46
« on: May 18, 2017, 11:22:59 PM »
(I will try to explain my ideas without presupposing that a reader has a lot of knowledge of the Indo-European linguistics, regardless of how long it would take to explain them.)
As anyone who is even remotely familiar with Indo-European linguistics knows, it’s relatively easy to reconstruct the consonants in the proto-language. The same, however, isn’t even remotely true for the vowels. To understand the problem, consider this: let’s try to reconstruct the Proto-Indo-European word for three. I happen to know its pronunciation in three distantly related Indo-European languages: Latin tres, English three and Serbo-Croatian tri. The first phoneme was quite obviously *t, English th is easily explainable via the Grimm’s law. The same goes for the second phoneme, it was *r. Now, the third phoneme is a bit trickier, but nevertheless remains deducible. Latin word suggests the phoneme *e, and English i sound is a result of the Great Vowel Shift, therefore it also suggests an original *e sound. However, Serbo-Croatian clearly suggests there to have been an *i sound. There is no rule that would make an e sound in other Indo-European languages correspond to an i sound in Serbo-Croatian, Serbo-Croatian words for six and seven are šest and sedam, not **šist and **sidam. However, here is a clue, the Ancient Greek word is treys. Now it makes sense. It was a diphthong! Loss of the final s in Croatian is easily explainable, in Proto-Slavic there was a so-called Open Syllable Law, and no syllable could end in a consonant. Syllable-final consonants either became initial (the liquid consonants r and l) or were elided. That’s why no native Serbo-Croatian word ends with an s. The English word can probably be explained in a similar manner, so that the original Proto-Indo-European word was *treys. Can we reconstruct the Proto-Indo-European word for two? Well, not that easily. So, the English word is two, the Latin word is duo and the Serbo-Croatian word is dva. So, the first phoneme was *d. English t can be explained as a simple application of the Grimm’s law. The second phoneme was *w. Serbo-Croatian v and Latin u are both its allophones. But let’s try to deduce what the third phoneme was. The English u sound comes from, via the Great Vowel Shift, from the *o sound. Latin also points to the original *o sound. So does the Ancient Greek. But how to explain the Serbo-Croatian a sound? It’s not like Latin o corresponds to Serbo-Croatian a, Latin word for eight is octo, and Serbo-Croatian word for eight is osam and not **asam. It’s as if there had originally been two words for two: *dwo and *dwa. Most of the linguists agree with that notion, and believe that change of the vowels, called ablaut, was the main way to derive new words in Proto-Indo-European. Its traces are visible in modern Indo-European languages. For instance, many English irregular verbs have their forms formed by change of the vowel, for instance: sing-sang-sung. There are languages today that have this feature as the main way to derive new words, for instance, the Semitic languages (that are not Indo-European, and are probably not related to Indo-European languages at all). However, there would appear to have had been many types of ablaut in Proto-Indo-European, way more than in those languages. So, it’s been hypothesized that there had been some phonemes in Proto-Indo-European that colored the vowels (much like the r colors them in modern English, a is pronounced significantly differently in fat than in far). So, the hypothesized phoneme that colored e to a, but did nothing to other vowels is noted as *h2. So, the Proto-Indo-European word for two is then reconstructed as *dwoh2. So, only one type of ablaut is hypothesized to exist, the alternation between o, e and no vowel. If an ablaut occurs that turns *dwoh2 into *dweh2, the word would, in modern languages, be rendered as if it had been pronounced *dwa. This bears a lot of explanatory power. But how were those sounds pronounced? Most of the linguists believe they were h-like sounds, and the theory about them is called the Laryngeal Theory. This is where I don’t agree with the mainstream linguistics any more. I think that those sounds were, in fact, semi-vowels (sounds like the consonantal y and w). I believe that, for instance, the often reconstructed cluster *eh2 was, in fact, a diphtong, usually pronounced like i in ride. To understand why, consider this example: the Proto-Indo-European word for mother is reconstructed as *meh2ter. The Latin word was pronounced mater, the Greek word was pronounced meter, and the Sanskrit word was pronounced mitar. The a in the Latin word mater is explainable the same way the e can be explained in tres, the same goes for the Sanskrit mitar, and the Greek word is explained via the analog monothongization as ae (pronounced like i in ride) in Classical Latin turned to e in the Romance Languages. Or consider the Indo-European word for beech, *bheh2gjos. The Latin word from it is fagus (bh turns to f in Latin), the English word from it is book (Old Germanic people used to write on the beech wood, beech comes from umlaut of book + the ending e that caused the palatalization), and the Serbo-Croatian word is bukva. The simple truth is, by regular sound changes, the English word would be **bak (Grimm’s law) and the Serbo-Croatian word would be **boz (satemization, *eh2 almost always turns to o in Serbo-Croatian). But let’s suppose *eh2 was here pronounced like ow in bow. So, that the Proto-Indo-European word for beech was *bhaugos. The mystery solved! au easily turns both to a, to u and to o (as it did in Late Latin). There are reasons to think that h2e gave ay in Illirian. Namely, it’s said by Pseudo-Scylax that Aenona (the ancient name for the Croatian city of Nin) comes from the Illyrian word for rocks, Aemonoi. That’s almost certainly from *h2ekjmon (like English hammer). Now, let’s analyze the arguments purported for the Laryngeal Theory. I’ll ignore the arguments from the purported loanwords, or, even worse, cognates to Indo-European from Proto-Semitic, because they are almost always based on a single phoneme in the supposedly related words. The strongest argument put forward is that the Hittite word for in front of (like Latin ante) has been transliterated as hanti. Now, Hittite was written in a syllabic script. So, how do we know that the first glyph represented ha? Because of the Hittite transcriptions of the Akkadian words. Now, in my dialect of Serbo-Croatian, there are no h-like sounds. And, in the loanwords, the h-sound gets replaced by a semivowel (either y or w). I see no reason to think that Hittite was any different.
I am not a linguist, but I think I know enough linguistics to make some conclusions by myself. If I am not right, then I am just wrong, not, as some say, not even wrong.
I would like to discuss my theories with other free-thinkers. If you think you can redirect me on some more suitable forum for such things, please do that.