Sunday, March 23, 2014

Alphabet or Abracadabra? - Reverse Engineering the Western Alphabet


(It may be good to print out the three charts shown at the end of this article. Having the charts on the side will make it easier to follow the text.
The spacing of Indic text on some of the illustrations is flawed due to PDF format incompatibility.
)


Frontispiece


INTRODUCTION
"... No one is sure why those particular sequences of letters.
Maybe it is some mnemonic device that we no longer understand."

~ Lawrence Lo
When one looks cursorily at the way the Western alphabet characters are strung together, one hardly ever wonders why they are in the sequence in which they are.
From the age of being a pre-schooler we have become so familiar with the alphabet's sequence of sounds that to most of us, when we recite or sing the alphabet, it might even feel like we are reciting or singing an age-old rhyme.
Could it be though, that there is actually 'rhyme and reason' to the fact that ABCD is followed by EFG?

Actually, as we shall see, in spite of looking quite disorderly, hidden within that disorder we will find order. To discover that order though, we need to do some sleuthing, but with the help of the science of linguistics - specifically phonetics - we will make some surprising discoveries.

We will eventually be able to establish, using as evidence the 1400-1200 BCE Ugarit abecedary from Syria (see frontispiece above) and the historical presence of early Pre-Ashokan Brāhmī script (Subramanian 2004), that phonetics were already being studied many centuries before 1200 BCE.
The ancient Sanskrit grammarian Pāṇini (India, ca 600-500 BCE), in his treatise on Sanskrit, expanded on earlier Indian studies from around 700 BCE (Taylor, 2003). Pāṇini's work gave a more refined and detailed account of the "place and manner of articulation of consonants".




Figure 2
 
The major Indic alphabets today still order their consonants according to Pāṇini's classification (Fig. 1).
 
It is not obvious that the Western alphabet sequence derived from a pattern or grid similar to the one above - the one now universally used by the Devanāgarī  or Sanskrit alphabet (an abugida or alpha-syllabary 1, 2 ), but that is the theory offered here.

The way the theory in this study began to be developed started on a hunch.

Figure 3

Not having yet developed the multi-coloured grid above (Figure 2), I at some point tried to force the Western alphabet characters into some kind of tabular format of columns and rows (Figure 3). I started out by lining-up the vowels in the left-most column. Subsequently, when I fitted the remaining consonants into adjacent columns, I obtained a somewhat orderly result.
Years later, after I compared my first obtained character grid with an early simpler format of the Sanskrit Devanāgarī alphabet grid (Figure 2), I got another hunch: I should be able to find clear, perhaps even obvious links between the two alphabets just based on their character placement and arrangement.

That second hunch led to the theory presented here: the dependence (if not direct, then at least indirect) of the Western - Late Roman Alphabet arrangement on the Devanāgarī Alphabet 3.

This dependent linkage between the Western (Late Roman) and the Devanāgarī alphabets is, as we shall see, evidenced by their similar if not almost identical ordering of characters in grids of comparable phonetic classifications.
Also in this theory, the origin of the dissimilarities - which turned out to be caused by human error - has been traced back to two errors made by an ancient (budding?) linguist!
It must be said though that even if his were mistakes, this theory shows that they were 'somewhat understandable'.

As the Western Alphabet in this theory's tabular arrangement appeared to be irregular in certain places, (see the blackboard picture above - Figure 3) especially when compared to the highly organized and well ordered Devanāgarī alphabet grid, it will be shown here how and why those irregularities came about.

After the error identifications as described in this study, and by taking into account the varying but close pronunciations of a number of comparable characters in both alphabets, and focusing on the proper placement of nearly all Western characters in their appropriate vowel, labial, guttural and dental columns, we will discover that the two tabular arrangements (the Western alphabet and the Sanskrit abugida) match each other by 85.00 %!

We will also be able to conclude that the above depicted clay tablet from Ugarit, Syria, which is dated between 1400 and 1200 BCE, is evidence that even before it was inscribed with its cuneiform abecedary, that a West-of-India style irregular alphabetic sequence was already in use, a sequence which in origin was based on an earlier simpler format of a Pre-Ashokan Brāhmī abugida grid of script characters. This Pre-Ashokan Brāhmī abugida grid must even have existed before 1700 to 1300 BCE 4, 5, 6.

Notes to the Introduction

1 For the sake of simplicity I will use the nomenclature "Sanskrit alphabet" in this study. However, although often used, that nomenclature is only partially correct as it is actually an abugida or alpha-syllabary.
An abugida is a series of writing characters in which each complex non-vowel character (e.g. फ pha) represents the sound of a complex consonant-vowel combination. It is not a complete syllable though.
When an abugida character is pronounced by itself, it ends with a short 'a'. But that short 'a' is not all the time pronounced when that character's sound is part of a word. For example: देवनागरी is pronounced as devnāgrī not devanāgarī . When an 'a' is to be pronounced, it appears in western print as 'ā' and in Devanāgarī as  ना, which is  न  plus an extra l. So  ना is  nā, or  /na:/ in phonetic format.

