The language of old Europe

[Daniel]

This is the question from the first post after the above post in which i did give you the short summary. Maybe you did not see my post like i didn't see your's...

That's not an abstract. That's an outline of a (problematic) methodology.
An abstract is about content, not methodology. (Generally it is for completed research, although it can also leave some things open.) You should list:
1. What problem you are addressing.
2. Which existing theoretical claims are maintained and supported.
3. Which are rejected.
4. What new insight your answer gives.
5. Why that answer is supported.
Again, all in about 200-500 words.

Some examples can be found here:
http://www.linguisticsociety.org/resource/model-abstracts#Adams
(And in many other places of course. Some journals include abstracts at the beginning of each article for example.)

I have been extremely consistent in my claims from the first post.

Possibly consistent, but not clear. I don't know what theories you are rejecting. For example, are you rejecting the theory that Proto-Indo-European existed (somewhere/at some time)?

I have given you summary of my theory and the way to prove it and disprove it here:

The problem with that is that the methodology is subjective because you have no objective way to measure meaning/relatedness.

From my post which you did not read, or maybe did not see: "Get all the words related to boundary, group them by common sound, get percentage for each group with common sound. Compare with threshold percentage. If distribution is uniform, we have randomness. If we have peaks, it is deliberate. Each peak represents one sound which is strongly related to meaning of boundary. "

That's fine. But:
1. You need to have a better way to "get all the words related to boundary". That's the fatal flaw in this.
2. You need to establish statistical methods that show that any grouping is unlikely to be found by chance. Completely normally distributed data is rare, so that's why statistical tests are used. Anything you find might be a coincidence. By comparing expected random variation to what you find, you will have a reliable reason to think the data is not randomly distributed if it differs significantly. However, statistical tests only work if you establish a clear hypothesis before looking at the data. If not, you're just data mining for observed correspondences-- there are always correspondences to be found. But a strong analysis involves a predetermined expected result, with the appropriate statistical test, then blind testing of that. In other words: you need to be able to blindly make predictions with your theory, not just tell a story about observations you have.

How do you objectively test meaning?

Now we're getting somewhere. This is extremely important.
There is no simple answer, of course. You're right that it's challenging. But science is challenging. "How do you determine the chemical structure of molecules?" It's a question, and you need to find an answer.
One method that may appear counterintuitive but actually works out relatively well is using corpora and looking for words that often appear near each other. That doesn't seem like it would work that well, but it does give you quite a bit of information.
A similar approach involves "semantic networks", which use various tools to gather their data (sometimes human surveys, sometimes corpora, etc.).

Language is never "objective" in the sense that gravity is objective. It's cognitive, so certainly we must use human intuition. But a researcher should not do that himself/herself. If you must use intuitions, then you must ask others, especially when what you're saying is so controversial.

Here's a possibility for a research question:
"If I ask speakers of English to identify whether or not a word is related to boundary, then they will statistically significantly more often identify words that include N, if they are presented with a list of random words from the dictionary."

You are biased by your own theory. You need something replicable-- something other people also understand and can do the way you do it. Something that, if you are correct, I can also do by myself and reach the same conclusions.

Treat humans as data processing machines. Under what circumstances would they confirm your hypothesis? What would you ask them to do?

I gave an example of a multiple choice question above:

What word is most related to summer?
A) Snake
B) Heat
C) Travel
D) Snow

I have no idea why your methodology would work, given unbiased questions like that.

It's actually very easy to run surveys like this (Amazon's Mechanical Turk can be used to get hundreds or thousands of responses in a single day if needed). Ask a few friends. See if it looks even close to what you expect. I'm doubtful. Feel free to prove me wrong.

No it is not. Boundary defines things. It does not necessarily enclose them as i already shown you.

That's an absurd thing to say. Would others agree with you? Are you talking just about words as they are in your head? Why do you claim this as a fact?
Your definition is so idiosyntractic and abstract that it isn't clear how it would or would not apply to any word. Again, see above about being objective.
Seriously, would anyone else have the same intuitions you do? If not, ask yourself why not.

Savannah comes from an African language.
open - Originally the past participle of the verb *eupaną, *ūpaną, related to *ūp (“up”). Root has no n.
plain - From Anglo-Norman pleyn, playn, Middle French plain, plein, from Latin plānus (“flat, even, level, plain”). From Proto-Indo-European *pelh (“flat”), *pelh₂-. Root has no n.

Nebula - From Proto-Indo-European *nébʰos (“cloud”). Cognate with Ancient Greek νέφος (nephos), Sanskrit नभस् (nábhas). The original meaning is sky. Derived from ne b = not material, hard. Boundary of material world.

Navigate - From Middle English navigate, from Latin navigo, from nāvis (“ship”) + agō (“do”), from Proto-Indo-European *nau- (boat), possibly, from Tamil நாவாய் (nāvāi). From nāvis (“ship”) + agō (“I do”). Actually from na + v + i + go = on + water + persist, continue, float + go. navi = na + v + i = on + water + persist, continue, float

1. Can you rely on existing etymologies for PIE? I thought you were rejecting that theory...
2. You reject etymologies for big reasons, but what about small reasons like sound change? You seem to think that you can analyze every letter in a word meaningfully, and I can guarantee that's wrong (unless you take sound change into account and even then it's a big stretch). Compare Spanish 'pes' and English 'fish'. Or the Latin root pesc- and Germanic fisk. What do you have to say about the meaning of these words?


Considering for a moment the gl- cluster, it would be possible to say something like this: at the moment gl- represents something about words like glimmer, glide, glow, etc., and in the future if it becomes (for whatever reason) kl- or l- or gw- or whatever, then it will still hold that meaning.
If the original claim works out, then this needs to interact with sound change.


Here's a much less extreme and more reasonable claim:
Possibly there are some arbitrary sound units that contribute to the meaning of a word. They vary by language and go through sound change.

Note that:
1. You cannot analyze whole words with that method. (That's still crazy.)
2. The correspondences would be indirect, based on systematic differences, not just "N looks the same in all languages".
3. You still need to find a way to objectively defend that claim.
4. It would not conflict with any existing theories because it would be part of them-- sound change as a starting point, etc.

Even if all of that is true, I don't see how important it is: so what if gl- has some meaning in English? Isn't that then just like an etymology? A kind of sound-based root, rather than morpheme based root? In fact, it might even be (mostly) morphological at some level, where gl- is sort of interpreted as a prefix.


Another question to ask yourself (and address in the abstract):
What is the minimum level of extreme/controversial ideas required to support my theory?
Make your theory as normal as possible, drawing from existing research. What smallest changes are required to support your ideas?


At the moment your ideas honestly come across like this:
"All of linguistics is wrong. Serbian and Irish are closely related languages. Because N means boundary. I said so."
(You have provided lots of details/stories, but none are more convincing than that.)

[jkpate]

jkpate

Thank you very much for attempting to help, but i have no idea what all this means.

Can you give me a list of words which according to the software you use relate to boundary. Let start from there. I don't know how the software calculates all this, and i would like to do some manual checks of it's accurateness.

Like does the list contain words like entrance, nip, snap, line nose, snow, no. If it does not then we have software that does not know what words related to boundary are, and any further calculations are based on wrong initial data set. Again, i am very grateful for your help. It would also help if you could explain how did they originally determine the word relationship when they were populating the database. That is crucial. Then what words did they use. Modern English or old English. We need to discard everything that came into English in last 300 years at least. We want only words of European origin, no Asian or African borrowings. And so on....

I already gave you the python script to produce the dataset exactly. There are 146,344 "lemmas," 62,408 of which might be viewed as multiple words because they include a space (like "demarcation_line"). (if we exclude these multi-word lemmas, the corpus-based JCN-measure is negative but not significantly different from zero, while the graph-based WUP measure is still significantly negative.) There are 20 lemmas containing "line" in the 2,000 lemmas most similar to "boundary", and only 9 in the 2,000 lemmas least similar to "boundary." "snow" is in the 2,000 lemmas most similar to "boundary," but "snowboarding" is among the 2,000 lemmas least similar to "boundary." Also, the vast majority of the words are unambiguously English; the only possible exceptions I saw while processing the data were "scientific" Latin terms.

Wordnet is a labeled graph. A graph is a set of nodes and a set of edges, in this case with labels, between nodes. In Wordnet, a node is a "synset," or a group of lemmas with approximately synonymous meanings (it does distinguish between senses of words, so that "bank" the financial institution is in a different synset from "bank" the side of a river). The edges encode meaning relationships between synsets, such as "hypernymy" ("color" is a hypernym of "red") and "meronym" ("finger" is meronym of "hand"), and so forth. Page 25 of five papers on wordnet has an example subgraph.

From the Enlightenment on, the singular ambition of science has been to develop techniques for producing and evaluating knowledge independent of personal opinion.

The original relationship between words, at the time when the word database was created was someone's personal opinion. Or did they use some magic to determine what is related to what. 

They did not use magic, but they also did not just write down their feelings. If you want to tell people your personal reaction to words, take a friend out for coffee; these are not scientific data. They devised a systematization of meaning relationships based on previous work on word meanings, and applied this system to a large number of words. If a new team were to follow the same set of rules, the resulting database would not be identical to Wordnet, but it would be largely similar. In particular, there's no reason to think that broad effects, like your proposed correlation between specific sounds and specific meanings, would change. Wordnet is a good resource for operationalizing word relatedness in a measurable way. As far as I can tell, the only way to tell if two words are "related" in the way you mean is to ask you.

