alphabetic-change

wholesale shift of phonetic sign interpretation

Ideographic logographic and syllabic sign systems such as Chinese and Cuneiform often have polyvalent readings
𒆳𒁁𒌅 kur-mit-tu® Schmetterling
𒆳𒁁𒌅 kur-BÀD-tu® butterfly

In the case of reduced syllabic systems and (extended)alphabetic systems such as Egyptian, there is much more stability.
However, as we shall see, even the Greek and Roman alphabets are not free of phonetic drift…

Imagine a world where instead of 'h' people say 'h'. Huh?
Point in case, Americans pronounce the letter 'h' as 'hage' whereas British utter 'h' as 'age'.
French people articulate Hamburg as 'amburg. British people have a tendency to drop middle 't's : Bri'ish…

Wide spread homophonic readings. If the whole continent reads one word similarily, how can it have had a different pronunciation 2000 years ago? Other than early sound change at the root, and the very strong influence of the biblical canon one mechanism which surely transformed the whole landscape is ‘phonetic infection’ through alphabetic-change. Say a very letter of the alphabet surely but slowly shifts from P⇨PH/PF⇨F⇨H in the age of written words and liturgies this can (and will and had) catastrophic effects on the stability of languages. The invention of the alphabet thus acted in two opposite ways: On the one hand it guaranteed a millennial reinforcing stability of the dictionary. On the other hand, if things shifted (or broke), the whole corpus changed. While generally counteracting lexical and sound changes, if things changed, they changed on steroids.

While this does not have too many repercussions on the vocabulary (yet), such changes in alphabetic pronounciations often have wholesale sound-changes throughout the language as a result. (Compare French hamburger = 'amburger')

This precondition is exuberated in reconstructed liturgal languages. In isolated alphabets (such as Armenian ) such changes can remain undetected for a long time.

Fortunately grammarians give hints when and when not such changes occured. If grammarians clearly and unmistakenly lable a character as vowel, its possible consonatic ancestry is lost at that stage. Such sources would lose all meaning if there is any chance that they were altered in the process of scribal copying: If Aristoteles original grammar was copied by monks 1000 years later they would surely imprint their own bias onto the original work. Only literally original sources can be consulted.

Dangers of projecting modern pronounciation onto old words.

Alphabetic Change can often be traced back through ancestral alphabets and historic contexts. On the other hand, alphabetic changes can reflect back on ancestral alphabets so that the 'errors' of modern scholars extend to previous alphabets, perpetuating modern readings to false reconstructions of older readings. The comparative method can reveal relative connections better than absolute positions of millenium old linguistic constellations.

One of the most corrupted yet recognizable words is Mars < Azura Mazda > Hermes

How on earth does one derive Hermes from Azura Mazda? Through millenia of repeated prayers with lazy tongues. This is just a cautious example of what migth go wrong when projecting modern pronounciation onto old words.

Known cases of alphabetic change

𓎡 ⋍t > k (probably)

Orthographic reinterpretation over sound change

If you take the g of γ to be read as in 'genie' and the φ ph as a soft p/b suddenly the word γραφή sounds very similar to schreibe (<> scribe Schrift script )! A hard reading of g as in graph may be a back formation. On the other hand an even harder reading of γραφή grave ≈ carve is conceivable.

The point is: alphabetic entries often diverge into clusters of similar sounds and in retrospect it's impossible to say at which stage which branch was how advanced and which reading exactly was used. Thus it's accademically approriate to determine an approximant 'center' of a sound cluster but unscientific to insist that such center represents the actual reconstructed pronunciation.

This is well understood by most Good-Linguistics but one can still read some presumed /pronunciations/ of ancient names, which is nonsensical; even more so for Sumerian where even the center of a cluster is often highly tentative.

split of a sign 'van' intwo vν = ոn (this equality is valid today!)

