Towards a Lang25 Phonology
What follows is an attempt to initiate, on the basis of the progression exemplified in childhood language acquisition and the historic "jargon ~ pidgin ~ vernacular" development, a phonology or sound system for a global core vocabulary of common words. This precedent (or scientifically-validated approach) indicates not only that LangX should proceed in discrete cognitive stages but also that it should be launched initially via a substrate or base language.
The most likely candidate for this substrate might seem to be the world's foremost existing auxiliary language, namely English, and an apt metaphor might be of the plane "LangX" taking off from the deck of the aircraft carrier "Global English" in international waters. However, there would no doubt be political opposition to the idea of using English in this way, and it would probably be too narrow a linguistic focus in any case. Indeed, the very fact that LangX is essentially starting out as a "global pidgin" would indicate that a conflation of historic pidgin substrates is called for. Thus the following provisional Lang25 phonology is loosely based on other major historic substrates, including French, Spanish and Portuguese.
NB Ex-alphabet IPA symbols may not display in Internet Explorer. Any other browser works fine: Firefox is downloadable free.
There are five vowels in Lang25 (as in Spanish, Japanese and other languages) but they should not be closely specified. So long as the word being enunciated is identifiable there should be no rule as to correct pronunciation. A similar latitude should extend to consonants. Thus, if someone (at this initial Lang25 level) fails to differentiate between voiceless and voiced (fortis and lenis) consonants /p/ and /b/, /t/ and /d/, /k/ and /g/, or between /l/ and /r/, or pronounces /w/or /f/ as /v/, or /s/ as /z/, or any other reasonably close approximation, nothing should be made of it so long as the word can be correctly identified. International communication is the need of the hour; anyone requiring phonemic accuracy may easily use another language.
Similarly the IPA values should be regarded as indicative rather than definitive: for instance, anyone should be at liberty to pronounce a voiced alveolar "r" sound as /ɾ/ - a tap or flap, as /r/ - a trill, or as /ɹ/ - an approximant, so long as the word being uttered is recognisable. However, some phones are more accommodative than others, and potential phonetic confusion and overlap would tend to exclude the letter / phoneme /v/ from the core phonology (and vocabulary). This leaves the 25 phonemes (20 consonants and 5 vowels) of Lang25.
As can be seen from the above table, all the consonants apart from [j], [x] and [q] have values familiar through English usage, and [j] and [x] find confirmation in other European languages. The only radical step is the casting of [q] as "plosive [g]". This might tend to be rejected as unprecedented, but the "letter shape" is surely appropriate, and the voiceless uvular plosive [q] is right next to the voiced velar plosive [g] in any case. For better or worse, uvular phonemes have mostly disappeared from popular usage in these Western languages. For instance, the word "queen" is now almost universally pronounced as "kween" rather than with the original uvular plosive, so the symbol [q] has become redundant and might as well be otherwise employed.
There is a fairly close correlation between these and the 20 commonest consonant phonemes identified by UPSID - the phonological inventory of 317 languages, each one representative of a different recognised language family grouping, first published by researchers at the University of California in 1984. The only difference is that the UPSID list excludes /ʤ/, /z/ and /ʒ/ (as well as /v/), and includes /ʔ/, /ɲ/ and /ŋ/. The last three are as
This UPSID list is quite representative of ethnic languages, but not on a population pro rata basis - also it omits English, the major creoles and all the constructed langauges, so a new auxiliary language should not necessarily be especially beholden to it. Many languages do contain /ʔ/, /ɲ/ and /ŋ/, but mostly unrepresented by unique symbols in the script. This could be because the glottal stop is often a suprasegmental rather than a true phoneme, /ŋ/ tends to assimilate (as in words such as "bank"), and it isn't always obvious that /ɲ/ is one sound rather than two. The main constructed languages tend to follow the same pattern by excluding symbols for these phonemes, and their phonologies are on the whole quite similar to that suggested for LangX above. Esperanto phonology is much the same except that it has two extra sounds: /ʦ/ (/ts/) voiceless alveolar affricate (as in
German) and /x/ voiceless velar fricative (as in "loch, Bach"). Ido also has /ʦ/, but omits /x/ and /ʤ/. It may be said that Eurocentrism has produced such a bias, but history cannot be discounted, and the preponderating influence of Graeco-Roman culture cannot be denied. It follows, then, that a new global pidgin substrate should emerge from much the same basis.
Notable for their absence from the suggested Lang25 phonology are the common English phonemes /θ/ and /ð/ (/dh/ and /th/) the voiced and voiceless dental fricatives - as in "the" and "thin". These are also normally absent from constructed languages because, although common on the Western European seabord, they are not typical of languages worldwide. However, these consonants are among those likely to be gradually formalised or incorporated in the levels of LangX beyond Lang25. According to the LangX scheme, 7 further consonant phonemes would be formally incorporated, most likely including /ʔ/, /ŋ/, /x/, /θ/ and /ð/. And hopefully by then there will be a new script - so that extra consonants might be symbolised phonemically without the need for diacritics. But all that is for the future; the requirement now is to develop a provisional core vocabulary based on the suggested Lang25 phonology.