summer haiku d'été – how Mount Soracte = que le mont Soractehow Mount Soracte shimmers with snow – let's quaff our wine la neige miroitée sur le mont Soracte – buvons le vin come Monte Soracte luccica con la neve - beviamo il vino mons Soracte lucet nivi - bibamus vinum Richard Vallance Original Latin: Vides ut alta stet nive candidum Soracte nec iam sustineant onus silvae laborantes geluque flumina constiterint acuto? Dissolve frigus ligna super foco large reponens atque benignius deprome quadrimum Sabina, o Thaliarche, merum diota. Permitte divis cetera, qui simul stravere ventos aequore fervido deproeliantis, nec cupressi nec veteres agitantur orni. Quid sit futurum cras, fuge quaerere, et quem fors dierum cumque dabit, lucro adpone nec dulcis amores sperne, puer, neque tu choreas, donec uirenti canities abest morosa. Nunc et Campus et areae lenesque sub noctem susurri composita repetantur hora, nunc et latentis proditor intumo gratus puellae risus ab angulo pignusque dereptum lacertis aut digito male pertinaci. Quintus Horatius Flaccus (65-8 BCE) Pronounce like Italian. painting by/ par Jean-Baptiste-Camille Corot (1796-1875)
Tag: Italian
senryu – how we all tweet = comme nous ‹‹ tweetons ›› = come noi twittiamo
senryu – how we all tweet = comme nous ‹‹ tweetons ›› = come noi twittiamohow we all tweet holding hands the world around comme nous ‹‹ tweetons ›› les mains tenues autour du monde come noi twittiamo tenendoci per mano il mondo intorno Richard Vallance
The absurd fallacy of HERSTORY. There is no such thing in any language other than English, and in fact no such thing in English!
The absurd fallacy of HERSTORY. There is no such thing in any language other than English, and in fact no such thing in English!
If there is anything which raises my hackles, it has got to be the absurd notion adapted by English language feminists alone that the word HISTORY is somehow gender related, when in fact it most certainly is not! English feminists who insist on changing the word history to herstory are displaying the most egregious linguistic ignorance. As anyone with even a smattering of higher education knows beyond a doubt, the English word history is in fact directly derived from the ancient Greek i9stori/a=, which means: inquiry, knowledge, information, science, narration and above all, a story. Now the sheer absurdity of the position of English feminists is blown wide open by the equivalent words for history in practically every other language, for instance, Dutch = geschiedenis, French = histoire, German = Geschichte, Italian = storia, Polish = przeszłość and Portuguese = história, to cite just a few examples. Any French feminist would laugh out loud at the notion that herstory is somehow the same thing as history, since in French the word for “her” is “son” (masculine gender) and “sa” (feminine gender). Of course, some allophone English feminists will scream aloud that “son” is gender-biased, without realizing in the least that gender in French, and for that matter in any and all inflected languages, including Greek, Latin, German, Russian etc. has nothing whatsoever to do with masculinity or, what is even worse, in their silly “intellectual” construct, sexism! The Dutch and German words, geschiedenis and Geschichte respectively, blow the English feminists’ ridiculous claim right out of the water, let alone the Polish przeszłość. I could cite hundreds of other languages, and the results would always be the same, to wit, the English word history has absolutely nothing to do with masculinity or sexism. So all I have to say to unilingual English feminists, “Get a life!” and at least swallow the truth with grace and dignity.
In the citations below, all italics are mine:
Consider Wikipedia:
Herstory is history written from a feminist perspective, emphasizing the role of women, or told from a woman’s point of view. The principal aim of herstory is to bring women out of obscurity from the historical record. It is a neologism coined as a pun with the word “history”, as part of a feminist critique of conventional historiography, which in their opinion is traditionally written as “his story”, i.e., from the masculine point of view.[1] (The word “history”—from the Ancient Greek ἱστορία, or historia, meaning “knowledge obtained by inquiry”—is etymologically unrelated to the possessive pronoun his.
And Rational Wiki:
“Herstory” is a neologistic term for “women’s history,” a variant of the Marxist “people’s history”; while a people’s history professes to reinterpret history from the perspective of workers and/or common men, a herstory professes to reinterpret it from the perspective of women. Most feminists don’t use it.