There is a wide range of familiar and unfamiliar consonants humans can produce, much more than common alphabets of less than 30 signs could map:

Nasal m̥ m ɱ n̼ n̥ n ɳ̊ ɳ ɲ̊ ɲ ŋ̊ ŋ ɴ
Plosive p b p̪ b̪ t̼ d̼ t d ʈ ɖ c ɟ k ɡ q ɢ ʡ ʔ
Sibilant affricate ts dz t̠ʃ d̠ʒ ʈʂ ɖʐ tɕ dʑ
Non-sibilant affricate pɸ bβ p̪f b̪v t̪θ d̪ð tɹ̝̊ dɹ̝ t̠ɹ̠̊˔ d̠ɹ̠˔ cç ɟʝ kx ɡɣ qχ ɢʁ ʡʢ ʔh
Sibilant fricative s z ʃ ʒ ʂ ʐ ɕ ʑ
Non-sibilant fricative ɸ β f v θ̼ ð̼ θ ð θ̠ ð̠ ɹ̠̊˔ ɹ̠˔ ɻ˔ ç ʝ x ɣ χ ʁ ħ ʕ h ɦ
Approximant ʋ ɹ ɻ j ɰ ʔ̞
Tap/flap ⱱ̟ ⱱ ɾ̼ ɾ̥ ɾ ɽ̊ ɽ ɢ̆ ʡ̆
Trill ʙ̥ ʙ r̥ r ɽ̊r̥ ɽr ʀ̥ ʀ ʜ ʢ
Lateral affricate tɬ dɮ ʈɭ̊˔ ɖɭ˔ cʎ̝̊ ɟʎ̝ kʟ̝̊ ɡʟ̝
Lateral fricative ɬ ɮ ɭ̊˔ ɭ˔ ʎ̝̊ ʎ̝ ʟ̝̊ ʟ̝
Lateral approximant l ɭ ʎ ʟ ʟ̠
Lateral tap/flap ɺ̥ ɺ ɭ̥̆ ɭ̆ ʎ̆ ʟ̆

The following sometimes only hints to affinity of sounds without representing the true english pendant.

Candidates for such fundamental shifts in mapping characters to sounds are:
г H < G ( Ukraine ) 𓎛 H < Ḫ ḫ ḥ < HR < TR
ϩ R > Ḫ > H
հ ᵖh>h
յ ᵗʸ > j
𐤏 so > סo > o ( ⵙ 𔗆 sol > hel > ἥλιος)
ա ˢa/ᵟa/ᶴa > a
ω ˢwo > ᶴwo > wo > o
ⲱ 𐤔 ש ˢwen 𓈙 𓈖 𓍶 šιn… Ω
ϩ S SḪ 𓎛 Ḫ HR WR 𓏲 𓍢 > H
ե ˢj/ᶜj/ᵗʰ > j ( ե ⋍ tj)
ո 𐌿 ⋍ vu > vɔ/ɔ (ok որոշ = vɔˈɾɔʃ)
ו v/u
𐌰 V/A > ᵛA > A
α pha > a
y γ /ʒ/ > y
𐍅 γʳ > ᵞv > v
ⲩ γʳ > ᵞu > u
𐌾 Gj > ɕj > j
𐍈 Q® > ƕʍ > wvo > w
и NI > I
ⲏ Н N > hn > η /ɑ̃ⁿ/ > E Eh < Ēta < •hĒta < •henta < •hemnta
Н ⲏ N > hn > η /ɑ̃ⁿ/ > H Ha < Han < •henta < •hemnta
η η > hn > ɑ̃ⁿ > eⁿ < ēta < •hēta < •henta < •hemnta / hengta ŋƞɲɳ
ի ի > hn > iⁿ > i (as in ether)
Я R > yɻ yɺ̥ yɺ yɾ̼ yɾ̥ yɾ yɽ̊ yɽ ? > YA
𒄩 H < HA < Ḫa < KHA KU₆ KUD COD GU > yú 鱼
𐎅 H > E ⋍ 𐎅 ⋍ 𒂍 É ⋍ 𒂊 E
𐎅 HE < ḪE < Heim / Chamber / 𒊓 Saal < 𒂍 Kallu
p ~ r proto p ~ q peculiar 𓈎 TAR ⋍ QAR 𓏘 ( quatro / tetra )

Most modern letters that look familiar don't just do it superficially, they mostly have a common history and thus common pending sounds.