Whenever I refer to or show the Sanskrit script in the Devanāgarī format, it is actually for convenience sake only as it is not truly known which kind of script Pāṇini used in his treatise on Sanskrit. The earliest form of the Nāgarī script proper is from the 8th century CE, and an early version of Devanāgarī is from 992 CE (Taylor, 2003). That version shows the emergence of the horizontal bar to letters or groups of letters - indicating distinct words.

3 Since putting this study on paper, more detailed and nuanced analyses enabled me to find out more exactly how Pre-Ashokan Brāhmī, Brāhmī, Nāgarī, Devanāgarī, Ugarit Syrian, Aramaic, Phoenician, Greek, Etruscan and Roman alphabets are related. Of course it is already well known that they are, but it is possible by analyzing the difference in distribution, number and kind of character (e.g. the number of vowels instead of diacritics) to establish which came before which. The scholarly consensus seems to favor the Aramaic alphabet, (Salomon, 1996) although the well known Indian philologist S. R. Rao is not at all convinced about that.
My plan is to bring us to a more definite conclusion in an upcoming study.

4 Which alphabet influenced what alphabet has been a debate for a long time (Hemple 1930). It seems to me that all too often cultural, national and even religious biases are involved.

5 There exists a Proto-Sinaitic script (from between 1700 to 1300 BCE) from which the Aramaic script developed, and there is even an earlier Proto-Sinaitic script: a Proto-Alphabetic one from 1900 BCE (Colles - 2009) which was discovered in Upper Egypt. Thus far it is accepted that this script is influenced by Egyptian hieroglyphs. (Zuckerman, Swartz Dodd - 2003, Colles - 2009)
This early Upper Egyptian script and early Brāhmī 400 BCE (Salomon, 1994) have much in common, although the spread in time between them is about 1500 years.
So far it appears that this Proto-Alphabet has no vowels, whereas the earliest Brāhmī does and... in their proper order. Of course having vowels or not, is not usually used to establish precedence of a script, as according to linguistic theory vowels are normally used to distinguish alphabetic from syllabic writing systems. But there are indications (and vowels come into play) which lead to a conclusion that Brāhmī is the more ancient one.
It is significant that the Brāhmī and Devanāgarī alphabets do not show any error in their consonant and vowel order of phonemic categories, whereas the Ugarit, Phoenician, Greek and possibly the Aramaic alphabets do - and of course the later Roman or Western alphabet versions.
If that is indeed the case, then the earliest Pre-Ashokan Brāhmī is prior to all of them. Then only the link between Pre-Ashokan Brāhmī, the 'Early Proto-Sinaitic' script from 1900 BCE and Egyptian glyphs remains to be found.

6 There have been two major discoveries of inscriptions that may be related to the Proto-Sinaitic script, the first one in 1904–1905 in Sinai by Hilda and Flinders Petrie, dated to the mid 19th century BCE, and more recently in 1999 in the Western Desert of Middle Egypt by John and Deborah Darnell, dated to the 18th century BCE. (http://en.wikipedia.org/wiki/Proto-Sinaitic_alphabet)
 

CHAPTER ONE - Abracadabra?


The Western or Late Roman Alphabet on the Blackboard Strip
(At the top of the Chart One.)
"No one is sure why those particular sequences of letters. Maybe it is some mnemonic device that we no longer understand…"
"http://www.ancientscripts.com/alphabet.html
~ Lawrence Lo.
Markandey Katju recently wrote in an essay in "Frontline" the weekly magazine if the Indian national newspaper "The Hindu" (Jan. 28-Feb. 10, 2012) 1:
"[...] Take, for example, the alphabet in the English language. The letters have been arranged in a haphazard manner. Why is B followed by C? Why is D followed by E? There is no reason why F comes after E, P is followed by Q, or Q is followed by R.
In Sanskrit, on the other hand, Panini arranged the alphabet in a scientific manner. For example, take the consonants. There is a sequence ka, kha, ga, gha, na (called the 'ka varga'). All these sounds come from the throat. Also, the second and the fourth consonants in this sequence are what are known as aspirants. An aspirant means a consonant in which 'ha' is added. The second and fourth consonants in every sequence (of five consonants) are aspirants.
The sounds in the second sequence of five consonants ('ca varga') ca, cha,ja,jha,ha, all come from the middle of the tongue. The sounds in the 'ta varga', ta, tha, da, dha, na, come from the roof of the mouth; the sounds in the sequence ta, tha, da,dha, na come from the tip of the tongue; the sounds in the sequence pa, pha, ba, bha, ma come from the lips.
We can see how scientifically these consonants are arranged."
To put it differently:
The wonder of the Devanāgarī alphabet is that it organizes human language sounds in a table of columns and rows.