The statistical procedure I used takes continuous measures of relatedness between synsets, and finds a set of weights that for each measure that predict the log-odds of having "n" for each measure. The WUP similarity metric ranges from 0.09 (least similar) to 0.93, and the coefficient for WUP in the mixed logit regression is about -1.86. This means that the variation due to WUP similarity in [latex]\frac{P(\mbox{has n} )}{P(\neg\mbox{has n})}[/latex] is [latex]\exp(-1.86 * (0.93-0.09)) \approx 0.2[/latex]: words with very high WUP similarity to boundary, other things equal, have "n" in them at about one fifth the rate of words with very low similarity. (and saying "other things equal" is justified because I picked two similarity measures that were not highly correlated).

-- Edit

apparently the latex rendering server is undergoing maintenence. That sentence should read:

This means that the variation due to WUP similarity in P(has n) / P(does not have n ) is exp(-1.86 * (0.93-0.09)) = 0.2: words with very high WUP similarity to boundary, other things equal, have "n" in them at about one fifth the rate of words with very low similarity

[dublin]

had to split this into next two posts:

djr33, i am continuing with answering some of your questions which i missed from page 9:

I said: "Group languages are invented by families. Families expand to clans, tribes, nations, races and propagate the language. They also propagate genes. This is why genes and languages are linked."

I am talking about the origin of the language. Time before books, films, easy travel. Time when communities lived pretty isolated from each other. By the way the same situation remained even in some parts of Europe until mid 20th century. For instance some isolated mountain shepherd communities in Carpathian mountains in Serbia did not know that there was a second world war happening around them. In this type of isolation there is direct link between language and genes. Look at R1a people. They lived between giant uncrossable rivers for thousands of years during the last ice age melt down. Do you think their language during that time was influenced by anyone? But once people start mixing, languages and cultures start mixing.

You say:

Shouldn't we also be able to relate family recipes and language then? Also eye color. Also religions. Also location. Also political views. Certainly all of these things are somewhat correlated-- families and communities are relevant to all of that, sometimes especially language. But none of it is particularly reliable.

We can actually relate these things, if we look at historical, ethnographic and anthropological data. They give us extremely valuable data which can help us to understand how language developed, and who brought what to it. Look at Jewish cuisine which tightly related to their religion and their language. Look at consumption of milk which is tightly related to first herders and their language which contained first words for milk for instance. Political views, as well as many other cultural attitudes, have believe or not, been found to be genetically influenced through studies of separated identical twins. This is amazing thing which completely changes our idea what culture is and how it is passed. Our culture is hardwired in our bodies through genetic and epigenetic programming. And if something like that can be programmed through genes, so can our language ability, our ability to hear and reproduce sounds.

I said: "Different families can use different logic to link context to sound. Some of them will use "N" to represent boundary, some will use "Z" and some will use "K" of what ever other sound they though was logically linked to the concept of a boundary in the environment they lived when they developed the language. These families merged, languages merged. Not all words in English come from the same "family" language."

And then you said:

Ok! Very important. So: sound-meaning correspondences are entirely arbitrary. Languages vary, so "dog" in English is just coincidentally not like "Hund" in German or "perro" in Spanish.
What you're claiming is that sounds are treated as meaningful units in speech by users of those sounds. That's more reasonable, but also harder to prove, because it's very hard to imagine clear evidence that would falsify it. (You're saying "sometimes" rather than "always", for example.)

How did you get that sound meaning correspondences are entirely arbitrary from the above i really have no idea. I said that languages were developed by families, in particular surrounding at different times. These languages are based on particular logic, which can be traced to back to the original sound (not word) language used by the original family, clan...You are right that this is hard to falsify, but just because it is hard to falsify it does not mean it is not possible to do it. Pick a language, check id the same sound to meaning logic works for the majority of  words from that language. Then look at related language. Repeat the same. I would expect that the same logic would work for the words which come from the same family language.

So then what is the point of these correspondences? Are you trying to track down individual families that they came from? There's a major time-depth issue in historical linguistics, and I'm certain you'll hit that.

I am trying to discover the old language(s) of Europe. I am fascinated by the ability of cultural and linguistic traits to survive millenniums. The fact that most people dismiss this as even a possibility, led to so many words in our languages which are "of unknown origin".

So far, I have been able to give etymologies using my sound to meaning algorithm, of words for which official etymological dictionaries have no etymology, and which correspond directly to the function of objects and gods and meaning of actions and rituals whose names i have been analyzing. I was able to do this for Sanskrit, Avestan, Latin, Old Greek, Celtic (Gaelic), Germanic, Slavic languages.

The linguistic discoveries are supported by archaeological, ethnographic, anthropological data and were most of the time result of the cross examination of all that data.

I believe that we can pretty much reconstruct the language down to last Ice age maximum 10,000 bc.

I don't understand this question 100%:

Additionally, what about multiple generations? Are you claiming that following generations of the same family (or speech community) no longer question the form of words? You said families vary in these associations. How are they maintained then? That seems like a fatal flaw in the argument.

What do you mean by questioning the form of words? Do you ever question the form of any of the words you are reading in this sentence? And if you do how do you do it and why do you do it? Family languages are invented to support conversation between members of the same family. If family A becomes clan AA and this clan becomes peoples AAA, in order for all of them to preserve the family link, they need to preserve the language. If they want to gather together for major religious ceremonies in neolithic times, as we know they did through archaeological data, they had to be able to speak the same language. The language is maintained by the group which needs to preserve itself. It is self regulating system. But fringe families of the AAA people can mix with fringe families of say BBB people. They have a need to communicate with both their people and their neighbors. So they form ABA or BBA or what aver combination of the two languages, depending on which fringe family is more influential, and we have a dialect. If this new ABA family starts spreading, you might end up with ABA peoples with their own ABA language. If ABA peoples then take over AAA people and force them, through cultural, economic, military means to adopt ABA language, we can have ABA language replacing AAA language....

Why did all major religions always insist on forcing everyone to adopt a new language, new names? So that it can change the reality of the group which is being converted. Why did all the governments did the same, standardizing languages within empires?  One language, one culture, one family.

Further, why do languages vary at all? If they all (probably) go back to the same original speech community (maybe 100,000 years ago?) then why don't we retain whatever sounds they used? Either it is consistent over time or it is not. So, which?

Languages don't go back to the same family. There were at least three speaking human sub groups before 100,000 bc: Africans, Neanderthals and Denisovans. They developed in isolation which lasted hundreds of thousands of years. During this time they acquired different genetic traits as adaptation to their surroundings. These adaptations probably included adaptations of language systems, which are directly influenced by epigenetic reprogramming triggered by experience. Mixing occurred in last 100,000 years. During that time additional genetic changes occurred as result of adaptation to changed living condition and experience. This means more changes related to language as well. This continues even today. What latest research is showing us is that our experience can influence our genes which can directly influence how we hear and how are we able to reproduce sounds. This is then passed down to our offspring. If I can hear Spanish or Slavic R but my vocal system is not able to reproduce it as sharply, we might end up with English R or French R. If I can't hear the difference between b and p it is a tossup which sound you will find in the language i speak. If the whole family is affected by this mutation, you end up with sound changes. Also originally all sounds were undifferentiated, as human ability to speak developed. There was no m,p,b,v,f there was something like mpbvf which is basically undifferentiated sound which is the root sound of the above consonant group. The whole group is produces by the same gross mouth position. The differentiated sounds are produces by tiny variation of lip position and pressure. As people developed their speech abilities, as they used speech more and more, their genes changed, enabling them to control their speech apparatus more and more precisely. This lead to sound differentiation, and  mpbvf turned first to diphthongs, like bv, mp, pf and then to individual sound m,p,b,v,f. And because these sounds are so close together, tiniest change in our ability to hear them and to pronounce them  will lead to additional sound changes which can then be propagated to the rest of the family and clan...

Yes. They words change so much that any kind of analysis like this seems absurd, even if originally there were correspondences.

I don't agree that we can not find the original correspondences. It is hard, we might not find all of them for all the family languages, but if i find most of them for R1a language, i will be happy. For me this is fun. Is linguistics fun for you?

I wish i saw this earlier, we could have avoided a lot of misunderstanding.

What is a "meaning"?
This may be an incredibly important point to discuss.

Meaning is the idea that a symbol on it's own conveys without help of any other word within the specific language. Symbols can be simple like color or sound, or complex like symbol or object (circle, cross, head) or word, or even more complex like picture (traffic sign) or sentence.

I believe that in spoken languages sound, and sound blocks, carry meaning. Sound block is a vocalized consonant like ta, go...These sound blocks were used in conjunction with actions to convey messages. The sounds eventually merged together to form words which replaced this audio visual language with audio language. Each one of these words has it's own meaning. This meaning can change through time, because word becomes a symbol with which people can associate new meanings. So to find the original etymological meaning you need to go to the oldest known form of the word. This is all we have to work with. But it is not as bad as you think. A lot of words go way back and have survived pretty unchanged, maybe not in English but in languages from which they came to English. 

You say:

Semantics: the study of inherent meanings of words or sentences
Pragmatics: the study of the meaning of language in context

"It's cold in here."
Semantic meaning: It's cold.
Pragmatic meaning: Close the window.

Roughly, semantic meaning is inherent while pragmatic meaning is contextual.

