This is a discussion some of us have been wanting to have for years, but had no idea how to begin. Awkwardly enough, I've been sitting on this for some years now (so, does that make me the bad guy?). Anyway, I will include, in my next post, two documents, being:

Comparative Igboid by Kay Williamson (deceased), Roger Blench & Chinyere Ohiri-Aniche.
Background: This document is not the completed work (i.e. official publication). Kay Williamson unfortunately passed away, leaving the work at a standstill. The PDF I will upload is Roger Blench's attempt to compile and organize Kay Williamson's work into a single manuscript. According to Blench, there may be some inconsistencies in the manuscript, due to Kay Williamson having used different/inconsistent phonetic fonts. So, it would be important that one be cognizant of this when reading through the manuscript.

Reconstructing the Igbo Cluster by Chinyere Ohiri-Aniche.
Background: This document is a more or less completed manuscript, and pretty much serves as Ohiri-Aniche's effort to continue on with the work left behind by Williamson. According to Ohiri-Aniche, part of the reason for Williamson's work having gone incomplete is due to the overwhelming variety of speech in the Igbo-speaking region. New surveys often would turn up new and obscure cognates that would force Williamson et al to reconsider their proto-Igbo conclusions.

Now, despite the diversity in speech, researchers still have to select a methodology and a sample population. For both works (Williamson et al & Ohiri-Aniche), the sample population centered around the same set of Igbo lects, being Owere, Ekpeye, Izhii, Ukwuani, Onicha, and Ogbakiri (Ikwere). Why these were chosen by the researchers is not exactly clear. Although we can reasonably speculate (and likely be correct), it would have been nice if they were explicit in the reasoning and methodology behind their selection.

In my next post, I'll upload the documents. This is because I understand well that some on NL aren't keen readers. The two manuscripts will require reading and critical thinking, and it is important that one also read the preface/introduction for the manuscripts. But because I know some will just skip to the reconstructions (and remain uninformed), I decided to make this initial post and provide some background from the two manuscripts. It will be important if anyone intends to carry out discourse on the subject.

*smiles*

This ought to keep me busy for the next couple of nights.

Do come back as soon as you can with the PDFs.

This is Nice

Nice thread.

I use to have that Pdf file until virus attacked my system and I lost it.
The comparisions were interesting. Ekpeye even was found to be compatible with Onicha Igbo by upto 76%.

ChinenyeN, stop peeping and get the thread started already nah.

Haa m! I'm just here gauging viewer responses. Anyway, it seems that NL does not support uploading PDFs as attachments to posts. So, I'm looking for a way to upload and share without forcing others to sign up or log into anything.

Alright. Sorry to keep you all waiting. Here is the link, as promised. Feel free to view and download.

https://www.dropbox.com/sh/fai471ny20v7vxg/AAD-lr9nAzR_QmG6gl3zFvnZa?dl=0

** I am not the owner of the linked manuscripts and am only sharing them for the sake of the dissemination of knowledge and freedom of information. Enjoy! **

Haha!

ChinenyeN, there is an interesting observation I want to make. It's one I have wanted to make for a very long time, but there has never been an opportunity and a place to make it. Well, until now. Some of these reconstructions appear to support that observation - I do not want to call it a theory just yet.

But, I've got some early-morning business to attend to. I'll make my observations later today.

I'll certainly be waiting to hear that any other observation you have to make. I've got some of my own as well, but I want to get as many thoughtful minds on this as possible, beyond my own.

Some Background on Lexicostatistics - The Swadesh 100 Wordlist

Lexicostatistics is a method where cognates are actually quantified and not qualified. It is with lexicostatistics that linguists arrive at what is called degrees of cognancy (in other words, shared etymology). Cognancy is a measure of the relationship between cognates, which helps in building language family trees. Many tools are used in the field of lexicostatistics. However, the Swadesh 100 Wordlist, developed by a U.S. linguist named Morris Swadesh, has become one the most widely used tool among linguists for lexicostatistics.