The Western or Late Roman Alphabet
(The blackboard on the left on Chart One.)

1. Each column is according to where the sounds originate in the mouth - from rear to front:
- a. The vowels are formed in the back of the throat at the top of the wind-pipe or trachea,
- b. Progressing forward come the gutturals, palatals, cerebrals, dentals and labials.

2. The row arrangement is according to how air passes through the aperture between the tongue and the roof of the mouth (the uvula and the soft and hard palate), the teeth and the lips, from sharp or raspy, to soft or velvety, or nasally without air or with very little air passing through the mouth.

Markandey Katju makes an excellent point telling us about the scientific arrangement of the Devanāgarī alphabet, but is the English alphabet - or more correctly - is the Roman alphabet really a "haphazard" mix of characters without any order to them, when he says, "The letters have been arranged in a haphazard manner."?

Around 1975, when I was trying to familiarize myself with Sanskrit, I was - like Mr. Katju - also struck by the neat order of the Sanskrit Devanāgarī alphabet. (It also made learning the Sanskrit characters so much easier.) Then, one day, while I was in a Canadian restaurant, an idea popped into my head, and while grabbing a napkin and a pencil (this restaurant purposely had pencils on all its tables - not just crayons for the kids,) I started reciting the Western alphabet to myself while I - on a hunch - put emphasis on the vowels: A E I O U... thinking to myself, "What if they showed some order?"

And I repeated: "A E I O U."

I did not pronounce them the usual English way though, but the Dutch way: the way the vowels sound in the English words 'far, net, it, cot, nul: 'a, e, i, o, u'.
I noticed, as I repeated the sounds again, but now as 'glottal stops', that my lips went from wide apart to closer together as I was forming and voicing those vowels.

(Somehow nobody noticed me, which was a good thing... pursing my lips could have been misunderstood.)

I took the pencil and wrote them down on the napkin, but... in vertical order:
a
e
i
o
u 
Then - again on a hunch - I added the consonants to them, but in short row format so that the whole scheme looked like:
a b c d
e f g h
i j k l m n
o p q r s t
u v w x y z
Looking closer, I noticed (except for the middle row) that the second character sounds, the b, f, p and v, were all labials (formed by the lips). So I mused, "What if the third characters in each row are also similar to each other? The c, g, k, q, and x". I also decided that because the 'v' and 'w' were so similar in pronunciation, that I could consider them as one sound, and I knew as well that in the 'days of yore' the 'c' was pronounced as 'k', e.g. Julius Caesar and Kaiser Wilhelm.
Well, 'c g k q x' are all similar, they are what you call gutturals (velars and palatals), which means that the back of the tongue touches the roof of the mouth while they are being voiced.

What about the fourth characters on each row?
Here I ran into trouble, Hmm?!
Nevertheless, I thought there was enough correspondence to not deem this 'order' accidental or "haphazard".

Some years later (I was then a Montessori teacher) I drew the grid that you see on the left of Chart One - on the blackboard.

"Not bad!" I said to myself.

(For the purpose of this study, I am concentrating on an abridged vowel and consonant inventory: just the vowels, gutturals, dentals, and labials. See "Closing Remark 6" at the end of this study.)

The Three Vertical Strips: the 'ka', 'ta', 'pa' Vargas
(The three tan strips on the right of the top half of Chart One.)

A few years later I took up Sanskrit again, so much so that I even started dreaming about it.
In one particular dream, 'I' was in a scene from very long ago, looking over somebody's shoulder, while that 'somebody' was scratching signs that were unfamiliar to me, in a vertical manner on a number of palm leaves.

When I woke up, I could not recall if the signs were Sanskrit letters or not, also, in those days, as I was just learning Sanskrit, I did not know about Brāhmī yet.
In any case, in a subsequent dream I was handling three of those palm leaves. Apparently I had carried them 'home' with me, and still in my dream, I was trying to arrange them in their proper order, but... I had forgotten what that proper order was. (I have also forgotten where 'my home' in my dream was.)

In 2007 I took this idea up again, remembering those palm leaves in a similar way as the three tan vertical strips  at the top right of Chart One - the 'ka', 'ta', 'pa' vargas 2.

So far so good, with a bit of creativity I could make the Roman alphabet characters fit the same order as a simple version of the Devanāgarī alphabet.

Notice (and remember for later) the two small ovals containing the letters K/L and Y, they didn't fit the scheme properly, no matter how hard I tried. But 'miraculously' after some more 'confusion', the K/L found an open spot and the Y (the Ypsilon or 'I Grecque', which is not common in some Indo-European languages) happens to be a first century BCE Roman insert influenced by the Greek language.

I am showing the Late Roman Alphabet, the early one (called Ancient Latin 3) did not have the G and the J, nor the V and W, while the K, X, Y and Z were only used for words with a Greek origin.