The term is an illustration of its coiners’ belief that regular history is heavily slanted toward men’s point of view, a “systemic bias” reflected in the term history, which they seem to have simply assumed was a portmanteau of “his story”.
Unfortunately, it happens that the English word history is a loan word, derived directly from the Latin historia, which is itself a loan word from ancient Greek.[1] On the other hand, the English word his is derived from a proto-Germanic root,[2] and is not in the least etymologically connected to the first three letters of history. They just happen to sound the same, and only in English. The origin of this term is a testament to the intellectual laziness of extremists in any field, who are quick to grab hold of anything that seems to support their point of view but reluctant to examine it critically.
and Reddit:
Also, I appreciate the title quote is somewhat playful. But I find it extremely irritating – ‘history’ is directly taken from the Greek word historia, roughly translating to ‘inquiry’ or ‘investigation’. ‘His’ and ‘her’ as actual words do not exist in Ancient Greek; words in the language meaning the same thing do exist. But the only reason ‘herstory’ is a thing is because it’s an awful pun based on the conventions of the English language which the word ‘history’ does not follow; it betrays a lack of knowledge of context, a tendency to jump on anything resembling ‘gendered’ words, and it’s a bad pun.
Is it even possible to determine what the word for “fig(s)” is in Minoan Linear A? You may be surprised!
Is it even possible to determine what the word for “fig(s)” is in Minoan Linear A? You may be surprised! Among several other tablets in both Minoan Linear A and Mycenaean Linear B, Linear A tablet HT 88 contains the supersyllabogram NI on the second line:The question is, what is the actual word for “fig(s)” in Minoan Linear A? Apparently, no-one knows. The odd thing about this supersyllabogram NI is that it was taken over lock-stock-and-barrel by the Mycenaeans. We will never know why, but it is clear that they thought it convenient simply to hang onto it. It may very well be that that the Mycenaeans continued to use the Minoan word for “fig” alongside their early Greek suza. If that is the case, it is all the more relevant for us to attempt to reconstruct the Minoan word for “fig”. Whatever the circumstances, we are still left with the perplexing question, what is the word for “fig” in Minoan Linear A anyway? In spite of apparently insurmountable obstacles, it may not be so difficult to reconstruct as we might imagine. If we stop to consider even briefly what the word for “fig” is that I have methodically selected in 13 languages, ancient and modern, belonging to 6 different classes, we discover that all but one of them are either monosyllabic or disyllabic. In one instance only is it trisyllabic, pesnika, in Serbian. This does not come as any surprise to me as a linguist, though it may to the so-called “common person” . Here are the words for “fig” in 16 languages belonging to 6 different languages classes: KEY to language classes: AU = Austronesian/ IN = Indo-European/ LI = language isolate/ NC = Niger-Congo/ SE = Semitic/ UR = Uralic. A language isolate is one which does not belong to any international language class whatsoever, but which stands entirely on its own. AU: Indonesian ara Malay rajah Maori piki IN: French figue German Feige Greek (Mycenaean) suza (Attic) suchon Italian fico Latin ficus Norwegian fiken Portuguese figo Serbian pesnika Spanish higo LI: Basque piku NC: Swahili mtimi (sub-class = Bantu) SE: Maltese tin (the only Semitic language in Latin script) UR: Finnish kuva Under the circumstances, I am given to wonder whether or not the Minoan Linear A word for “fig” is monosyllabic, disyllabic or possibly even trisyllabic. It is clear that it cannot be monosyllabic, because the supersyllabogram for “fig” in both Minoan Linear A and Mycenaean Linear B is NI. And supersyllabograms are always the first syllable only of di- tri- or multi-syllabic words in both of these languages. Given this scenario, is it possible or even feasible to reconstruct the Minoan Linear A for “fig”? Surprisingly enough, the answer is yes. Why so? It just so happens that most Minoan Linear A words which are diminutives are feminine with the ultimate being either pa3 or ra2. Under the circumstances, it only takes one small step to restore the two mostly likely candidates for the Minoan Linear A for “fig”. And these are:
It is of course possible to argue that the Minoan word for “fig” is trisyllabic, but this is highly unlikely, since the only trisyllabic word for “fig” in all 13 of the languages cited above is the Serbian, pesnika. Hence, I am reasonably convinced that the Minoan Linear A word for “fig(s)” is either nipa3 (nipai) or nira2 (nirai). Finally, as it is clear that since the word for “fig(s)” does not even remotely correspond to any of the 13 words in 6 language classes, ancient and modern, above, not even Basque, it may very well turn out that, like Basque, the Minoan language is also a language isolate. I should not be the least but surprised if it were. This discussion will be part and parcel in my upcoming article in Vol. 12 (2016) of Archaeology and Science (Belgrade) ISSN 1452-7448, “Pylos tablet Py TA 641-1952 (Ventris), the Rosetta Stone to Minoan Linear A tablet HT 31 (Haghia Triada) vessels and pottery” and a Glossary of 110 words”, the third article in a row I shall have published in this prestigious international annual by the beginning of 2018 at the very latest.