You need to go one level lower than that and look at word cold on it's own. The meaning of the sentence "it is cold" is the sum of meanings of "it" "is" "cold". Without cold having it's own meaning, the sentence would be meaningless. "it is ljhlkjow" means nothing to you or me now. We can agree that ljhlkjow means the same as cold. And we can continue using the word ljhlkjow in our own language. If you try to use it with your mother, she will rightly ask you: "what is wrong with using word cold? we all know it, we all understand it. you sound like a weirdo when you use ljhlkjow insted of cold. do you want to sound like a weirdo?" This is how languages are preserved. Through preservation of the meaning of words. If enough weirdos start using ljhlkjow instead of cold, new dialect will emerge. But you know as well as I do, that most people will just continue using cold. But if instead of cold we decide using caold, or colt, we have more chance of developing our own dialect, as the new words "sound like" the old and preserve more or less, the main meaning contained in sounds.

Semantic meaning of yellow light: in a short time this will be red.
Potential pragmatic meanings of yellow light: slow down (if you have time); speed up (if you don't have time or are in a hurry); watch out! (if your breaks are failing); the cars will stop soon (if you're a pedestrian trying to cross); take the picture now! (if you're a photographer wanting pictures of yellow lights).

Symbol has no meaning outside of the language. Yellow color is just a yellow color with no meaning unless it is assigned meaning in particular language. In traffic light language it means one thing. In language of emotions it means another. In language of bees it means something completely different. In language of sunflowers it means again something else.

Look at this: podrig in Irish is Patrick, while in Serbian it means burp. But bud (bod) in both Serbian and Irish mean something hard sticking out like a penis, and are both ancient words for penis.
Podrig sound combination comes from two different languages. Bud (bod) comes from the same language.

In the case of language, there is a huge difference between N sometimes being related to boundaries (well, of course it is, as in the word itself) and being predictably related to them.

I agree. But i believe that i can predict that N in languages where No is used as negation word  conveys meaning of boundary. Boundary defines things. It does not necessarily enclose them as i have already shown to you. Boundary line of any kind can be opened or closed. It is still a boundary which defines something.

Here is definition of the meaning of the word boundary. Tell me how is this different from what i said:

: something (such as a river, a fence, or an imaginary line) that shows where an area ends and another area begins

: a point or limit that indicates where two things become different

boundaries : unofficial rules about what should not be done : limits that define acceptable behavior

http://www.merriam-webster.com/dictionary/boundary

I said: "6. Words and sentences are linked to situations, context. This is how people learn language. And this is why you can not analyze the spoken language in isolation. You always need to link it to the context of which it is an integral part."

And you replied:

Incorrect. See above. There are two different (and equally valid) kinds of analysis.
To say that "N means boundary" you must establish some sort of conventional meaning for N. It's not just about context.

All meaning is linked to context. Every sound that you can recognize is linked to a context in which you heard it the most. Take the sound of bees. If you are about to pick a flower, and you hear bsssss you will know that it is a bee. But if you hear uaiuaiuaiuai you will have no idea what to make of it. Even worse would be if you saw a bee flying and heard no sound. You would conclude that you are gone def. Of if you heard bsssssss coming from a dog instead of the sound of barking. Of if you saw a bee flying and heard uaiuaiuaiuai. You would start questioning your sanity.

Hunter gathers use sounds to identify world around them. This is why i was interested in the sound of summer, as most Indoeuropean names for Sun are S words. Why? I believe that this is because of the sound of summer, sound of your heat which has no sound. You can ask then what about Irish, and grian? Maybe the Irish language developed in the environment where sun did not mean heat originally. Or where sun stopped meaning heat. Like during last Ice age maximum. We still have close - heat - sun - fire association but the root is close, getting warm by huddling. Maybe S words for sun developed in hot places, like north Africa, or Balkans, or Black sea basin, which were worm during last glacial maximum. Or maybe we have newer languages compared to irish?

Regardless, once word is formed it becomes a symbol in it's own right which can have new meaning attached to it through different contexts. But you can still find the original meaning by analyzing the sound blocks from which the word was made.

[dublin]

The fact that I don't know something doesn't mean you're right.

Exactly. Just because you don't know something it does not mean that that something is crazy. And just because you know something it does not mean that that something is valid. It just means that we know different things. Maybe we can cooperate.

I said: "You were asking me for a null hypothesis. This is it: I presume that all the "N" words that I found so far linked to the meaning of boundary in English are all coincidences."

Why can't they be coincidences?

Because for something to be a coincidence from the statistical point of view it has to have high probability to happen as a random event. If you have 30 sounds in a bag, to your disposal, and you need to create 100 words. What is the chance that you will randomly pick the same sound for all 100 words. For one it is 1/30. For hundred it is
(1/30)*(1/30)*(1/30)*(1/30)*...100 times. Do you see how improbable that is? Now calculate probability of the same two sounds being always picked. But if you deliberately pick the same sound, based on the meaning of the sound, then we have a completely different story. We don't have independent events and probability rises dramatically. If the event is me being able to construct the meaning of a word from its sound blocks. The probability of that event is 1/2. I either do it or not. If I do it, again and again, on hundreds of random words from multiple languages, what is the probability of that happening randomly?

I said: "I believe that if I analyze all of the "N" words in English, and large majority of them is linked to the meaning of boundary then i have proven that this is not a coincidence." Your question was:

How do you determine "is linked"? Do you mean that you can make up something then report it to us as a fact? You need a methodology.

No, i use etymological dictionaries, which were written by people who believe in the same things you do and i do, and i find the meaning of the original words. If they are defining something by negation of something else, exclusion from something else, position at the end of something else, surrounded, separated by something else or being the thing that is used to separate, define then the word is relate to boundary. Like words "line", "finger", "one" or "thing"

For example, you could do a survey asking 100 people whether certain words are related to boundaries.

We could. But as I said, we need to look at the original words, with their original meanings. You said open is not related to boundary, because for you boundary and barrier are the same. And they are not. We can have a boundary which is a barrier (fence), boundary which is not a barrier (line), and barrier which is not a boundary (tree). 


I said: "To add to that If I find the same correlation in related European languages, then we certainly have a pattern which is deliberate rather then coincidental." To which you replied:

No, no, no! Don't use related languages to test for lack of coincidence. Use other languages like Swahili, Japanese and Navajo.

I did that with the analysis of the words meaning "one". Correlation is complete. All words split into two groups, which consistently use the same sound blocks to build the meaning of oneness.

Plus, if it's self-evident, great. Why? Common sense is not acceptable (that's not science), so you surely must mean it is easy to reliable detect. How would you train a computer program to detect words related to boundaries?

So you say that common sense is not acceptable, but you suggest that we use random people  for a survey???

Use etymological dictionaries and find the meaning of the original words. If they are defining something by negation of something else, exclusion from something else, position at the end of something else, surrounded, separated by something else or being the thing that is used to separate, define then the word is relate to boundary. The main point is that they convey this message of boundary by themselves. As I said before N is not the only sound that is related to boundary. R can convey meaning of boundary through cutting, S through surface. I did not complete this investigation, far from it. There are some sounds for which I am pretty sure what meaning they convey, and others for which I am not. But I am by now convinced that they all do convey some consistent basic meaning.

I don't know what theories you are rejecting. For example, are you rejecting the theory that Proto-Indo-European existed (somewhere/at some time)?

No I don't. I did say that already. Proto Indo European language is R1a language. Pre Indoeuropean languages are R1a, R1b, E1b, I, J...languages. Mix of R1a language, with all these other languages gave us all Indoeuropean languages which are mix of Pre Indo European languages. This is why we can still find ancient words and sound block patterns.

Can we establish one thing. Do you trust etymological work of all the people who built etymological dictionaries? How did they do their work? By statistics or by eyeballing and data mining? They surely compared and verified each other's findings by more eyeballing and data mining. There are still disagreements on a lot of etymologies. But you do accept their work as valid and scientific? Or you don't. If you do, what is the difference between what they do and what I do to determine meaning of words? If one of you guys wants to determine the etymology of a word for instance "hephaistos" how would you go about doing it? What statistical method could you use to achieve this? I don't think there is one, but please, I am willing to learn.

So once we establish which words have what meaning, we can try sound block hypothesis to see if it works. And we can do probability calculation to see how likely it is that if the hypothesis works n out of m times, that it works by coincidence.

Treat humans as data processing machines. Under what circumstances would they confirm your hypothesis? What would you ask them to do?

I would need to only use humans who understand the meaning of the question. How do you confirm that? By checking if their understanding of the question matches the etymological dictionary understanding of the question.

Question: what is boundary?

if answer is something like: "something (such as a river, a fence, or an imaginary line) that shows where an area ends and another area begins. a point or limit that indicates where two things become different. rules about what should not be done : limits that define acceptable behavior"

Then we have understanding of the question and then we ask them which of these words relates to boundary, which of these words have meaning which is built around the meaning of boundary.

Example: navis - boat in Latin.  na + v + i = na + vo + ide = on + water + persists, continues, goes. Roman navy was built by Dalmatian Illyrians from the Balkans. Word boat probably comes from them, and the roots are still present in Serbian. na - on, voda - water, ide - goes.

What word is most related to summer?
A) Snake
B) Heat
C) Travel
D) Snow

If this is the question most people will say heat. But if question is which of these things possesses sound of summer, or which of these things is only linked to summer, you will have different answer.