The Four Multi-Colored Tables 4
(The bottom half of Chart One.)

Let me quote Markandey again:
"In Sanskrit, on the other hand, Panini arranged the alphabet in a scientific manner."
I am sure that Pāṇini (600-500 BCE) in his treatise did come up with something that looked like the opening picture in the Introduction of this study, but foreseeing some 'reverse engineering' and actually taking my dream seriously, I could see that prior to Pāṇini, the tabular arrangement must have been much simpler, and also, in retrospect, I'm now sure that Pāṇini and his predecessor used an early form of Brāhmi: Pre-Ashokan Brahmi.

The three tan colored vertical strips must have been from a time between an original simple 3 by 3 grid holding just nine characters (the small "Nāgarī - Original" table at the bottom left of the chart) and one (Step 2) that had grown to fifteen characters.
Steps 1, 2 and 3 represent my reconstruction of the transition from a 12, to a 15, to a 25 character grid by constructed Pāṇini.

The Timing of the Linguistic Errors

Figure 4

Edge of a
writing tablet
Etruscan / Greek
The already flawed Syrian (Ugarit) cuneiform alphabet from between 1400 and 1200 BCE (R. D. Woodard, 2008) - or one very similar to it - must have been the basis for the much later 'Old-Italic' Roman alphabet arrangement.


Different Near-Eastern linguists must have been involved though, because the  Ugarit alphabet was vowel-less while the much younger 'Old Italic' had vowels.
One of the linguists (the one I dreamed of) must have gone to India to study the Indian script, but when he returned home he had brought only three palm leaves with him - instead of four - he must have lost the one with the vowels, and it was that lack of vowels together with his confused re-arrangement of the three palm leaves that caused the an abecedary very similar to the Ugarit one, to be vowel-less.

Another linguist, who may also have gone to India, but later, must have returned with the Brahmi vowels that his predecessor had missed or lost. These were then subsequently added to the already flawed alphabet arrangement.
There is a difference of some 500 years between the Ugarit alphabet from Syria and a Greek alphabet (Fig. 4) from ca 800 BCE found in Marsiliana d'Albegna, Etruria, featuring the oldest Etruscan / Greek writing.

The 'Old - Italic Roman' alphabet was based on the Etruscan one, which in turn was based on a Greek alphabet, which was in turn based on a Phoenician alphabet, but somewhere in between vowels were inserted.

Looking at the dating of the tablets and the sequence of the characters we can tell that the first error had already been made before the Ugarit tablet, and that the second error must have happened much later. Perhaps we can find out when.


Abracadabra


Figure 5
The expression 'abracadabra' has an interesting history; initially it had nothing to do with magic.
First take note of the last three letters... 'bra'.
The word 'algebra' ends the same... people doing algebra were speaking... "gibberish".
"Algebra" derives from 'al-jabr', a Persian word  that could mean 'restoring' or 'putting things in order'.

Thus 'abracadabra', although it had to do with mysterious sounds (to the layperson), it was the linguistic science of putting sounds and their representative signs 'in order'.

Let's look closer at the first part of abracadabra: 'a-bra-ca-da':

It starts with the vowel 'a' - the first letter of all alphabets. (Also 'abugidas')
'Bra' may have originally been 'ba' - you will see later why - the second letter of the various alphabets that were constructed after the three palm leaves (the ones that had Brāhmī characters scratched on them) got put in the wrong order. You will see how that happened in Chart Two of this series.
  1. 'Ca' - the third letter of the various alphabets that were constructed after the three palm leaves got mixed up.
  2. 'Da' - the fourth letter of the various alphabets that were constructed after the three palm leaves mix-up.
The three palm leaves had the Brāhmī characters 'ka, ta, pa' at the top, but after they got mixed up, their order became 'pa, ka, ta' and eventually 'b, c, d' and thus 'a-bacada-bra'.

Another name for certain alphabets is 'abugida', again following the mis-ordering of sounds: 'a-bu-gi-da.

Let me repeat some paragraphs from above

"The wonder of the Devanāgarī alphabet is that it organizes human language sounds in a table of columns and rows:

1. Each column is according to where the sounds originate in the mouth - from rear to front:
- a. The vowels are formed in the back of the throat at the top of the windpipe or trachea,
- b. Progressing forward come the gutturals, palatals, cerebrals, dentals and labials.

2. The row arrangement is according to how air passes through the aperture between the tongue and the roof of the mouth (the uvula and the soft and hard palate), the teeth and the lips, from sharp or raspy, to soft or velvety, or nasally without air or with very little air passing through the mouth..."

Any alphabet grid or alphabetic sequence of characters 5 that does not follow that order of natural progression can only be post-Brāhmī.