KEY POST! Introduction to the Complete Bibliography of 138 Citations for “The Rôle of Supersyllabograms in Mycenaean Linear B”
Introduction to the Complete Bibliography of 138 Citations for “The Rôle of Supersyllabograms in Mycenaean Linear B”, Presentation by Richard Vallance Janke at the 2015 Conference in the Pultusk Academy of the Humanities, Pultusk, Poland, June 30-July 2, 2015. In the next 2 posts, I shall present my exhaustive bibliography of 138 items (79 citations in each of the two parts) for the talk I shall be giving on “The Rôle of Supersyllabograms in Mycenaean Linear B” at the 2015 Conference, “Thinking in Symbols” in the Pultusk Academy of the Humanities, Pultusk, Poland, June 30-July 2, 2015. It is so exhaustive that I doubt I have missed any sources of any significance to the topic at hand. Of course, the paper of the talk itself cannot be released at this time, as it is still under wraps. Certain researchers past and present, above all Marie-Louise Nosch, have made significant contributions towards the realization of the General Theory of Supersyllabograms which I have just finalized this year, after a year of intensive research (spring 2014 – spring 2015). Previous researchers have sometimes come right up to the edge of a general theory correlating the single or multiple syllabograms they usually designate as “adjuncts” or “endograms” to the Linear B ideograms to which they are “surcharged” (i.e. attached), and which they invariably qualify. But all of these definitions are lacking in one sense or another, for the following reasons: 1. Although designated as (mere) “adjuncts” to the ideograms they invariably qualify, these associative single or multiple syllabograms (up to a maximum of 5!) are far more than that. Standing in as first-syllable abbreviations for words and even entire phrases in Mycenaean Greek, they play an absolutely critical rôle in significantly qualifying the ideograms to which they are attached, all the more so when the tablet on which they are found contains no text whatsoever, but only ideograms with these so-called “adjuncts”. But since these “adjuncts” invariably replace either Mycenaean words or (very often) entire phrases, they cannot be relegated to the status of simple adjuncts. In far too many instances, these single syllabograms encompass so much text that their inherent meaning as such turns out to be much more comprehensive and significant than that of the ideograms to which they are presumably attached. In other words, the single syllabogram(s) embodies/embody so much more than what would have otherwise been nothing but wasteful discursive text. So it appears that we should expediently and practically refer to as the ideogram as the adjunct, rather than the other way around. On tablets with no text whatsoever and with 3 or more syllabograms performing this function, it is more than apparent that all of the single syllabograms functioning as the first syllable of a Mycenaean Greek word or an entire phrase replace so much discursive text that they literally cut down the amount of space used on the tablet in question by as much as two-thirds! Since the Linear B scribes at Knossos and Pylos in particular were real sticklers for saving as much space as they possibly could on what were (and are) extremely small extant tablets (rarely more than 15 cm. or 6 inches wide), they resorted to this stratagem so often (on at least 23% of the Linear B tablets at Knossos) that the practice is, if anything, of far greater importance to an accurate decipherment of those tablets on which they appear than was previously thought. It is for this reason that I have come to designate syllabograms playing this rôle as supersyllabograms, and certainly not as mere “adjuncts” or “endograms”, since that is patently what they are – supersyllabograms. 2. The designation of supersyllabograms as “endograms” is extremely misleading and quite inaccurate, since as many of these supersyllabograms precede as follow the ideograms to which they are attached. So “endograms” account for only half of supersyllabograms at best. Besides, what are we to call the supersyllabograms which precede the ideograms to which they are attached? Has anyone thought of that or even mentioned it in previous research? Not that I have ever seen, and I have read every single document (monographs, journal articles and articles in every past conference) I could lay my hands on. The reason for this lacuna is clear enough. Past researchers have focused solely on “adjuncts” or “endograms” related solely to the field of research in Mycenaean Linear B which is of primary and frequently exclusive interest to themselves. Even Marie-Louise Nosch, who has done an astonishing amount of truly remarkable research in this area, has restricted herself to the textiles sector of the Minoan-Mycenaean