail, after reviewing objections.

Desiderata

What we might have, and certainly aimed for, is a self-contained, self-study instrument able to provide basic (i.e., phonetic) literacy independent of the language of the learner and with little or no dependency on tech or teacher.

Printed paper is self-contained. The Delhi print shop charged me $0.03/page, and once in hand, a recipient doesn't need smart phone or computer, electricity, chargers, blackboards, government-paid teacher employees, or anything but time, interest, and light.

Therefore I say: print copies, put them in the hands of the unlettered, and let them go at it.

The need, then, is for black and white line drawings which can survive the xerox copy machine. With a QR code to bring smart phone users like those who own copiers to access the original clean source through the internet, it doesn't have to survive many generations of copying. But the instrument itself will come out of a copier, in the form which users receive it, and that must be good enough. So: clear black and white line drawings are the target format.

What shall they communicate? The unguessable, arbitrary, and conventional correspondence between sound and squiggle that a script represents. This is the key to the kingdom. Which sounds, which squiggles?

Among sounds, we want to uniquely and unambiguously distinguish all or at least most of the linguistically-distinctive phonetic features. These include laryngeal states, vocalic features, consonantal place and manner, and for tone languages, tone. Among all the scripts in the world, any choice might work, but the International Phonetic Alphabet was designed for just this purpose, to symbolize all the linguistically distinctive sounds in human language. So the IPA will do.

Notice: Orthographies are not the target. In each detail where they differ from phonetic scripts, they cannot be taught by phonetic charts. Hence this is not about English spelling, or Cambodian. You can learn those using IPA, but orthography is a separate level and if you seek to achieve the possible then you won't start with orthography.

Nor can scripts with phonetic clusters such as Devanagari "ढ" = [ ɖ h ə ] be the target; a phonetic script can easily write clusters as sequences of simple sounds each of which is capable of being specified using a phonetic chart, in this case. But the other way around would be impenetrable. So three IPA symbols in sequence represent a cluster symbol very nicely, and each was teachable with a segmental phonetic chart; whereas the chart for even a minimal CV sequence would be many-dimensional, thus practically impossible to learn from. So:

Learn IPA first!

(And then use IPA, as in writing sequences of sounds in a cluster. With IPA you can learn anything, with definiteness and clarity, because the IPA is defined in terms of simple actual sounds. IPA First.)

Some have objected, saying, Hey, No, instead just start small, just do one language, customize this for Hindi or something, make it work for one before you aim for all. I reply, No, you miss the point. Solving the problem for one language means you specifically prevent yourself from solving the problem universally. Let's keep the focus for a time on a universal instrument, which could give something to everyone. With that we can give something to everyone, and I want to not aim to leave 99% of everyone behind, if possible, and it seems possible. After that, we can put more effort into individual languages and have a Hindi-specific version and versions for different specific local languages. But let's light the world-wide fire first.

Oh that's just the first objection of many.

Objections

Often highly literate subjects balk on being asked to do anything physical with their actual mouth and their own physical sound-making apparatus. I take it that the practice of literacy is largely about forgetting the articulatory basis of one's sound system, and remaining conscious only of the set of distinctions, and able only to hear the sounds, and to see the forms of the letters, while direct proprioceptive access to their own mouth movements has become much more difficult or repressed. Certainly in public or in front of a stranger, there is so much fear of making mistakes, entering into a domain highly unpracticed for people who are so highly practiced in an adjacent cognitive space, it exposes painful vulnerability. Illiterates, on the other hand, don't have that acquired ignorance and acquired incapacity. Still the teacher may need to encourage a student to vocalize aloud fearlessly.

Some have objected: this should be an app on the internet. These students just need a teacher. This requires audio visual support! We should be providing accessory teaching resources for actual teachers, not this impossible and demoralizing context-less task for unsupported learners to simply destroy their motivation without giving them a functional improvement in their lives. I agree that more is better, and my list of additional accessory elements for learners and teachers is on my list too, though I expect others have provided them already.

Such objectors are not getting the point. More is better, yes, wherever there is more, but what about those without? I think there are a billion have-nots who are living without. The primary, the target recipients for this instrument are illiterate, low-resource people in impoverished and remote corners of the globe. If they do have tech, phones, public schoolteachers, and audiovisual support stuff, then that is great, for them, but then they don't need basic phonetic literacy on a piece of paper. They've got literacy as a whole covered already. They are not our target recipients.