The wordlist is utilized mainly because the words/concepts in the wordlist are generally universal among languages. Knowledge of the Swadesh wordlist does not mean that one will be able to perform lexicostatistical analysis on terms across languages. This is because the list is not trivial and the method of calculating the degree of cognancy is also not trivial. There is a methodology applied to it and it specifically requires the use of cognates and not just similar sounding words.

I would list out the terms in the modified Swadesh 100 Wordlist, if not for the fact that you would read it anyway in Ohiri-Aniche's work (i.e. all the comparisons you are reading are based on some form of the modified Swadesh 100 Wordlist). It is how Ohiri-Aniche arrived at the percentages of similarity when comparing cognates between Onicha and the other lects in the sample. The degree of cognancy clearly shows the relatedness and ancestry (with respect to linguistics) among the speech varieties in the Igbo-speaking region.

Seriously waiting for your observations. I can't download any pdf files for now.

General Background on Linguistics - Understanding Cognates

Cognates are not an easy thing for many people to understand (so far that I've seen). In layman's term, cognates are 'similar words'. However, even that is an oversimplification of cognates. In reality, a cognate is a word with similar derived etymology. Unlike 'similar words', which may share the same meaning, cognate words need not share the same meaning across languages. Instead, they only need have a shared etymology, meaning they share the same ancestral root or word. Cognates don't even have to sound alike to be cognates. I would imagine this is where the difficulty lies with many. A commonly used example is the English word wheel and the Sanskrit word cakra. The words are true cognates of each other, being derived from the same Proto-Indo-European ancestor, though they sound nothing alike.

To bring this closer to home, there are some examples of this attested in Comparative Igboid by Williamson et al. The word 'daylight' has the following cognates across the Igbo languages ihe, ife, and ikworo, among others. Notice that ikworo does not sit as a 'similar sounding' word with ihe and ife, but is still a true cognate, because they all have the same etymological ancestor.

This often makes it difficult for most people to identify cognates, without a true or native understanding of sound shifts between languages. But fret not. It is not an impossible skill to learn. The concept may just take some getting used to, for those who do not immediately grasp it. However, I did feel the need to explain it beforehand (before the discussion begins), since it is a concept we will have to make use of in our discussion. So, it is important everyone understand it, as it will facilitate a much richer discussion.

We are waiting for the onus of this thread. Pls spill the beans I'm already getting bored.

Yujin, this thread requires posters and viewers to be informed. I have presented two PDF documents for viewing and downloading by to-be posters and casual viewers as well. I have also been steadily establishing the background and foundation for a rich discussion with my most recent posts on this thread.

This is not a topic to be [mis]treated like many other culture topics. If you want to quell your boredom, feel free to read through the manuscripts I referred to earlier. The link has been posted for easy viewing and downloading. There is a great deal of information that will go into this topic. As such, I neither expect nor anticipate nor desire for anyone to jump haphazardly into a discussion of this nature.

Seriously. I very much encourage you to read the manuscripts. If nothing else, read them out of curiosity and for new insight.

I also do not intend to hold anyone's hand through the course of this thread.

Okay! This might be long and rambling. I hope you somehow manage to piece together what I'm trying to say.

I have to warn, this is going to be super-boring for people who aren't really interested in linguistics.


Before I Get to the Point:

ChinenyeN, you remember a short discussion we had here a while back about 'm' and 'n' as 'myiriudaume' or pseudo-vowels? For the benefit of anyone following this discussion who might not know about 'myiriudaume', let me give (as much as I can) a brief explanation.

'M' and 'n' are usually regarded as consonants ('mgbochiume'); but there are cases in Igbo where they behave rather like vowels. This is usually in circumstances where they begin a morpheme or a word and, instead of being followed by a vowel, they are followed by (another) consonant. In such cases, 'm' or 'n' is said to be a 'myiriudaume' - a voweloid, if you will. I'll try to illustrate.

In the following words, the first 'n' or 'm' is followed by a vowel, and not by a consonant:

Nata (collect), noro (stay), nüö (marry), mee (do), mara (know).