Can you rely on existing etymologies for PIE? I thought you were rejecting that theory...

I never said I reject PIE theory. I reject some of it's conclusions, but as I said before, most of it is spot on. I reject the conclusion that there was only one language from which all indoeuropean languages developed. There were many pre indoeuropean languages, but the main was R1a language, whose most direct, least diluted, descendants are Slavic languages. I reject the conclusion that the pre indoeuropean languages have disappeared. They are still here, as part of indoeuropean languages. I also think that mixing started earlier than the 3rd millennium bc, and that some languages groups, like R1a and I and J could have been mixed in Central Europe even before.

One method that may appear counterintuitive but actually works out relatively well is using corpora and looking for words that often appear near each other. That doesn't seem like it would work that well, but it does give you quite a bit of information. A similar approach involves "semantic networks", which use various tools to gather their data (sometimes human surveys, sometimes corpora, etc.).

These are great tools but you can not use them for getting the meaning of a word. Here is an example:

word Beograd. What is the meaning of this word? If you look at words related to it you could, depending on the sample, which is by definition biased, find these words related to it by location:

capital, Serbia, city, big, dirty, polluted, white, friendly, Slobodan Milosevic, NATO, bombing, river, border, party....

What can you conclude from this? What is the meaning of the word Beograd? Meaning is White city, but you will not be able to extract that from the above set. And If we add word Belgrade, Biograd, Belgrad, Beligrad, which all etymologically mean the same, you will end up with statistical chaos, because Biograd is town on the sea, and Belgrade is the English name of Beograd in Serbia but also a town in Montana USA....

Do you see what I mean, when I say that you can not use location based relational statistics to get the meaning of a word?

This thing might sound crazy. But it works, so far. All I can do is continue with trying to prove that it doesn't in large majority of cases. But if it does work in large majority of cases, like say 70%, 80%, 90% of cases, what then? If it is proven that it works but we can't exactly establish why, what then? I saw program on BBC last night about research in the effects of placebos on people. Conclusion is placebos work as well as real medicines. Actually the test for effectiveness of real medicines in now if they work as well as placebos. But because we don't know why they work, we are not allowed to use them. Do you see how idiotic this is?

[Daniel]

I am talking about the origin of the language. Time before books, films, easy travel. Time when communities lived pretty isolated from each other. By the way the same situation remained even in some parts of Europe until mid 20th century. For instance some isolated mountain shepherd communities in Carpathian mountains in Serbia did not know that there was a second world war happening around them. In this type of isolation there is direct link between language and genes. Look at R1a people. They lived between giant uncrossable rivers for thousands of years during the last ice age melt down. Do you think their language during that time was influenced by anyone? But once people start mixing, languages and cultures start mixing.

People have been mixing for as long as there were separate groups. There has been some recent discussion on here (and on the internet in general) about Neanderthal DNA showing up in modern humans.
You're right that there was less travel and so forth, but people did move around a lot, especially over thousands of years.
If you want to maintain your claims here, you should do some research on today's tribal groups living more or less as humans have for many thousands of years. For example, Australian and the (so-called) Khoisan languages offer a perspective on what life might have been like back then. As I understand it, intermarriage is actually a common and important part of those cultures, though certainly not exactly like in Europe or America today.

How did you get that sound meaning correspondences are entirely arbitrary from the above i really have no idea. I said that languages were developed by families, in particular surrounding at different times. These languages are based on particular logic, which can be traced to back to the original sound (not word) language used by the original family, clan...You are right that this is hard to falsify, but just because it is hard to falsify it does not mean it is not possible to do it. Pick a language, check id the same sound to meaning logic works for the majority of  words from that language. Then look at related language. Repeat the same. I would expect that the same logic would work for the words which come from the same family language.

Do you understand what the word "arbitrary" means? It means that it follows a convention that in itself is not motivated/predictable. It does NOT mean it is random.
For example, consider a stoplight. Red means stop. Green means go. That's consistent and not random. But it is arbitrary. It's perfectly imaginable that in some alternate universe the opposite is true.
Therefore, if "dog" in English means "perro" in Spanish, then any sound correspondences must be arbitrary. That doesn't mean "meaningless" or "random", but it means language-specific. This is just like "gl" in English-- it probably doesn't carry the same meaning in Japanese or Zulu.

I am trying to discover the old language(s) of Europe. I am fascinated by the ability of cultural and linguistic traits to survive millenniums. The fact that most people dismiss this as even a possibility, led to so many words in our languages which are "of unknown origin".

You are NOT the first to try to do this. It's very interesting. There has been quite a bit of research on "Proto-World" even. The problem is that the time depth makes this extremely unreliable.
I teach a unit on this in my language history class. There's a video that you might find useful, called "Roots of Language". Unfortunately I haven't found any links to it on youtube or previews, etc. There's certainly a lot you can find on the internet, though.
These are very interesting ideas, but there's a reason we don't know anything for certain: a time depth of over 10,000 years is either impossible or just extremely difficult given what information we have available today. There are so many changes and so much borrowing during that time, it is no longer possibly to rely on the comparative method.
I personally believe it's possible to keep pushing back our knowledge just a bit to see farther into the past, but I don't think it's easy or that something as, honestly, superficial as your analysis will do the trick. You need to have a lot of layers of sound changes and borrowings accounted for.

I believe that we can pretty much reconstruct the language down to last Ice age maximum 10,000 bc.

Which language?

What do you mean by questioning the form of words? Do you ever question the form of any of the words you are reading in this sentence?

That's exactly what I mean. Sound symbolism only works if speakers actively maintain it over generations. If not, they're just using normal non-phonosemantic language! (Whatever the origins might be...)

Languages don't go back to the same family. There were at least three speaking human sub groups before 100,000 bc: Africans, Neanderthals and Denisovans.

That's very controversial. There is evidence to suggest that Neanderthals and possibly others had some form of language, but it's by no means certain.
When I say "human" I'm referring to the ancestors of modern humans, or as you call them "Africans". (Didn't all of those subgroups come out of Africa at some point though??)

I don't agree that we can not find the original correspondences. It is hard, we might not find all of them for all the family languages, but if i find most of them for R1a language, i will be happy. For me this is fun. Is linguistics fun for you?

It's fun for me. Do what you'd like. I still find it incredibly unlikely. You're directly analyzing the modern pronunciations of words that have gone through numerous major sound changes over about 10,000 years. That's a terrible method...
That's like trying to complete a jigsaw puzzle after putting the whole thing through a blender to make the pieces smaller!
Certainly it MAY be possible to reconstruct earlier languages, but if so you will need a more complex methodology than just looking at spelling in the modern languages

I believe that in spoken languages sound, and sound blocks, carry meaning. Sound block is a vocalized consonant like ta, go...These sound blocks were used in conjunction with actions to convey messages. The sounds eventually merged together to form words which replaced this audio visual language with audio language. Each one of these words has it's own meaning. This meaning can change through time, because word becomes a symbol with which people can associate new meanings. So to find the original etymological meaning you need to go to the oldest known form of the word. This is all we have to work with. But it is not as bad as you think.

So how is this any different than etymology in the first place?
Sounds reasonable to me, if you just mean that there were originally sounds that then became part of a language and normal morphological processes followed. This does NOT mean that you can look at modern English and separate out each letter as an independent meaning!!

A lot of words go way back and have survived pretty unchanged, maybe not in English but in languages from which they came to English.

I don't know what you mean by "a lot of words", but I'd say on average, very much not. A few rare sounds might survive, but whole words? Almost never.

You need to go one level lower than that and look at word cold on it's own. The meaning of the sentence "it is cold" is the sum of meanings of "it" "is" "cold". Without cold having it's own meaning, the sentence would be meaningless. "it is ljhlkjow" means nothing to you or me now. We can agree that ljhlkjow means the same as cold. And we can continue using the word ljhlkjow in our own language. If you try to use it with your mother, she will rightly ask you: "what is wrong with using word cold? we all know it, we all understand it. you sound like a weirdo when you use ljhlkjow insted of cold. do you want to sound like a weirdo?" This is how languages are preserved. Through preservation of the meaning of words. If enough weirdos start using ljhlkjow instead of cold, new dialect will emerge. But you know as well as I do, that most people will just continue using cold. But if instead of cold we decide using caold, or colt, we have more chance of developing our own dialect, as the new words "sound like" the old and preserve more or less, the main meaning contained in sounds.

You COMPLETELY missed the point of what I was trying to say. Irrelevantly, yes, you're correct. I was explaining the distinction between semantics and pragmatics. That's what was important there.

Symbol has no meaning outside of the language. Yellow color is just a yellow color with no meaning unless it is assigned meaning in particular language. In traffic light language it means one thing. In language of emotions it means another. In language of bees it means something completely different. In language of sunflowers it means again something else.

Look at this: podrig in Irish is Patrick, while in Serbian it means burp. But bud (bod) in both Serbian and Irish mean something hard sticking out like a penis, and are both ancient words for penis.
Podrig sound combination comes from two different languages. Bud (bod) comes from the same language.

What's your point? Words have meanings and sometimes they are similar in different languages and other times different? Sure. That's precisely what arbitrary means!!

All meaning is linked to context. Every sound that you can recognize is linked to a context in which you heard it the most. Take the sound of bees. If you are about to pick a flower, and you hear bsssss you will know that it is a bee. But if you hear uaiuaiuaiuai you will have no idea what to make of it. Even worse would be if you saw a bee flying and heard no sound. You would conclude that you are gone def. Of if you heard bsssssss coming from a dog instead of the sound of barking. Of if you saw a bee flying and heard uaiuaiuaiuai. You would start questioning your sanity.