Notes to Chapter One

1 http://hindu.com/fline/fl2902/stories/20120210290208900.htm

2Vargas - divisions

3 Link to article about the Latin alphabet
http://www.orbilat.com/Languages/Latin/Grammar/Latin-Alphabet.html

4 The simplified Nāgarī tables at the bottom half of the chart above, reflect my earliest findings. It may be that some of the specific Nāgarī character columns need to be reviewed.
It could be in Step 3 on Chart One that the choices I made between the two yellow columns (guttural or palatal) as well as between the two green columns (cerebral or dental) need to be reversed. This nevertheless would not alter the basic tenet of the theory as presented.

5  There is one caveat:
It is of course hard to say whether the mis-ordering was done to alphabets that originally had the right sequence, but there is another way to figure that out.
That will be dealt with in yet another series of charts and... so far it looks like the Indus Script graphemes will be involved.


CHAPTER TWO - Two Errors of Historic Proportion


1. The Ancient Language Student's First Error
(The tables on the top half of Chart Two.)

The tan colored table on the extreme left is the same as the three separate tan (palm leave based) strips that we saw on the previous chart, except here western characters (IAST) are added. The strips are joined together and in the proper order:
  • the 'ka - guttural' strip,
  • the 'ta - dental' strip,
  • the 'pa - labial' strip.
A few years ago (2008) I decided to compare the two grid arrangements: the tan table (the transition stage between Step 1 and Step 2 from Chart One) and the light yellow table, the corresponding table with Late Roman Alphabet letters.

I expected them to be the same, but as you might remember, the second one of my two dream scenes was tainted by an uncomfortable feeling.I wanted to find out why that was, so I kept checking, but at first sight it did not seem at all obvious what might have caused my discomfort.
 

Figure 6
Palm leaves as in my dream.
Now - in retrospect - I know what happened, it is something that, I'm sure, everybody has experienced.
Have you ever looked through a stack of photographs, and every time after you've looked at a photograph, you placed it at the bottom of the stack, but then, for some reason, you forgot to return to the very first photo - the one that should've been at the top?
Well, in the case of the three palm leaves, something similar must have taken place. In this case though, the one that was originally supposed to be at the bottom of the stack (or laid out in the right-most position) was now on the top of the stack (or - after I laid them out in my dream - it had mistakenly taken the left-most position.) But as I (in my dream) had only a vague memory of which sounds those scratches really stood for, I had no way of knowing what their right order was.
Let's take a closer look.
Compare the two adjacent tables between the red arrows in the top center of Chart Two. Below them I am showing the top row of each table to compare their character sequence...

Yes...!

"OOPS !!!"

The three strips are now as shown on the right: the wrong order - as in my dream - except then I did not know!

The two multi-colored tables on the bottom left of Chart Two show the error in another way: column three accidentally had become column one.

2. The Ancient Language Student's Second Error?

At that point I expected that the character correspondence was a match, albeit the wrong order. It was clear to me now, how the Roman alphabet had become mangled.

But something else must have gone wrong as I was to discover next. Too bad though that I had no more clarifying dreams. So to unravel the next problem, and to reconstruct how that happened, some real detective work was needed.

Of course right now we have the advantage of knowing how it all eventually turned out, but this ancient (budding?) linguist had to reconstruct it all by himself, and in spite of some difficulties (at least one more 'oops') he did a genial job...
Of course he did not use Roman characters, nor Greek or Aramaic ones. In his possession he had the "Brāhmī Character Set" and he set-out to match his own set of characters to it.

Properly, I should have started this project by comparing either the cuneiform Ugarit Syrian abecedary or the Phoenician one with Sanskrit alphabet formats, but as my original discovery went via the western alphabet, and as western characters are nowadays more widely recognizable (and we don't have to deal with two series of non-Western scribbles) I think this choice was a good one.
In any case, we will pursue the Ugarit Syrian and Phoenician connections with Brāhmī in much more detail in an upcoming study.

The difference between Indian speech and Near-Eastern speech is that in the latter the 'h' and the 'g' are more raspingly pronounced (laryngeal); as well, aspiration of their consonants differs quite a bit. In addition the flexibility of the tongue inside the mouth also is quite distinct.
It could well be that 3000 years ago the difference was similar... it must have been because when our linguist tried to reconstruct what he remembered, he appeared to have some difficulty matching the sounds he remembered and their location on the palm leaves.
Also, instead of using strips like palm leaves, he might have scratched or painted his - what he thought were - corresponding characters on pot-shards or on a different kind of writing medium.
It is generally accepted that  Brāhmī, was written from right to left (sinistrograde) , but when he was in India, Brāhmī may already have been written from left to right (dextrograde).
However, there are indications (it showed up in the manner the discovery in this study came about) that writing was done like Chinese: vertically, and this writing later turned either clock or counter clock-wise, respectively becoming right-to-left or left-to-right oriented.

In the process of reconstructing what he had learned, and while fitting his own characters into a grid of columns and rows, he may have turned either version of characters 90 degrees clock or counter clock-wise. Any which way, the lettering twisted and turned.
It is amazing (let's for now forget about his palm leaf dis-placement) that this linguist only made one more error (maybe two).