In the above examples, they ('m' and 'n') perform the function of regular consonants. However in the examples below, 'n' or 'm' is followed by (another) consonant, and not by a vowel; and in these cases, they function not as true consonants, but as pseudo-vowels or 'myiriudaume':

Ndu (life), nze (a title of nobility), mkpi (he-goat), mbido (beginning). I might also add nna (father) and mma (knife) - where the second 'n' and the second 'm' function as regular consonants, as in the set of examples before these, but where the first 'm' and the first 'n' do not behave like regular consonants and are thus pseudo-vowels or 'myiriudaume'.

Easy and practical way to tell when 'M/N' is a pseudo-vowel rather than a consonant: The mouth is closed while producing the 'm/n' sound for myiriudaume, and the sound is produced somewhere in the throat - without much assistance from the lips.


Getting to the Point:

Now of all the Igboid lects, Ekpeye appears to be the only one that doesn't recognise 'myiriudaumes'. Where a 'myirudaume' would appear in most (if not all) the Igbo lects, Ekpeye substitutes with a regular vowel - an 'a', an 'i', an 'ï', an 'u', an 'ü'.

I'll furnish just a few examples:

Ncha (soap) in Ekpeye is ïcha
Nne (mother) is ïne
Ndo (shade) is Udo
Mbo/Mvo (Claw/Nail) is Ubo
Nmekpahu (Stress/Trouble) is Umekpeli.

You could go through Blench's Dictionary of Ekpeye and not find a single non-contestable application of 'myiriudaume'.

Did Ekpeye have Myiriudaume, like the Igbo lects, but lost them at some unknown period in history? Or were myiriudaume absent in Proto-Igboid - and were only acquired by the Igbo languages after Ekpeye had split from the main Igbo linguistic sphere?

My suggestion is that proto-Igboid lacked the concept of 'myiriudaume'. For one, our closest linguistic cousins - the Edo and the Yoruba - like the Ekpeye, also appear to lack them. That lack explains the fact that a Yoruba (before he has mastered Igbo sounds) is very likely to pronounce Nneka as Ineka and Ndigbo as Indigbo. What the Yoruba-speaker is unconsciously doing is 'reducing' the 'myiriudaume' 'n' to the position of a regular consonant by placing a vowel ('i' in this case) before it, making it easier for his 'myiriudaume-unfamiliar' speech organs to handle.

In Ohiri-Aniche's paper I noticed three or four reconstructions that appear to favour the hypothesis of a non-myririudaume proto-language for the Igbo-speaking groups:

For 'Central Igbo' Nti (Ear), Ohiri-Aniche suggests proto-Igboid Eti.

And for 'Central Igbo' Mpi, she suggests proto-Igboid Upe*

I wish she had tried to reconstruct more proto-Igboid equivalents of more myiriudaume-containing modern Igbo words (Mkpi, Nta, Ngwo, etc); unfortunately, those two - and maybe one or two others - are the only ones she attempted in the paper.

The manuscript 'Comparative Igboid' (which I suspect to be somewhat outdated, despite being more comprehensive) does not appear to fully support the hypothesis I'm pushing for here, i.e, that proto-Igboid didn't have the concept of myiriudaume. But it still has many interesting proto-Igboid reconstructions. eg:

For mba (nation/tribe), it has proto-Igboid uba, and for nkata proto-Igboid akata.



Speculating the Possible Source of Myiriudaume for the Igbo Languages:

Not a few NLers have opined that the Igbo have a special relationship - blood and linguistic - with the East Benue-Congo-speaking groups, i.e, the linguistic groups predominant today in Cross River, Akwa Ibom and Bantu Africa. I think there is an extent to which this is true.

I remember a very keen-minded non-Igbo NLer saying that Igbo appears to share Ng-, Nk-, Mb- words with the East Benue-Congo groups. He probably didn't know at the time that he was referring to the concept of 'myiriudaume'. I suggest that Myiriudaume was introduced into Igbo through intense prehistoric relationship with East Benue-Congo groups - and that this happened sometime after the linguistic ancestors of Ekpeye had split from the 'main' Igbo language-stock.