Language doesn't work like that. The original usage may be iconic, but soon after it becomes arbitrary/conventionalized. This is Linguistics 101.
Completely context-dependent meaning is not language. Language is conventional/cultural/arbitrary. That's what defines human language!!
At least in modern languages, onomatopoeia accounts for only a very small proportion of the words we come across daily.

Regardless, once word is formed it becomes a symbol in it's own right which can have new meaning attached to it through different contexts.

Yes.

But you can still find the original meaning by analyzing the sound blocks from which the word was made.

No.
You must first find the sound blocks then find the meaning, but that ends up being circular.
Additionally, again, the sound blocks are not present in modern languages. The sounds we observe in modern languages are very distorted from whatever original sounds there were (assuming your hypothesis is correct in the first place).

Because for something to be a coincidence from the statistical point of view it has to have high probability to happen as a random event. ...

Yes, I know that. But you must actually set up a statistical test to see whether these could be by chance. You need to determine what that would look like, and you need to use reliable data as input. Again, focus on objectivity and replicability.

No, i use etymological dictionaries, which were written by people who believe in the same things you do and i do, and i find the meaning of the original words. If they are defining something by negation of something else, exclusion from something else, position at the end of something else, surrounded, separated by something else or being the thing that is used to separate, define then the word is relate to boundary. Like words "line", "finger", "one" or "thing"

Is there anyone else who finds these same correlations? Replicability makes a crazy theory into an accepted one.

We could. But as I said, we need to look at the original words, with their original meanings. You said open is not related to boundary, because for you boundary and barrier are the same. And they are not. We can have a boundary which is a barrier (fence), boundary which is not a barrier (line), and barrier which is not a boundary (tree). 

So you say.

So you say that common sense is not acceptable, but you suggest that we use random people  for a survey???

Yes! Native speakers are by definition experts, and a random sample is statistically desirable.
An experiment may involve common sense from the participants. But you need to show that this is a reliable result, not just something in your head. That's what I keep saying.

No I don't. I did say that already. Proto Indo European language is R1a language. Pre Indoeuropean languages are R1a, R1b, E1b, I, J...languages. Mix of R1a language, with all these other languages gave us all Indoeuropean languages which are mix of Pre Indo European languages. This is why we can still find ancient words and sound block patterns.

Lots of conjecture. Again, write an abstract and tell us exactly what you mean here. How is the idea that Irish and Serbian and very closely related going to be consistent with Proto-Indo-European?

Can we establish one thing. Do you trust etymological work of all the people who built etymological dictionaries? How did they do their work? By statistics or by eyeballing and data mining? They surely compared and verified each other's findings by more eyeballing and data mining. There are still disagreements on a lot of etymologies. But you do accept their work as valid and scientific? Or you don't. If you do, what is the difference between what they do and what I do to determine meaning of words? If one of you guys wants to determine the etymology of a word for instance "hephaistos" how would you go about doing it? What statistical method could you use to achieve this? I don't think there is one, but please, I am willing to learn.

Peer review. Trial and error. On average, these sources are reliable, and the great thing is that you can check the details yourself as needed. Nothing is certain in historical linguistics, but that doesn't mean everything is wrong. On average, there has been a lot of good work.

So once we establish which words have what meaning, we can try sound block hypothesis to see if it works. And we can do probability calculation to see how likely it is that if the hypothesis works n out of m times, that it works by coincidence.

You're still missing the huge step of time depth and sound changes.

Example: navis - boat in Latin.  na + v + i = na + vo + ide = on + water + persists, continues, goes. Roman navy was built by Dalmatian Illyrians from the Balkans. Word boat probably comes from them, and the roots are still present in Serbian. na - on, voda - water, ide - goes.

There's a much simpler explanation:
Latin navis "ship," from PIE *nau- "boat"
Obviously the -is ending is just inflectional, like thousands of other nouns in Latin. Note also that "v" in that is pronounced [w], so this is an example of sound change getting in the way of your claims.
Certainly it is more reasonable to take the etymology from PIE than it is to assume it's from Serbian, right? Just based on time depth...

So again, you're just making things up. I don't know what to do with that. You're welcome to your opinions, but that really is all they are.

If this is the question most people will say heat. But if question is which of these things possesses sound of summer, or which of these things is only linked to summer, you will have different answer.

Seems like sort of a weird question to me, but ok, sure. Go for it. What do people say?
Asking a leading question like that may cause some suspicion, but regardless I'm not convinced even that will work. I think people might still say heat because of sizzling and so forth.

Feel free to actually do some research and let us know what the survey results are though!

I never said I reject PIE theory. I reject some of it's conclusions, but as I said before, most of it is spot on. I reject the conclusion that there was only one language from which all indoeuropean languages developed. There were many pre indoeuropean languages, but the main was R1a language, whose most direct, least diluted, descendants are Slavic languages. I reject the conclusion that the pre indoeuropean languages have disappeared. They are still here, as part of indoeuropean languages. I also think that mixing started earlier than the 3rd millennium bc, and that some languages groups, like R1a and I and J could have been mixed in Central Europe even before.

So you DO reject PIE as it is. You're contradicting yourself.

Do you see what I mean, when I say that you can not use location based relational statistics to get the meaning of a word?

No. You're just rejecting it by suggesting absurd uses. Any tool can be used incorrectly. This is still a reasonable tool, among others.



Still waiting for an abstract.....

[dublin]

djr33

Do you understand what the word "arbitrary" means? It means that it follows a convention that in itself is not motivated/predictable. It does NOT mean it is random.

arbitrary ˈɑːbɪt(rə)ri/Submit adjective

based on random choice or personal whim, rather than any reason or system.
"an arbitrary decision" antonyms:   rational, reasoned

For example, consider a stoplight. Red means stop. Green means go. That's consistent and not random. But it is arbitrary. It's perfectly imaginable that in some alternate universe the opposite is true.

This color scheme derives from a system used by the railroad industry since the 1830s. At this time, railroad companies developed a lighted means to let train engineers know when to stop or go, with different lighted colors representing different actions.  They chose red as the color for stop, it is thought, because red has for centuries been used to indicate danger. For the other colors, they chose white as the color for go and green as the color for caution.

The choice of a white light for go turned out to cause a lot of problems. For instance, an incident in 1914 where a red lens fell out of its holder leaving the white light behind it exposed. This ended with a train running a “stop” signal and crashing into another train. Thus, the railroad decided to change it so the green light meant go and a caution “yellow” was chosen, primarily because the color is so distinct from the other two colors used.

http://www.todayifoundout.com/index.php/2012/03/the-origin-of-the-green-yellow-and-red-color-scheme-for-traffic-lights/

So nothing random there. It was a premeditated decision, based on thinking and planning, making a mistake, then realizing mistake, correcting the mistake. This is how humans learn and this is how humans invent things. This is what makes humans different from monkeys.

Therefore, if "dog" in English means "perro" in Spanish, then any sound correspondences must be arbitrary. That doesn't mean "meaningless" or "random", but it means language-specific. This is just like "gl" in English-- it probably doesn't carry the same meaning in Japanese or Zulu.

As you can see you can't use word arbitrary if you don't imply random original choice. And as I have shown you, the probability of randomly choosing the same sound in 100 words with related meaning is close to 0. Different genetic human groups, with different ability to hear and repeat sounds also have different logical ability. Latest neuroscience research shows that these things are directly linked. Sounds could be linked to such old and such deeply rooted contexts, that they carry the same meaning in most languages. Example are words for one, which you completely ignored because they prove my hypothesis that sounds carry consistent meanings. With words things are different. They are complex meaning matrices. We don't know what people tried to convey when they created word dog. What characteristics of dogness did they choose to describe. Originally there were only wolves, and as I have shown you, names for wolves in R1a European languages are onomatopoeic. This is normal, because this is where wolves lived and where they were first domesticated into dogs:

The origin of the domestic dog (Canis lupus familiaris) began with the domestication of the Grey Wolf (Canis lupus) several tens of thousands of years ago.[1][2][3] Genetic and archaeological evidence shows that humans domesticated wolves on more than one occasion, with the present lineage of C. l. familiaris arising at the latest 15,000 years ago as evidenced by the Bonn-Oberkassel site and possibly as early as 33,000 years ago as evidenced by the mtDNA testing on a paleolithic dog's remains from the Razboinichya Cave (Altai Mountains).

http://en.wikipedia.org/wiki/Origin_of_the_domestic_dog

The people who lived with wolves, continued to call their domesticated wolves wolves minus howling. This is why we have

Serbian - Vuk,Volk - wolf; Cu - dog
Irish - Voulcu - wolf, cu - dog
Germanic - wolfaz - Hund - dog

Words like Slavic pas, Spanish Perro, English dog are later developments, and are describing maybe different characteristic of a dog species, usage of dogs, or something completely different.