Eventually I figured out that he accidentally moved one row of characters upwards and while doing that he also moved them over one spot. (That maneuver is shown on the next chart).
For now just check the purple and light-yellow columns underneath the second "OOPS": the bottom 'ma' moved up two rows (to eventually become the Roman the M), while taking the two 'na's with it.

The light-yellow table (bottom right) changed because of all this...


CHAPTER THREE - Order Restored (Almost, that is!)


Tables A and C

I added the blue vowels column to table A, but what has to be kept in mind is, that the vowel column was not involved in the row manipulation that accidentally took place and which will be described next. The vowel columns in both Tables A and C will come in when we are going to establish how well both tables match.

Tables A and B - The Ancient Language Student's Second Error

When one follows the red arrow from the bottom of Table A, one can see how in Table B the bottom row of voiced nasal stops moved up two rows and… one spot over sideways… one spot too many in fact.
That was how 'ma - म ' and 'na - ङ' received the position that our linguist used for the his equivalents of the Roman M and N.
Hence the additional irregularity in the Western - Late Roman alphabet. This way also a spot came free for the K or L, or... both. (The L in those days was probably a semi-vowel, just like the L in the English word 'feel'.)
Also, our linguist probably did not need that extra 'न - na'.

Table C (the Blackboard) in Relation to Table B

You may already have been wondering about a few things:
  • What about the vowel columns?
  • How can V/W be matched up with 'bha - भ' in the purple cell on the bottom left of Table B?
  • Why haven't I said anything yet about R/S/T, and what are they doing in Table B's green 'da - द' cell?
  • How can Q have anything to do with 'ga - ग '?
The answers:
  • What about the vowel columns?
There are up to four distinct vowel characters in Sanskrit of which only 'a - अ' and 'i - इ' are used most often, so how can there be five in the Roman alphabet?

In the earlier days that Brāhmī was written, at least up to around 300 BCE (some time prior to Emperor Ashoka) Brāhmī featured five clear vowels. Later Sanskrit had them too but only two of them are most regularly used, for the others diacritical marks were used instead.
We can conclude from this, that probably someone else (not our linguist, as the vowels did not get mis-arranged) picked up his vowel information at some other occasion and kept it on a separate palm leaf. (It also was not part of my dream.)
  • How can V/W be matched up with 'bha - भ' in the purple cell on the bottom left of Table B?
There is a sound and character 'waw' that appears in many Near-Eastern languages. The labials V and W represent that 'waw'.
It is quite common that the V and W get pronounced as either an aspirated 'p' or 'b' or (in Devanāgarī) as 'bha' and 'pha'.
(Incidentally, the modern word 'mobile' (cell phone) is pronounced as 'mowhile' in India.)

As I will show in a follow-up study, the locations in the purple column (Table B) of the Devanāgarī 'pha' and 'bha' (which respectively became the Roman F and V/W) were filled in with variations of the 'waw' sound. Even the Roman Y came to represent 'waw'.
  • Why haven't I said anything yet about R/S/T, and what are they doing in Table B's green 'da - द' cell?
As many English speakers know, when they have to learn a language in which the 'rolling r' - the alveolar trill - appears, that to learn that sound, they usually start by 'playing' with the tip of their tongue against or almost against their teeth to get a short 'd' going in the hope to get it to repeat itself so as to end up with trill.
The point is, it starts with the tip of the tongue in the 'd' position.
The rolling 'r' in many ancient languages was not necessarily rolling very much.
Take the word 'Sanskrit' संस्कृत, the ligatured character indicating 'kr' is कृ, it is a क (k) with a little hook at the bottom, that little hook stands for the 'r'. That 'r' is like a short quick 'd' that sounds like an R but just at the beginning of a roll... "sanskɖit".

The r, s, z, t, d, n, etc. in languages like Sanskrit are what you call 'retroflex consonants', meaning they are formed with a concave tongue somewhat against the front of the hard palate. Phonetically the characters' retroflex-ity is represented by a little hook at their bottom right: ɽ, ʂ ʐ ʈ ɖ ɳ , etc.

In the Western alphabet these retroflex sounds might not be so retroflex, because the Western tongue is 'flatter' so to speak. Notice that many of these characters were fitted (squeezed almost) into the bottom right of the table: the R, S, and T, and below it the Z and above it the N.
  • How can Q have anything to do with 'ga - ग'?
Except for Greek and Latinized Greek loanwords and just like the X and the Y, in West European languages the Q was not a common character used in the pronunciation of native words. Nevertheless when loanwords became 'as good as' naturalized (words like 'question' and 'quarter') a place for the Q had to be found and that bottom right corner being a bit of a refuse container had an open spot (the G was already fitted elsewhere) so the  'ga - ग' spot was just right for the Q and... after all, Q is a guttural sound.