I've probably gone further than I originally intended to. Pardon me. I hope you can make sense of this sha.

Radoillo.

I enjoyed that your last post, please don't stop now, keep posting more of that here, I am hooked.

Good post, Radoillo. Honestly, part of me wants to believe that more minds will want to (and hopefully will) contribute, but it seems like it might just be the two of us for the time being. I can't really blame anyone. The two manuscripts are a lot to digest.

Anyway, moving forward...

I do remember our brief discussion on 'myiriudaume'. I hadn't thought of considering 'm' and 'n' as pseudo-vowels until you had mentioned it at that time. That was a new angle for me. Oddly enough though, my observations came from a different direction, but still led me to somewhat of a similar conclusion as yours concerning the limited role of m/n in proto-Igbo. It's interesting that we'd come to some similar conclusions, but from different angles.

My angle came from my understanding of tone. I believe that most would be commonly inclined to view tone as an aspect of lexicon (i.e. change in tone will change the meaning of a word). I, however, have come to believe that tone is just as much an aspect of syntax as it is an aspect of lexicon. I've never tried to actually put this collection of thoughts together in black and white, but I'll try and try to be as coherent as possible.

I will admit that my conclusions derive strictly from having observed Ngwa at the syntax level. The reason why it is limited to Ngwa is simply because what I have observed (dubbed as 'nasal insertion' by Benson O. A. Oluikpe, an Ngwa linguist, 1979) has yet to be attested for in any other lect (to the best of my knowledge). This syntax element, nasal insertion, involves affixing roots with a nasal prefix. Depending on the root and the person writing, the nasal may be written/rendered as an 'n' or 'm', but regardless, it is still the same nasal insertion.

Ex: nzere m whne ojoo ngburu we
Trans: may I be so fortunate as to escape the evil that killed them

Ex: whne nnula iwhnu oma mbia
Trans: may something akin to good fortune come

At first, I never knew what to make of nasal insertion. It always struck me as some odd and peculiar aspect of speech that was present only at the syntax level. The full nasal expression is typically never realized during nasal insertion (in other words, the throat does the heavy lifting during nasal insertion, just as you noted for other instances). However, above all else, what interested me more about nasal insertions was that they carried what I felt was excess tonality, as if to say they sit in place of something else.

Admittedly though, there is more to my observations and conclusions, but the much I've talked about is the point where our two observations seem to converge. If I take a step back to consider the 'myiriudaume' as pseudo-vowels, then the excess tonality manifested in speech really begins to fall into place.

Your East Benue-Congo-speaking speculation is certainly interesting. It is worth looking into, but we would have to see how we can establish some sort of reasonable timeline for the interaction. I also have a speculation of my own, but it emphasis a more internal source for innovation, and is actually centered on tonality at the syntax level. It is actually an extension of my previously mentioned observations and conclusions. I'll be back to share those.

FOLLOWING

I see. There is a clear relation between this nasal insertion and what I was trying to saying.

Putting it very crudely, it appears (to me, a non-Ngwa) that Ngwa lect (from your examples) even 'overdoes' it - somehow. The Ngwa situation, as shown in your examples, doesn't seem to me like a simple case of a nasal m/n taking the place of a vowel in our hypothesised proto-Igboid. It now appears - from a non-Ngwa perspective, of course - like an 'unneccesary' insertion of an Myiriudaume or a nasal m/n.

In your sentences, for example, a non-Ngwa such as myself might wonder at the prefixing of an 'n' in 'ngburu' and an 'm' in 'mbia'. In Izugbe, as well as in the forms of Igbo spoken in the Awka area (and many other areas besides), and given the context both words were used in the given sentences, 'gburu' and 'bia' would do.

It is interesting that we both (coming from different angles) seem to agree that proto-Igboid probably lacked the nasal 'm/n' or myiriudaume. I'd like your take on a possible source or sources of this innovation in the Igbo area. Are there reasons why you think the innovation occurred within the Igbo area, rather than being an imported element? Where in the Igbo area would you locate the cradle of this innovation?