Let me give you an example of how sound block meanings are preserved in different languages:

Word snow

http://en.wiktionary.org/wiki/snow

Serbian - sneg = s + ne + ga = with, surface, soil + no + point, see
English - snow = s + no + vi = with, surface, soil + no + see, know (Arian ved, vid - to know to see)
Proto-Indo-European - *snóygʷʰos = s + no + je + go + s = with, surface, soil + no + is + point, see
Spanish - nieve = ni + je + ve = no + is + see
Proto-Italic - *sniks = s + ni + ka(ga) + s = with, surface, soil + no + point, see
Czech - sníh = s + ni + h(g)a =  with, surface, soil + no + point, see

All these Words come from R1a language, and were invented by people who wanted to convey the same message: we can't see soil, surface any more.

You are NOT the first to try to do this. It's very interesting. There has been quite a bit of research on "Proto-World" even. The problem is that the time depth makes this extremely unreliable. These are very interesting ideas, but there's a reason we don't know anything for certain: a time depth of over 10,000 years is either impossible or just extremely difficult given what information we have available today. There are so many changes and so much borrowing during that time, it is no longer possibly to rely on the comparative method.
I personally believe it's possible to keep pushing back our knowledge just a bit to see farther into the past, but I don't think it's easy or that something as, honestly, superficial as your analysis will do the trick. You need to have a lot of layers of sound changes and borrowings accounted for.

First, since when did difficult stop people from trying? If it did we would be still hanging off trees. Second, why is my analysis superficial?

Superficial - appearing to be true or real only until examined more closely. Did we do the examination which proved that my analysis is wrong? No we didn't. I have given you null hypothesis. Neither you nor me nor anyone else here have proven the null hypothesis. Therefore, according to you my hypothesis is still not invalidated.

One question about this whole invalidation stuff.

Hypothesis: Penicillin kills bacteria.
Null hypothesis: Penicillin does not kill bacteria.

According to you any hypothesis is worthless until there is a null hypothesis that can invalidate it. And once null hypothesis is proven, the original hypothesis becomes invalidated and therefore wrong, useless.

Penicillin null hypothesis has been proven by scientific research. Penicillin does not kill all bacteria, and even bacteria it once killed can develop resistance to it. But we still use penicillin. Shouldn't we have discarded penicillin as worthless, useless because null hypothesis was proven and original hypothesis was invalidated?

Actually no. Because penicillin still works in some cases. It kills some bacteria. For other bacteria, we have found other antibiotics which kill them. If the job is "kill bacteria", we use all the available tools. People use what works.

But this is not what you think should happen? Or did I not understood you correctly?

If we have the same situation in physics for instance, should we discard the original hypothesis as wrong just because null hypothesis has been proven? Does that invalidate the original hypothesis. According to you this is exactly what should happen. But according to modern physics, this is not what happens. We use what works. If Newton laws were found not to work in macro and micro world, does does that mean we are not using them any more in our middle world? No. We say, Newton laws work in these cases, but not in these, and we continue to use them. They work for what we need them. But according to you Newton's laws are worthless, they have been invalidated, because null hypothesis has been proven. It is exactly with your favorite example of scientific certainty, which does not work in the same way everywhere and under all conditions, and in micro world it does not work at all. But we still use gravity based calculations because they do the job even though the null hypothesis has been proven and our understanding of gravity has been proven wrong. Obviously what you think is the way science works is not how science actually works. Science exists so it can help us find, define and understand new things, undefined things, unknown things. Science works on the boundary of what is accepted, and a lot of times way outside of that boundary. This is how science expands our knowledge. Scientists are not afraid to look at new things no one looked at before. Or to try things others tried and failed. This is the job of scientists. Are you a scientist? What was the last new thing you investigated and what is the last hypothesis you proposed?

So based on the way science works, my hypothesis of meaning carrying sound blocks works as it is doing the expected. It produces the meaning of words from sound blocks in a consistent way. Does it mean it will work for every word? I don't know. We need to do a lot more research to find that out. Does it mean that if we find words for which the hypothesis doesn't work it is immediately invalidated? No it doesn't. It means that we found exceptions to the rule. Maybe we have to redefine the rule, put boundaries for it. Find out what makes the rule work for some case and not for other. We might discover something new by doing that. But that is how scientists would think, which is not how you think.

When I said that I believe that we can pretty much reconstruct the language down to last Ice age maximum 10,000 bc, you asked: "Which language"?

Any of the existing languages which existed at that time. I am particularly interested in old European languages, R1a and I,J,G...We can quickly start seeing how words group based on logic which was used to assemble words from sounds. We then relate that to existing representative languages from each genetic family: East Slavic for R1a, Gaelic, Basque for R1b, Central European languages (south, west Slavic, Germanic, Norse) for I...We see if there is correlation, logical grouping. It is not precise, but it is the best we have.

That's exactly what I mean. Sound symbolism only works if speakers actively maintain it over generations. If not, they're just using normal non-phonosemantic language! (Whatever the origins might be...)

This does not mean that the meaning is gone. If you compare multiple languages, if you look at how word changed, the original meaning starts to emerge. Not always, but enough times to give us something to work with.

When I said:

Languages don't go back to the same family. There were at least three speaking human sub groups before 100,000 bc: Africans, Neanderthals and Denisovans.

You replied:

That's very controversial. There is evidence to suggest that Neanderthals and possibly others had some form of language, but it's by no means certain.

When I say "human" I'm referring to the ancestors of modern humans, or as you call them "Africans". (Didn't all of those subgroups come out of Africa at some point though??)

Welcome to real science. Once things become certain, they are given to teachers...

When I say "human" I'm referring to the ancestors of modern humans, or as you call them "Africans". (Didn't all of those subgroups come out of Africa at some point though??)

We don't know for sure. Latest archaeological data from Georgia, are forcing us to completely rethink our understanding of human development. Even if people originally developed in Africa, the sub groups I am talking about lived separately for so long, that they became genetically different. By the way there is firm proof that Africans, Neanderthals and Denisovans mixed. There is even Neanderthal DNA found in east African tribal people. I would expect that languages of east Africa share some traits with languages of Europe. Also Central African languages. Because R1b genetic influence, i would expect these languages to share traits with other R1b languages, like gaelic and basque.

I said:

I don't agree that we can not find the original correspondences. It is hard, we might not find all of them for all the family languages, but if i find most of them for R1a language, i will be happy.

You said:

I still find it incredibly unlikely. You're directly analyzing the modern pronunciations of words that have gone through numerous major sound changes over about 10,000 years. That's a terrible method...Certainly it MAY be possible to reconstruct earlier languages, but if so you will need a more complex methodology than just looking at spelling in the modern languages

I am not only looking at spelling or even pronunciation of modern languages. I look at words from modern languages, then i compare them with their etymologies and older versions of the words. Then I compare them with words from other languages from cultures where similar anthropological cultural, ethnological features were found. You find sound block meanings at the cross between all this. It is extremely complex work.
But once you find the meaning (or set of related meanings) of a sound block, you test it against modern languages. How can we do that? Because sound blocks are still there. Some have changed within the same undifferentiated group, like g - k, some were lost, new were added. But all these changes are done in such way as to preserve the original meaning of the word. If the meaning changes it changes to related meaning. gay - gentle, gey - homosexual. This has to happen if you want to preserve language as a tool for conversation.
Of course you can completely change the meaning of words, but that is an exception, not the rule.

I said:

I believe that in spoken languages sound, and sound blocks, carry meaning.

You replied:

So how is this any different than etymology in the first place?
Sounds reasonable to me, if you just mean that there were originally sounds that then became part of a language and normal morphological processes followed.

Let me try to explain using an example we already looked at:

I said:

navis - boat in Latin.  na + v + i = na + vo + ide = on + water + persists, continues, goes. Roman navy was built by Dalmatian Illyrians from the Balkans. Word boat probably comes from them, and the roots are still present in Serbian. na - on, voda - water, ide - goes.

To which you replied:

There's a much simpler explanation:
Latin navis "ship," from PIE *nau- "boat"
Obviously the -is ending is just inflectional, like thousands of other nouns in Latin. Note also that "v" in that is pronounced [w], so this is an example of sound change getting in the way of your claims.
Certainly it is more reasonable to take the etymology from PIE than it is to assume it's from Serbian, right? Just based on time depth...

This shows that you don't understand what you are talking abut. You are just quoting things.

First time depth. Illyrians lived in the Balkans in the exact same area where you find South Slavs today. They lived there in the second millennium bc. And built boats. This is at least 500 years before Rome was founded. Illyrians lived in Italy as well. Illyrians built and manned Roman navy.

Serbian language is built on top of all the languages of all the people who lived in the Balkans and whose mix produced Modern Serbian. So Illyrian sound roots could be (and as you can see are) present in Serbian language.

There's a much simpler explanation: Latin navis "ship," from PIE *nau- "boat"...This just shows the extent of your intellectual laziness. If you have bothered, and did some more reading you would have found this:

From Proto-Indo-European *néh₂us, cognate with Ancient Greek ναῦς (naus, “ship”) and Sanskrit नाव (nāva, “ship”). Usually described as derivation from *(s)neh₂- (“to swim”).
Gamkrelidze and Ivanov argue that it's borrowing from Semitic *ʾunw(at)- (“jar, vessel; boat”).