You could say of our target recipients, the world basically until now actually cannot properly reach them, but barely, potentially, yes we can begin to finally reach them, if only by putting something self-contained into their hands: printed pieces of paper. Something that communicates without teachers, without smart phones or computers or tech. It just needs the information to be in there, clearly and unambiguously, so that if they are interested, they can figure it out themselves.

And among the innumerable steps of the literacy ladder, everyone already knows their own language in its full glory. They know its words. They know its grammar. They know its sounds. We don't have to teach language to humans: They have language! What they don't have is the arbitrary, conventional, and generally quite impenetrable sound-symbol correspondence. Even for the most phonetic of scripts, Korean hangul, the phonetic meaning of this or that squiggle MUST be explained by a teacher in direct audio-visual contact with the student. Without that, it is impossible; people are helpless to cross this fundamental obstacle, and because of this impossibility, they are functionally outside modern society, and perhaps even functionally enslaved by the literates around them.

So the crucial essential problem is, can a line drawing communicate sufficient phonetic detail, enough to uniquely identify the distinctive sounds of language?

I say Yes. And here to prove it are the details. Of course you can just look at the drawings and satisfy yourself, but some guided commentary might not be inappropriate.

Structure and Content

For the vowels: line-drawn features include as follows:

• Veatch 1998, English Vowels, found that each dialect has a reduction target, a vowel sound toward which stressed vowels shift in acoustic space when they are less stressed. The reduction target is central and unrounded but may be high or mid depending on the dialect. However the generalization that vowels are to be understood as deviations larger or smaller from a central form, which could be understood as the default posture, or equivalently as the unmarked vowel quality, seemed evident in all four dialects studied.

  Hence the drawing for [schwa] is shown as the equal-cross-sectional-area vowel tract configuration, and then all the non-schwa vowels can be understood as deviations in one direction or another from it.

• The first, easiest-to-demonstrate, vowel [a] is symbolized by a single down arrow on the jaw or tongue, plus the [+voice] feature. Effective and sufficient verbal instructions are: "Open your jaw. (pointing at the jaw arrow drawing.) Make a sound (pointing at the [+] glyph)." The subject generally does indeed make an [a] sound; I have not heard an [ae] nor an [openo]. Rarely I get [schwa], but that will be distinguished by its own separate glyph.

• The vowel ordering is [a u i o e] because after [a] is the most opposite and obvious: [u]. Closing the mouth nearly, rounding the lips, raising the tongue body to the back, all these opposites produce a clear and distinct [u], and having this second provides the learner a sense of success with a second example without any troubles.

• The third corner of the vowel triangle is a little harder than [a] and [u] because of the less-obvious physical meaning of an upper-lip retraction arrow, one arrow on the upper lip. Alternative conceptualizations like lip spreading or lip corner retraction cannot be drawn with a simple arrow line on a midsagittal head outline since the corners of the lips are off the center line and arrows where they ought to be symbolized would be quite confusing as they would overlap centerline parts: the teeth and tongue tip and open regions of the mouth. Fortunately, upper-lip retraction seems to be performed correctly by users, and an adequate (clearly distinctive and perceptually identifiable as "same sounds") approximation of lip spreading does actually result, from what I've seen so far. So I have decided for now, no need for extra arrows, nor an additional head drawing such as a smiley face, considering that one arrow on the upper lip successfully communicates the lip posture of unrounded vowels.

• Identifying the jaw/tongue downward arrow as a graphical form of a phonological feature, it would be a privative feature [open]. (This is to take a similar but opposite approach compared with Veatch 1991, Chapter 3, Privative Features) Privative [open] can cooccur with all the other features or configurations, and where it does, [i] becomes [e] and [u] becomes [o]. If the language of these line drawings is considered as a phonological theory, it asserts that the [open] feature can be drawn with or without the two mutually contradicting configuration features [high-front-unrounded] and [high-back-rounded]. This view yields all combinations. Of course, phonologists might not think graphical representations are the underlying form of distinctive features, but please consider how very distinctive and how freely combining they are when taken this way. When all possible combinations are indeed found, that suggests you've got the right analysis.

• A few other points. Among places of articulation, what we might call "dental" is actually a cover glyph for interdental, dental, and alveolar, and what we might call "labial" is a cover glyph for labiodental and labial. Finally uvular and velar are not distinguished. The tradeoffs here are made for the sake of space on the page; we get filled columns and rows this way, and save space. If constrained to two sides of a single sheet of paper in order to minimize global project cost to an attainable number, we don't have spare space.

	Your thoughts?
Name+:
Email*:	(will not be shared or abused)
Comment:
	Feedback is welcome.