This topic is going to breed nothing more than a series of long posts.

Anyway, I believe I understand what you may mean by 'overdoing' it. I guess I should also state then that the context of the sentence is inconsequential with respect to nasal insertion. At the syntax level, all verbs are subject to 'nasal insertion', even auxiliaries. I suppose one could call it an obligatory aspect of the lect.

Ex: whne unu nka nto uto
Trans: what you all are saying is sweet

Ex: nkwa mgba nji akpa m uwha (ji is the auxiliary)
Trans: mgba music typically makes me feel joy

Nasal insertions aren't arbitrary in most cases, as they do establish mood and create tense. For instance, in the first example, the insertion before the verb roots 'ka' and 'to' fully displaces the present tense auxiliary construction. For the second sentence however (with the auxiliary), syntax rules in Ngwa require nasal insertion, despite the fact that it does nothing to the sentence.

The added excess tonality (as I see it) from nasal insertion has always made me suspect some proto-Igbo fallout, but I never had any real support for my suspicion until I read the comment by Williamson et al about Margaret Green. Apparently, Margaret Green believed that Igbo had 'too many tones chasing too few syllables', so the manuscript reads. How this would apply to nasal insertion (if it applies at all) interests me, because that excess tonality has been my observation and forms part of the basis for my conclusions.

Now, with respect to the myiriudaume innovation.. if you remember, I stated that there is more to my observations and conclusions (from having observed Ngwa at the syntax level). This is where those observations and conclusions lead. I'll start this off with a sentence example.

Ex: nmema ala nma nma
Trans: land reform is pleasing

Three things are occurring in the above sentence example (at least, by my interpretation). We've got our first 'nma' of the sentence as an example of nasal insertion, the second 'nma' as an myiriudaume noun and finally 'nmema' as what I interpret to be a pseudo-noun. Let me see how I can explain this...

'Nmema' represents another aspect of Igbo syntax that I believe parallels nasal insertion (not saying that they serve the same purpose) and is not specific to Ngwa. This syntax phenomenon basically occurs during sentence construction. It uproots an otherwise unsuspecting verb from its stateless infinitive and binds it to a subject without inducing any sort of tense or mood. If tense or mood exist, then it comes from other attached roots and affixes.

I believe this is what we can hold responsible as the source of the myiriudaume innovation (or at least, to some degree). We cannot completely rule out interaction with East Benue-Congo-speaking groups. Clearly, there is parallelism between the myiriudaume innovation and their nC- constructions (where C represents any consonant). I simply believe that the parallelism played a more secondary role in all this. If I had to speculate, I would say that it is possible that the interaction with East Benue-Congo-speaking groups might have facilitated the transition from pseudo-noun to full-fledged myiriudaume nouns. That's just a speculation though.

LOL @ 'nothing but a series of long posts'. But it really can't be helped, can it?

I think I see your point about a likely local source for the innovation being discussed. I initially hadn't taken into consideration that the Igbo lects spoken in the south contain nasalizations. And that these nasalizations probably provided some 'experience' (if you will) of nasal 'n' usages before the emergence (or introduction) of 'myiriudaume' proper.

Moving away from one observation...

I noticed something interesting about the Igbo numerals in the 'Comparative Igboid' paper. You may have noticed it, too. I'll be back to share this observation in a few hours, when I'm less busy.

I have also come to believe that it is possible to broadly group all the Igbo lects into two - similar to the grouping of Indo-European languages into the Satem languages and Centum languages. I'll be sharing that later, too, and I want to know what you think of it.

I know there was one interesting thing (a pattern) that I noticed about the reconstructed numerals. I'm not sure if it's the same thing you noticed, but it certainly was strange enough to catch my attention.

I would mention it explicitly, but then no one would read the manuscript.

Odumchi, you might like this. Do you remember our discussion some years back about the 'sh' sound in Igbo? Well, it looks like the manuscripts might have somewhat of an answer for us on that.