Armenian: Old Armenian: նաւ (naw), նաւազ (nawaz, “skipper”)
Celtic: Old Irish: nau
Germanic: Old Norse: nór
Hellenic: Ancient Greek: ναῦς (naus)
Indo-Iranian: Indo-Aryan: Sanskrit: नौ (nau), नाव (nāva)
Iranian: Avestan: (nauuāza-, “skipper”)
Khotanese: (no)
Mazanderani: نو (no)
Old Persian: (nāva, nom./acc. pl.)
Middle Persian: *nʾw (nāw)
Manichaean Middle Persian: nʾwʾz (nāwāz, “skipper”)
Persian: ناو (nâv)
Ossetian: Digor: науӕ (nawæ)
Iron: нау (naw)
Parthian: nʾwʾz (nāwāz, “skipper”)
Italic: Latin: nāvis
Illyrian: Nauna, Nauportus (toponyms)

First, using your time depth order these languages according to their age. You see that in the oldest records it is naV, naUV which later became naua, nau, no. So V was lost because people could not pronounce it properly and eventually dropped it. But the meaning was preserved. 

Na - on
U - in
V - water, white, clear, transparent, not hard, not material
I - continue, persist, action, movement

Proto-Indo-European *néh₂us = n(a) + (j)e + (g)h + u(v) + s = on + is + water, in +  surface = thing you get into to stay on the surface or water. But this is reconstructed word, not real word. Look at real words:

Ancient Greek ναῦς (naus, “ship”) = na + u(v) + s = on + water, in + surface
Sanskrit नाव (nāva, “ship”) = na + va = on + water
Semitic *ʾunw(at)- (“jar, vessel; boat”) = u + n(a) + v = in + water + on
Old Armenian: նաւ (naw) = na + v = on + water
Old Persian: (nāva, nom./acc. pl.) = na + va = on + water
Avestan: (nauuāza-, “skipper”) = na + u + ua (v) + s = on + in + water + surface

The problem is that you don't believe in eyeballing (looking), and data mining (reading, learning). These are dirty words for you modern linguists. So you are walking around blind repeating stuff you managed to memorize during your studies. You are accusing me of ad hominem argumentation style. For those who don't speak the lingo, ad hominem means:

An ad hominem (Latin for "to the man" or "to the person"[1]), short for argumentum ad hominem, is a general category of fallacies in which a claim or argument is rejected on the basis of some irrelevant fact about the author of or the person presenting the claim or argument

Now I don't believe that you not wanting to see, hear or think is "irrelevant fact" about you. I believe that if you want to be a linguist, you have to be able to see, hear and think. You not being able to see the above makes me believe that you are either unable to see, hear or think, or that you don't want to. I am feeling more and more like I am tacking with a tape recorder, not a human being.

So do you see the difference between Etymology as you see it, and Etymology as I see it.

Current understanding of etymology: Proto-Indo-European *néh₂us

The question no one is asking is why *néh₂us and not babarat? I did ask that question and  the result of my investigation is my sound block hypothesis. And here it is used to answer the above "why" question:

*néh₂us = n(a) + (j)e + (g)h + u(v) + s = on + is + water, in +  surface = thing you get into to stay on the surface or water.

To your shouting (capital letters, exclamation marks...):

This does NOT mean that you can look at modern English and separate out each letter as an independent meaning!!

I can say: actually I can, successfully.

[lx]

http://www.todayifoundout.com/index.php/2012/03/the-origin-of-the-green-yellow-and-red-color-scheme-for-traffic-lights/

So nothing random there. It was a premeditated decision, based on thinking and planning, making a mistake, then realizing mistake, correcting the mistake. This is how humans learn and this is how humans invent things. This is what makes humans different from monkeys.

Having a lens falling out of a light, turning the one that was red (white light + red filter lens) confusing people is absolutely no counterargument whatsoever to the point that was being made. If the bulb instead of the filter had been red, then that issue would not have happened and there would be no problem. It's still arbitrary.

So based on the way science works, my hypothesis of meaning carrying sound blocks works as it is doing the expected. It produces the meaning of words from sound blocks in a consistent way. Does it mean it will work for every word? I don't know. We need to do a lot more research to find that out

You just seem to lack the fundamentals of what scientific analysis requires. You're allowing your original data that gave you the idea to support the conclusion that you drew from that data! That's what even school children are taught is a big no, no. You're completely right that you need to do a lot more research. You don't know it's going to work for every word, but you haven't even tried.

Here's how you have to think of it. You need to write down a set of instructions, and have an imaginary bucket of words to investigate. You need to be able to give someone the instructions on how to analyse the words, all the hypotheses, and then give them the bucket. If you can draw the conclusions and argue the same case for an unseen set of data (in a replicable way) then you have an argument that will cause people to stop ignoring the babbling and it will actually compel them to allow for the possibility and investigate what you're questioning.

There's no method, there's no abstract. You have some words that have given you a conclusion and you're firing it back to us as if it is evidence of the original claim. That is the definition of a circular argument. Anyone, anyone, who works in science or with the scientific method can see that a mile away.

I could take the well-understood phono-semantic feature of English of word-initial sn- clusters having a high correspondence with things relating to the nose, and then form a conclusion this is the case for every language. Sneeze, snore, sniff, snivel, snigger, snooze, snuffle, sneer, snob, snout would be some good examples, and then to argue my case that this is the truth, I would quote those same things. It might seem logical to you, but it's fundamentally flawed when it comes to making generalised claims.

[lx]

Latest neuroscience research shows that these things are directly linked.

I must insist you produce this evidence that supports your claim that:

"Different genetic human groups, with different ability to hear and repeat sounds also have different logical ability."

[lx]

As you can see you can't use word arbitrary if you don't imply random original choice. And as I have shown you, the probability of randomly choosing the same sound in 100 words with related meaning is close to 0

Aha. You think you've shown us something. You think you've actually shown that in 100 words a possibility of 0 can be drawn. You ACTUALLY think you've shown us this? Without any experiment? With carefully cherry-picked data?

This is why this discussion isn't going anywhere. We're talking at different levels. You're not the first person to come to a forum like this with an extraordinary claim and we end up going in circles. At the end of the day it comes down to the person with the extraordinary claim not understanding how to put together credible data. I might have more of an attention span if I hadn't seen this scenario repeat itself so many times before. Asking for an abstract-style description is really not hard when the idea is well thought out. We've been asking for pages and nothing has arrived. It's just dodging questions and trying to throw facts in the faces of people trying to get to the sound logical basis of the argument. It might manage to distract people for a little while, but you can't throw unrelated facts to a point where people accept that the core basis of a theory starts to make sense.

[dublin]

Lx

They chose red as the color for stop, it is thought, because red has for centuries been used to indicate danger.

Premeditated.

they chose white as the color for go and green as the color for caution.

Premeditated.

The choice of a white light for go turned out to cause a lot of problems. For instance, an incident in 1914 where a red lens fell out of its holder leaving the white light behind it exposed. This ended with a train running a “stop” signal and crashing into another train. Thus, the railroad decided to change it so the green light meant go and a caution “yellow” was chosen, primarily because the color is so distinct from the other two colors used.

Premeditated corrective action.

What is random about this?

You just seem to lack the fundamentals of what scientific analysis requires. You're allowing your original data that gave you the idea to support the conclusion that you drew from that data! That's what even school children are taught is a big no, no. You're completely right that you need to do a lot more research. You don't know it's going to work for every word, but you haven't even tried.

You have no Idea what my original data was. I tried to explain it to you guys many times, you are unwilling or unable to read it. The data I used to arrive to my sound block hypothesis, was based on interdisciplinary data including linguistic, historical, anthropological, ethnographic, archaeological, genetic...The word corpus I used was originally only Serbian (South Slavic languages) and Irish. Then I added Sanskrit. I only started looking at Latin and Greek when I started looking into the ancient European Gods and what happened to them when Greeks and Romans got their hands on them. The extent of my language analysis of Greek and Latin was narrowly concentrated on religion and religious beliefs and words used to describe them. I did not even look at English until I started writing here. So the fact that my hypothesis works on English, German, Spanish, Latin, just proves that the hypothesis is valid.

You can believe me or not. It's your choice.

Here's how you have to think of it. You need to write down a set of instructions, and have an imaginary bucket of words to investigate. You need to be able to give someone the instructions on how to analyse the words, all the hypotheses, and then give them the bucket. If you can draw the conclusions and argue the same case for an unseen set of data (in a replicable way) then you have an argument that will cause people to stop ignoring the babbling and it will actually compel them to allow for the possibility and investigate what you're questioning.

I did that already in my last two posts. Plainly, use sound blocks I already used in my analysis. Find 100 random words from this thread, or any random page on the internet, which are built around these sound blocks. Find root words using etymological dictionary. Give the list of modern words and root words to someone with the list of sound blocks and their meanings. See if sum meaning of the original words is close to the meaning of the original word and of the modern word.

As I said already my sound analysis is not complete. I did not get to the root meaning of all the sounds. But the ones i used here to analyze words I am pretty sure about.

There's no method, there's no abstract. You have some words that have given you a conclusion and you're firing it back to us as if it is evidence of the original claim. That is the definition of a circular argument. Anyone, anyone, who works in science or with the scientific method can see that a mile away.

You have no idea what scientific method is, and what science is. I have given you the method, go and use it if you want. And as I said you are wrong about what data I used to get to my hypothesis.

I could take the well-understood phono-semantic feature of English of word-initial sn- clusters having a high correspondence with things relating to the nose, and then form a conclusion this is the case for every language. Sneeze, snore, sniff, snivel, snigger, snooze, snuffle, sneer, snob, snout would be some good examples, and then to argue my case that this is the truth, I would quote those same things.