CHAPTER FOUR - Conclusion, The Match between Devanāgarī and Western Alphabets

Taking into account the varying but close pronunciations of a number of characters (as detailed above: B-pa, C-ka, D-ta, F-pha, G-kha, P-ba, Q-ga, T-dha, V-bha , X-gha, Z-dha) and focusing on their placement in the appropriate labial, guttural and dental columns, we see that 13 of the 14 white blackboard characters match the grid positions of Table B… only H does not match.

The overall character placement match (including Table A's 4 matching vowels: a - अ, e - ए, i - इ, and u - उ) between Table B (13+4) and Table C (20) is 17/20 or 85.00 %!

Forcing the Western - Late Roman Alphabet characters in some kind of tabular format (initially on a hunch) by starting out to line-up the vowels in a column, and subsequently by fitting the consonants into adjacent columns, a somewhat orderly result was obtained. Later, by comparing the obtained character grid with an early simpler phase of the Sanskrit Devanāgarī alphabet, it became possible to discover clear links between the two alphabets.

I have made it a point to show the dependence (if not direct, then at least indirect) between the Western - Late Roman Alphabet on the Devanāgarī Alphabet.

This dependent linkage between the Western - Late Roman and the Devanāgarī alphabets is evidenced by their similar if not almost identical ordering of characters in grids of comparable phoneme classifications.

In this theory, the origin of the dissimilarities - or, in fact, irregularities - has been traced back to two mistakes made by an ancient linguist. Although, even if they were mistakes, this theory shows that they were somewhat 'understandable'.

As the Western - Late Roman Alphabet in this theory's tabular arrangement, appeared to be irregular in certain places, especially when compared to the highly organized and well ordered Devanāgarī alphabet grid, this theory shows how and why those irregularities came about: the horizontal displacement of one column containing labial consonants and a subsequent vertical displacement (and an extra horizontal shift over) of one row of voiced nasal stops.

Significantly, the vowel column did not in any way get mis-arranged, this could indicate that either two different Near-Eastern linguists were involved or that the original linguist went back to the source to find more information, although, that is doubtful, as in that case the original error would likely have been spotted.

When we compare Charts One's Table 1 (the simplified Devanāgarī table) with Table 2 (Western Alphabet characters fitted into a grid) at first glance, there does not seem to be any similarity - except for the vowel column. Looking closer however, and as demonstrated, Table 2 shows the remnants of an attempt to classify the consonants characters phonetically according to where in the mouth they are articulated: the tongue, the palate, the teeth, the lips, etc.: the first column shows vowels, the second one mainly labials, the third column mainly gutturals and the fourth one shows mainly dental consonants.

Row-wise with the consonants, there seems to be a vague phonetic classification as to plosive, fricative or nasal sounds, but it is not very precise.



When these two tables were compared (before the tracing of errors and before the corrective reconstruction) a percentage of similarity of only 20 % (4 out of 20) was produced. But we calculated that:
  • after the error identifications (as described in detail),
  • after taking into account the varying but close pronunciations of a number of comparable characters in both alphabets,
  • and while focusing on the proper placement of nearly all Western - Late Roman characters in their appropriate vowel, labial, guttural and dental columns,
that we arrived at an 85.00 % match between both alphabets.

We have also been able to conclude that the Ugarit abecedary from Syria, which is dated to be from between 1400 and 1200 BCE, is evidence that even before it was inscribed with its cuneiform abecedary, that a West-of-India style alphabetic sequence was already in use which in origin was based on an early simple form of the Devanāgarī alphabet grid.

Closing Remarks
  1. Initially it were the Western and Sanskrit Devanāgarī alphabets that led me to the discovery that is described in this study.
  2. Subsequently, after more analysis, it became clear that there were transitional phases between the two, likely via one or more Near-Eastern alphabets.
  3. It also became clear that the Sanskrit Devanāgarī alphabet must have been based on an earlier Proto-Pre-Ashokan Brāhmī alphabet from which:- a. subsequent Near-Eastern alphabets arose, or
    - b. which showed that these Near-Eastern alphabets were highly influenced by a Proto-Pre-Ashokan alphabet, or
    - c. that there was an even earlier, possibly pre-dynastic Upper-Egyptian Alphabet, which gave rise to all other follow-ups, and lastly
    - d. that somewhere within this sequence of alphabet developments, many characters can be fitted and matched of which their origin can be found in the Indus Valley Script and which likely can be traced back to said pre-dynastic Upper Egyptian characters.
  4. This analysis was never intended to trace the formation of individual script characters or any formative link between the various scripts' characters. However, it now appears to be possible to establish or confirm some rough dating of the formation of and linkages between distinct script character sets.
  5. One, possibly two more studies will follow that detail and illustrate the remarks made in note 2 and 3.
  6. For the purpose of this study I concentrated mainly on categories 4, 5, 8, and 9 of the consonant list below, bringing 4 and 5 together under the header 'gutturals'.
  7. Phonemic categories of consonants:
    1. Laryngeals,
    2. Pharyngeals,
    3. Uvulars,
    4. Velars,
    5. Palatovelars,
    6. Sibilants,
    7. Interdentals,
    8. Dentals,
    9. Labials,
    10. Liquids,
    11. Nasals
  8. If there was a cursive script in use alongside the glyphic script as found on seals in India's Indus Valley Harappa Culture (3300–1300 BCE) - which I hypothesize there was - it would be an early Brahmi 'nāgarī lipi ' (urbane script) used by a more sophisticated segment of the Indus Valley population, possibly the elite who next to the more common Prakrit, spoke an early form of Sanskrit or Pali.