According to Williamson et al, it seems that Igbo development gave rise to two different 's' sounds, one 'old' and the other 'new'. The old 's' is originally proto-Igbo. In other words, efforts to reconstruct proto-Igbo for certain words ended up with an 's' sound. Apparently, the old 's' (the proto-Igbo 's') is mostly responsible for words that have the h/s/sh/r dynamic across lects (i.e. ohi, osi, oshi, ori -- theft).

Ex1. River -- ohimiri, orimili, osimini -- osimidi (proto-Igbo)
Ex2. Slave -- ohu, oru -- osu (proto-Igbo)

The new 's', which is the modern Igbo 's', is apparently derived from one of the proto-Igbo 't' sounds. It would seem that one of the proto-Igbo 't' sounds is mostly responsible for words that have the s/sh/ts dynamic across lects (i.e. asi, ashi, atsi -- hate).

Ex1. Tree -- oshishi, osisi, usisi, otsitsi -- utiti (proto-Igbo)
Ex2. Head -- isi, ishi, risi, rishi, itsi, itshi -- diti (proto-Igbo)

So, as it turns out, based on these reconstructions, 'sh' is actually not proto-Igbo, but rather an aspect of modern Igbo. We were wrong from the beginning. The 'sh' is not older that 's'. Go figure.

Approximating Proverbs in Proto-Igbo from Izugbe (assuming syntax remains the same)

Izugbe: Ura ga-eju onye nwuru anwu afo
Proto-Igbo: Uda ga-ejuyi onie ṅudu eghuu dipwo

Izugbe: Onye hapu onu ya, uguru aracha ya
Proto-Igbo: Onie saa onu ye, ugudu adacha ye

Izugbe: Were ire guo eze gi onu
Proto-Igbo: Gwuwe idide guo eje ge onu

It's hard to believe really..

ChinenyeN:
Approximating Proverbs in Proto-Igbo from Izugbe (assuming syntax remains the same)

Izugbe: Ura ga-eju onye nwuru anwu afo
Proto-Igbo: Uda ga-ejuyi onie ṅudu eghuu dipwo

Izugbe: Onye hapu onu ya, uguru aracha ya
Proto-Igbo: Onie saa onu ye, ugudu adacha ye

Izugbe: Were ire guo eze gi onu
Proto-Igbo: Gwuwe idide guo eje ge onu

It's hard to believe really..

Interesting!

ChinenyeN:
I know there was one interesting thing (a pattern) that I noticed about the reconstructed numerals. I'm not sure if it's the same thing you noticed, but it certainly was strange enough to catch my attention.

I would mention it explicitly, but then no one would read the manuscript.

I'm almost certain we noticed the same pattern. It is just too obvious.

The proto-Igbo ( as well as the Ekpeye and some Ikwerre) numerals for the numbers 6 to 9 are compound numerals based on five.

i.e, six is itinu or - to use the more 'modern 's' - isinu (most likely a contraction of ise + nanu) or five-one;

seven is itaabuo or isaabuo (a contraction of ise + abuo) or five-two;

eight is itaato or isaato (ise + ato) or five-three. This is the only five-based compound that appears to have survived in a clearly-recognisable form in all modern Igbo lects;

Now nine is a bit problematic, from my viewpoint. And this is where I want to know what other contributors here think. It seems to me that the linguists' (Blench, Williamson and Ohiri-Aniche) reconstruction of proto-Igbo 'nine' as
'ituaalo' isn't really defensible. I don't even know what 'ituaalo' is supposed to mean. 'Itua' isn't proto-Igbo for 'five', and 'alo' isn't proto-Igbo for 'four'.

Perhaps, they were attempting a 'compromise' reconstruction that would sound like 'itaano'/'isaano', but also sound close enough to 'toolu' or 'itoolu' to have also been its parent-word as well.