You could do that, but you would not be right. And that would not be what I would claim either, because as I said many times, but you have failed to notice, different languages would be created by people who hear sounds differently, reproduce sounds differently and have different data to meaning processing logic. So languages are expected to be different. Did you read anything I said about Epigenetics?

One big difference between you an me is that you don't know why all these words have S and N in them. Do you want to know? This is from page 8 of this discussion. Again i am telling you that I did not look at English at all until I came to this forum. I said in my original post that I arrived to my conclusions comparing Irish and Serbian:

Let's go back to sound "N" and related words. I also believe that part of this post will show my friend freknu that some things sound funny because you don't know enough about them.

If you inhale through your nose with your mouth slightly opened you get sound "N".
If you inhale through your nose with your mouth closed and you get sound between "N" and "M" depending on the position of your tong.

Nos - Nos, a boundary point of our face. Also sound produced by sNiffing is "N". So NOS = N + O + S = sound N + hole + with = the hole that makes N sound.


The fact that both "N" and "M" sounds can be made through our noses, led to all "N" and "M" words related to nose and smell:

English, Serbian, Irish

Nos - nos - srón
Nostril - nozdrva - cuinneán, srón-pholl
sNout - njuška - smuit
sNaffle, snavel - kljun - ghob, gop, guilbend, gulba, gulban
sNiff - njušiti, šmrkati - snaois
sNivel - cmizdriti, from smizdriti
sNeeze - kijati - sraoth
sNite - blow your nose - izduvaj nos - srón séideadh
sNooze - dremati - shuan
sNore - hrkati - roncaim
sNuff - onjušiti - snaois
sNoop - cunjati, Njuškati - snaois
sNot - slina - ronn, smuga
sNigger - smeškati se
sNeer - podsmehnuti se
sNort - šmrkati, frktati

Smell - Miris - boladh
smell bad, stink - smrdi - brént, mosach
smacking - mljacanje - blastarnach (something smells nice, soliva starts rushing in our mouth, we smack to collect and swallow it. Sound of that action)

mmmmmmmmmm - sound when something smells, and as our experience tells us also tastes nice


Have a closer look at all these words above. The sound they all have apart from "N" is sound "S". As a matter of fact, most of the English words describing things related with nose start with letter "S" followed with letter "N". Why?

I must insist you produce this evidence that supports your claim that: "Different genetic human groups, with different ability to hear and repeat sounds also have different logical ability."

Read about the studies that deal with how hearing impediments affect our ability to speak and our intelligence. You have google, do some data mining and eyeballing...

I said: "As you can see you can't use word arbitrary if you don't imply random original choice. And as I have shown you, the probability of randomly choosing the same sound in 100 words with related meaning is close to 0"

And you said:

Aha. You think you've shown us something. You think you've actually shown that in 100 words a possibility of 0 can be drawn. You ACTUALLY think you've shown us this? Without any experiment?

You see if you actually read what I write, you would have seen this:

For something to be a coincidence from the statistical point of view it has to have high probability to happen as a random event. If you have 30 sounds in a bag, to your disposal, and you need to create 100 words. What is the chance that you will randomly pick the same sound for all 100 words. For one it is 1/30.
For hundred it is (1/30)*(1/30)*(1/30)*(1/30)*...100 times. Do you see how improbable that is? Now calculate probability of the same two sounds being always picked. But if you deliberately pick the same sound, based on the meaning of the sound, then we have a completely different story. We don't have independent events and probability rises dramatically.
If the event is me being able to construct the meaning of a word from its sound blocks. The probability of that event is 1/2. I either do it or not. If I do it, again and again, on hundreds of random words from multiple languages, what is the probability of that happening randomly?

This is plain probability theory. If you say that people choosing sounds for words is not the same like picking sound cards from a bag, you need to answer why is it not the same? Maybe because it is in some way predetermined by our genes? Or our experiences which are stored using epigenetic switching? If it is the same, and our minds are just mindless random sound generators, which sounds completely mad by the way, then the probability of choosing the same sound in 100 boundary related words is the probability of the same outcome of 100 random unrelated events. But if the events are related, like for instance we realize that people understand that N means boundary, and we continue using N every time we want to build a word related to boundary, then....We are not mindless machines or monkeys. War and Peace will never be written by monkeys with typewriters or computers.  But you guys are trying to treat the most human of all human things as random set of symbols to which we attach random meaning and then happily continue to use that symbol for thousands of years.

This is why this discussion isn't going anywhere. We're talking at different levels.

What exactly do you mean by this? Do you see yourself as superior in any way to me? Show me, using arguments, where I am wrong. Let others judge you. I gave you plenty of opportunity to take apart any of my arguments by proving that the examples I gave are wrong. Take all words that mean "one" and explain the correlation between the sounds and meaning in all these words in your own way. Show us all how your argument is superior to mine. And don't use "cherry picked data" as an excuse not to do it, because that just shows how week your own hypothesis is, if you have one, when it doesn't work on any of my "cherry picked" examples. Or maybe not every theory works on everything?

At the end of the day it comes down to the person with the extraordinary claim not understanding how to put together credible data.

What is credible data to people who believe that you can use statistics to arrive to etymology of words and deplore eyeballing and data mining and cherry picking? And yet all etymological dictionaries were written by people who used eyeballing and data mining and cherry picking as their methods. The whole idea of PIE was derived from that data. I asked already but got only mumble mumble answer:

Can we establish one thing. Do you trust etymological work of all the people who built etymological dictionaries? How did they do their work? By statistics or by eyeballing and data mining? They surely compared and verified each other's findings by more eyeballing and data mining. There are still disagreements on a lot of etymologies. But you do accept their work as valid and scientific? Or you don't. If you do, what is the difference between what they do and what I do to determine meaning of words? If one of you guys wants to determine the etymology of a word for instance "hephaistos" how would you go about doing it? What statistical method could you use to achieve this? I don't think there is one, but please, I am willing to learn.

Asking for an abstract-style description is really not hard when the idea is well thought out. We've been asking for pages and nothing has arrived.

Actually I have written the summary already, in short and concise way. It easy to do it because the idea is very simple. I could not be bothered to repeat my self again, so here is a link to my post in which I have given my summary. Read it.

http://linguistforum.com/historical-linguistics/the-language-of-old-europe/msg1198/#msg1198

I have expanded on it in great detail through answers to your questions. Read them too. Ask additional questions.

This is not what I am working on and which is late. What I am writing is unified theory of language which is something I am much more interested in. And that takes time.

[lx]

Actually I have written the summary already, in short and concise way. It easy to do it because the idea is very simple. I could not be bothered to repeat my self again, so here is a link to my post in which I have given my summary. Read it.

Done.
Let me quote jkpate in the post that followed that one:

Also, since you haven't actually done these tests, why do you expect people to believe you?

I'm happy to believe you think you've got the answer. Send your hypotheses to some linguistics journals. Let us know what they say. Failing that, try sending the idea to one credible linguist and promise to quote the honest response he/she gives you.

[dublin]

Lx

Done.

Great. So now you can all stop asking for summary again and again.

Also, since you haven't actually done these tests, why do you expect people to believe you?

Hopefully now I will have time to do these tests.

I'm happy to believe you think you've got the answer.

This is not what I expect from you. I expected this from all of you (I will repeat it):

Show me, using arguments, where I am wrong. Let others judge you. I gave you plenty of opportunity to take apart any of my arguments by proving that the examples I gave are wrong. Take all words that mean "one" and explain the correlation between the sounds and meaning in all these words in your own way. Show us all how your argument is superior to mine. And don't use "cherry picked data" as an excuse not to do it, because that just shows how week your own hypothesis is, if you have one, when it doesn't work on any of my "cherry picked" examples.

I did not expect this:

Send your hypotheses to some linguistics journals. Let us know what they say. Failing that, try sending the idea to one credible linguist and promise to quote the honest response he/she gives you.

Are you not credible linguists? I am wasting my time here then? What is the point of having this discussion board at all?

[Daniel]

Are you not credible linguists? I am wasting my time here then? What is the point of having this discussion board at all?

We've told you that we don't believe you, that your arguments are not convincing, that your methodology is flawed, and that your theories are almost certainly incorrect. This has not discouraged you.
What lx was trying to say is: can you find even one credible person (here or elsewhere) who agrees with you?

Let's say you emailed Noam Chomsky and he took the time to respond (or anyone else). What do you think he would say?

Having an unpopular opinion is fine. Having a theory that absolutely no one believes is probably a sign that something is off.


Are you wasting your time here? Probably, because you aren't arguing convincingly or any closer to convincing us than when you showed up (in fact, I'm a lot less likely to be convinced now because you clearly have no concept of how to, for  example, gather data and run experiments before making claims!).


One way to spend your time here more productively would be to discuss less controversial topics, especially those that either 1) you truly have the data and results to back up at this time, or 2) you want to have us inform you of something you don't know.
For example, it might be helpful for you and interesting for us to discuss the etymologies of a few words that interest you.

But, no, we're not getting anywhere with you just going on and on without any supporting evidence. Even assuming you are correct, there's no point in trying to convince us by continuing like this. Work out a more convincing argument if you'd like.

[dublin]

One way to spend your time here more productively would be to discuss less controversial topics...like if you want to have us inform you of something you don't know.
For example, it might be helpful for you and interesting for us to discuss the etymologies of a few words that interest you.

You are the most arrogant person with the least reason to be so. You are the last person that I would ask for any clarification on linguistic matters.

[Daniel]

Good luck!