Egyptian signs

Egyptian signs were not purely alphabetical and thus most likly exhibited even more severe shift in their phonetics.
All readings are highly tentative, after several millenia of disuse any claim of exact pronunciation is unscientific.

Since every single sign underwent sound-change, there is no point listing them explicitly here, instead they can be investigated by case: Gardiner-signs.

𓇋 pj>fj>j>i/e/a …

Complex forms

One reason why non-alphabetic sign systems with some history show surprisingly complex forms is that the original form was subject to sound-change and to prefer the original sound, a more periferal form is used:

E.g. 𓐠 vs 𓈙 for the sound "S" in Darius inscriptions, presumably because other signs such as 𓋴 had become too ambivalent:

𓐠 𓏘 𓂧 𓇋𓇋 𓈉 | Sogdia
𓊪 𓃭 𓐠 𓈉 | Persia
…

𒊓 Saal > hall 𒂍 É ⋍ 𒂊 E
𒂍 Kallu > hall 𒂍 É ⋍ 𒂊 E

On the other hand simplified forms often have more entropic sounds:
𒂍 É ⋍ 𒂊 E
𒀸 aš > 𐎀 a

Roman alphabet

The Roman alphabet took the Phoenician and Greek approach one step further by being fully abstract and not being reminiscent of the words which once gave rise to the letters.
E < FE < Beith < Port (historic values not present upon symbol creation)
e < wo < qv ( all < ealle = whole < 𓎟 kʷəll keh₂i-lom )

However there were possibly still some changes of the alphabet over 2500 years:

A radical conjecture, which would take years to ingest, is that some of our very familiar letters had (slightly) different readings 2000 years ago. If all branches of languages switched from one reading to another, the original reading could nearly be forgotten.

So the following proposed changes must be analyzed only with original sources, and rarely with transmitted or transcribed texts:

i:r
e:v
u:v (just backreading Latin u=v)
o:s (only very early stages of Latin)

ER R > Я > ί > YA > yo I

The confusion arises when different stages of the letter are mixed at different times. If the letter 'i' was read as 'r' at some stage it surely was not at other stages or dialects.

Gothic

Old Armenian հուր ᵠhur ⋍ πῦρ pûr
New Armenian հուր hur ⋍ huo 火

հինգ hing pyng (five)
հաստ hast feist (fat, thick)
հայր hayr père vatr (father)
հետ տալ het tal phathal πτύω փսխել pʿsxel 𓃀 𓈙 𓅱 𓂐 𓏥 (spittle)

Problem:
Wide spread homophonic readings. If the whole continent reads one word similarily, how can it have had a different pronunciation 2000 years ago?

Problem:

Redundancy

why would one postulate
ⲏ⋍N when we already have ⲛ
ⲉ⋍∑ when we already have Ϣ ϣ Ϭϭ Ⲥⲥ Ⲅⲅ Ⲋⲋ
ⲩ⋍r when we already have 𓍱⋍ⲩ⋍γ⋍g

Solution:
There is a wide range of familiar and unfamiliar consonants humans can produce, much more than an alphabet of less than 30 signs could map.

Additionally single signs can represent consonant clusters : z=ts ψ=ps q=kw … todo more

On the other hand alphabetic (false) consonant clusters can truely represent a single sound: sch = ʃ …

Solution:
Other than early sound change at the root, and the very strong influence of the biblical canon one mechanism which surely transformed the whole landscape is ‘phonetic infection’ through alphabetic-change.

Problems:
Interchangability of “vowels”
ⲏ ↔ ⲓ N ⇔ R
ⲏ ↔ ⲟ N ⇔ S

Solution:
N ⇔ R are similar if using indian-like semivowels
May be different grammatical forms completely lost.
Expected ⲏ ↔ _ ⋍ ung

Solution:

Imperfect orthography.

As today, there was surely not a one to one mapping between characters and sounds.
There may have been orthographic rules when clustering signs, or some readings may just have been out of band for no (good) reason. Just browse through the modern english dictionary if you are not a bored or nodding linguist.