Here is what I'm thinking: Iteghete/Itoolu/Itenaani isn't descended from a proto-Igbo five-based numeral, and thus shouldn't have influenced the reconstruction of proto-Igbo for 'nine' - which I think should have simply been 'itaano/isaano', rather than the complicated 'ituaalo'. [Interestingly, according to Blench's Onitsha Dictionary, 'sanno' is still an archaic Onitsha word for 'nine']

Now, how to break down 'Iteghete/Itenaani/Itoolu...

The 'modern' Igbo words for nine (Iteghete/Itenaani/Itoolu) still sound like compound numerals, just not a five-based numeral. I've been wondering how it could be broken down to its constituent numerals. 'Naani' and 'Olu' sound like 'one'. The rest is undecipherable to me.

Based on this, I'm wondering if Itenaani and Itoolu could mean something like 'remove one' or 'minus one' [from ten].

Thoughts?

ChinenyeN:
Approximating Proverbs in Proto-Igbo from Izugbe (assuming syntax remains the same)

Izugbe: Ura ga-eju onye nwuru anwu afo
Proto-Igbo: Uda ga-ejuyi onie ṅudu eghuu dipwo

Izugbe: Onye hapu onu ya, uguru aracha ya
Proto-Igbo: Onie saa onu ye, ugudu adacha ye

Izugbe: Were ire guo eze gi onu
Proto-Igbo: Gwuwe idide guo eje ge onu

It's hard to believe really..

No Igbo-speaker alive today would be able to hold a conversation with a proto-Igbo-speaker if it was possible to go back in time. I suspect that syntax too must have changed.

ChinenyeN:


According to Williamson et al, it seems that Igbo development gave rise to two different 's' sounds, one 'old' and the other 'new'. The old 's' is originally proto-Igbo. In other words, efforts to reconstruct proto-Igbo for certain words ended up with an 's' sound. Apparently, the old 's' (the proto-Igbo 's') is mostly responsible for words that have the h/s/sh/r dynamic across lects (i.e. ohi, osi, oshi, ori -- theft).

I'm going to run off the rails here, so pardon me.

This whole r/h/s/sh thing reminded me of an argument I had with myself a long time ago. I was considering cognates in the YEAI subfamily of West Benue-Congo (Yoruboid-Edoid-Akokoid-Igboid).

Then, the possibility that the word for 'head' in Yoruba, Edo and Igbo may be cognates - despite the fact that on the surface, they do not at all look like cognates - hit me.

If we make the assumption that r/h/s/sh sound shifts were not confined within one language cluster, but could have occurred across related clusters, then the likelihood that 'o - ri' (Yoruba for 'head'), 'e - hi' (Edo for 'head') and 'i - si' or 'i - shi (Igbo for 'head') are cognates seems quite high.

Based on the probable relation between ori-ehi-isi/ishi, I've been working on a theory on the likely nature of what we now know as the 'chi' concept in proto-Igbo times (i.e, some four to six thousand years ago, based on Armstrong's glottochronology).

Scholars of African religions see a connection between the Igbo cult of personal chi, and the Yoruba/Edo personal cults of the head (ori and ehi). But why isn't the Igbo chi as explicitly connected with the head as its Yoruba and Edo counterparts? And why isn't this Igbo personal cult known by the Igbo word for 'head' as is the case in the other two languages?

Is it possible that in the distant past, the Igbo 'chi' (or whatever name it was known by at the time) also had a clear connection with the head? Would this long-forgotten connection (assuming it existed) explain why a name like 'Isioma' means exactly the same thing as 'Chioma' - 'good fate/fortune'?

Was the personal 'chi' anciently known as 'isi', and did the waning into insignificance of isi's connection with the head necessitate the sound shift (and initial vowel loss) that led from 'isi' to 'chi'?

This little speculation may be a little off-topic, but it's something I felt I should voice out.

Well, i refuse to speculate or believe that ''Chi'' has been known/pronounced differently....

Radoillo, who uses nüo for 'marry' instead or luo? You mean drink or what?

Nwaiberibe:
Radoillo, who uses nüo for 'marry' instead or luo? You mean drink or what?

Most of the communities that use 'ani/ana' for land also use 'nuo' for marry.

Ok