Bibliography
  • Steve T. Bett, (1996) "The Origin of the Alphabet".
  • Florian Coulmas (1991) "The writing systems of the world".
  • Prasad Das (2007) www.hindu.com/2007/12/20/stories/2007122054820600.htm.
  • Orly Goldwasser  (Mar/Apr 2010) "How the Alphabet Was Born from Hieroglyphs". Biblical Archaeology Review (Washington, DC: Biblical Archaeology Society) 36 (1). ISSN 0098-9444.
  • Elizabeth J. Himelfarb (2000) "First Alphabet Found in Egypt", Archaeology 53, Issue 1 (Jan./Feb. 2000) page 21.
  • L. Lo http://www.ancientscripts.com/alphabet.html
  • Richard Salomon (1996) "Brahmi and Kharoshthi" in "The World's Writing Systems". 
  • Richard Salomon (1995) "On The Origin Of The Early Indian Scripts: A Review Article", Journal of the American Oriental Society 115.2 (1995), 271-279 1995.
  • William Schniedewind, Joel Hunt (2007) "A primer on Ugaritic".
  • T.S. Subramanian (2004) "Skeletons, Script found at ancient burial site in Tamil Nadu". http://www.hindu.com/2004/05/26/stories/2004052602871200.htm
  • Isaac Taylor (2003) "History of the Alphabet: Aryan Alphabets, Part 2", Kessinger Publishing, ISBN 9780766158474. "... In the Kutila this develops into a short horizontal bar, which, in the Devanagari, becomes a continuous horizontal line ... three cardinal inscriptions of this epoch, namely, the Kutila or Bareli inscription of 992, the Chalukya or Kistna inscription of 945, and a Kawi inscription of 919 ... the Kutila inscription is of great importance in Indian epigraphy, not only from its precise date, but from its offering a definite early form of the standard Indian alphabet, the Devanagari ..." 
  • Roger D. Woodard (2008) "The Ancient Languages of Syria-Palestine and Arabia".
  • Bruce Zuckerman, Lynn Swartz Dodd (2003) "Pots and Alphabets: Refractions of Reflections on Typological Method (p. 89), MAARAV (2003) "A Journal for the Study of the Northwest Semitic Languages and Literatures".


© 2012 Wim Borsboom
All illustrations (except the author's charts) are public domain.

Charts for printing:



Wednesday, March 19, 2014

Being Higgs

Detail from a BBC video frame

Visualizing Higgs Space, the Higgs boson/field and virtual Higgs bosons.
(A philosophy of physics section will be inserted here shortly.)

The Higgs boson's 'presence' and 'absence' causes space to feature localized, internal, 3 dimensional wave-like density variations in a phase-state (non-solid/non-liquid) that is at once firm, gel-like and resiliently flexible... not like molasses (as it is often described) but like... "Jello" (T.M.)
"Jelloid Higgs Space"
A virtual Higgs boson appears as a contracted rip or compacted ripple in Higgs space. It is the possible virtuality of the Higgs boson that enables the spatial variation in Higgs space compaction or contraction.
Individual masses within their individual jelloid Higgs environments, affect their surrounding Higgs space density. This results in varying the 3 dimensional (rippled, or rather 'rimpled') compaction of Higgs space into spherically extended formations.
The particles that compose a 'mass object' inside Higgs space, cause an equivalent number of Higgs bosons to 'disappear' into virtuality so as to make room for the presence of the massive object itself.
The resulting space compaction extends onion-shell-like around the 'contained' mass object, thus producing inverse square density properties that make their 'contained' masses APPEAR as though they  - 'mutually' of course - affect other massive objects in their proximity.

This makes them seemingly attract each other - however, the mass objects are NOT attractive/ing at all...
At some point a spin-induced, torque-like shaped equilibrium is reached, which gives the apparent path of any mass (as contained within jelloid Higg space) its oval, or rather, spiral path appearance and apparent movement.
Their SEEMING attraction is not a quality/property of the mass objects in space, but a quality/property of multiple, mutually interfering spatialities: mass-filled/mass-affected jelloid spatial Higgs field-spheres.