Solution:

Redundancy

μήνη mḗnē m.mn.vn.hn
ⲍⲉⲛⲏⲥ zonos is really z.s.n.hn.s
ⲍⲉⲛⲏⲥ genus is really g.s.n.hn.s
ⲏⲛ ⲛⲏ (difference stress nN Nn => EN NE!)

Since the spelling of Coptic is often transparent with the greek 🇬🇷 cognates, possible reconstructions of dialectic pronunciation may extend to greek words. It is conceivable that the letters ⲏ Н N > hn > η were read as •hemnta > Ēta at the same time. Since many of the greek words are part of our english vocabulary, we are heavily biased and resistant to attach new vocalizations to accustomed sequences of characters.

σελήνη selḗnē ☽ סַהַר sáhar.mḗnē μήνη 𒌋 MAN moon 𓏠 𓃽 ⇔ nin 𒊩

ἥλιος hḗlios “sun” in the reconstruction would yield something similar to ἥλι Nelir or Neri
While such a form would be completly unfamiliar to us, and contradict a connection ἥλιος ⇔ solarius,
ἥλι Nelirᵒˢ or Neriᵒⁿ (λι rr redundancy/clarification) might find a home in
наран naran ⲣⲏ ܢܘܗܪܐ nuhrā, nuhhārā, 𒉢 NU ⋍ 𒉡𒊑 نُور nūr light𒂟 nûru нэфнэ năfnă? Evenki: ӈэри ŋəri Bengali: নূর nur ⇔ illuminare and finally could have a synthesis in Javanese: sunar
An inverted reading Neri ἥλι Inri would raise eyebrows (not only of scepticism but of strange intimacy):
If one believes the strong “sun” background of the whole Christ saga, ἥλι Inri suddenly becomes almost obvious, especially if it is a bound article form ἥλι < µ λι < RI rì 日 ⇔ 𓂋 𓂝 𓇳 𓏤 ray, ra, re

Inri acronym reading of "Jesus Nazareth King Jews." may be PSM.

Suddenly 𓂋𓇳 ⲣⲏ reflects back the R⇔S change of 𓂋 ar arse os “mouth”
𓂋𓇳 ⲣⲏ •RUNᵈ / SUN / KÜN (𓂋 ⋍ 口 kǒu) or even
𓂋𓇳 ⲣⲏ ⇔ пие päiväⁿ 𓂋 ⲣⲟ pù 𒅤 mouth

This quickly yields to a feeling of "the whole system falls apart".
The trick is to observe the direct connections one at a time.
Then the utter state of confusion yields invaluable insights, however exactly the pathing between the connectios might have arisen.

Naturally we are adversed to matching a fractional core λι in ἥλιος with 3/5⋍60% 'packaging' to known roots.
However there is much precedence:
ⲁⲍⲁⲑⲟⲥ agados ⲁ.ⲍⲁⲑ.ⲟⲥ ⲍⲁⲑ > GAD > good

On the other hands it would not surprise, in fact should be expected, that other subunits of ⲁⲍⲁⲑⲟⲥ 'good' read slightly differently would yield other words for 'good' in the Eurasian language tree. While this is quickly reminiscent of the wildcard-fallacy, logically it makes sense to derive many short words from one long root.

Reconstructed readings

If for some reasons ancient 'holy' scripts are to be read back in attemtps to revive a forgotten language, based on the reconstructed alphabet, so much can be lost or shifted that two completely different, yet mysteriously 'correct' readings can appear:

𓋴 𓊪 𓂧 𓏏 𓇮 “spitz” sharp at one end, pointed 𓊪 𓂧 𓏏 𓇮
𓋴 𓊪 𓂧 𓏏 𓇮 •'ǧypdotos > ἀκιδωτός akidōtós

This is the case with so many Egyptian words that a true suspicion arises that indeed at some stage in its history an attempt was made to read the holy texts in a new religious liturgy.

𒆷𒊓𒈬 la-sa-mu l'𓃹
𒆷𒊓𒈬 ra-sa-mu rasen
𒆷𒊓𒈬 to run ⇔ δρομ drom < drasam!

atomization

Early in the history of rhebus syllabic writing started the process of atomization, extraction of phonetic and semantic components out of and into ligatures!