economy, as that is her primary field of interest. Fair enough. Given this scenario, it appears to me that researchers past and present have been focusing exclusively on the trees or even sometimes, as with Marie-Louise Nosch, on whole clearings in forest. But none have ever concentrated on the entire forest, at least until last year, when I myself decided to ransack every single syllabogram on some 3,000 tablets (not fragments) from Knossos, in order to hypothesize, if at all possible, a general pattern to the use of supersyllabograms with ideograms. I succeeded beyond my wildest dreams. So far, I have discovered that at least 33 of the 61 syllabograms plus one of the homophones (“rai” for saffron) frequently function as supersyllabograms. Under the circumstances, and given that so many scribes so often resorted to this strategy, I soon enough concluded that it was not only a standard convention in the compilation of some 700 tablets at Knossos, but that the supersyllabograms found on these tablets were almost invariably formulaic codes. And in ancient Greek – witness Homer alone - any practice which was both conventional and formulaic was always deliberate. No-one ever resorts to such strategies in any language, unless they have abundant reason to do so. This is all the more true for the practices the Linear B scribes routinely ascribed to, given that they would do absolutely anything, if they possibly could, to save precious space on their tiny clay tablets. This too is another crucial factor past researchers have overlooked. Linear B scribes only recorded information which was absolutely essential to the precise compilation of what were (and are) after all statistical accounts and inventories. We can take the far-reaching consequences and implications of this conclusion even further. Have you ever seen a modern-day inventory which resorts to similar tactics to conserve precious space and to make the inventory as clear, precise and accurate as possible? Of course you have. As illustrated in the following two examples, the most efficient of modern inventories resort to the same tactics, the formulaic use of code abbreviations as substitutes for wasteful discursive text with predictable frequency – which is almost always: Click to ENLARGE each one with its relevant notes![]()
In other words, just as abbreviations serve as default codes in modern inventories, supersyllabograms function pretty much the same way on the Linear B tablets. Supersyllabograms are in fact inventory codes for the Mycenaean Linear B words or entire phrases they replace. This revelation surely substantiates the claim I am now going to make: the Linear B scribes were far ahead of their time in the compilation of inventories and statistics. No other ancient language, including classical Greek and even Latin, came remotely close to this extremely advanced practice the Linear scribes so brilliantly and consciously contrived for their astonishing ability to create practical templates they consistently applied to inventorial management. And no-one until the Italian bankers in Renaissance was to revive the practice with equal skill. As for the standard practices of the Linear B scribal inventories, they are so remarkably alike modern 20th. & 20st. Century practices that it is uncanny. 3. But there is more. Why previous researchers have not drawn attention to the fact that many supersyllabograms, especially in the field of textiles, neither precede nor follow the ideograms they qualify, but are almost invariably inside them, is beyond me. Once again, no one in any language resorts to any stratagem without solid practical and even logical reason(s). Such is the case with the textile “intragrams”, as opposed to “exograms” in Linear B, the latter of which invariably qualify pretty much all ideograms in the field of agriculture. Again, this raises the critical, hardly hypothetical, question, why. And again, there are substantive and strictly functional reasons why the Linear B scribes made this critical distinction – because they knew they had to. Supersyllabograms functioning as “exograms” are always associative, while those operating as “intragrams” are invariably attributive. The Linear B scribes made this fundamental distinction between the two sub-classes of supersyllabograms for the simple reason that they, as a guild, knew perfectly well what the operative distinction was which each of these types of supersyllabograms played on the tablets on which they were inscribed. The talk I am giving at the Conference in Pultusk between June 30 and July 2 2015 will make this perfectly clear. 4. I have no objection to the designation “surcharged” for “exograms” as supersyllabograms, because they are not only literally surcharged onto the ideograms with which they are always associated, they also figuratively surcharge the meaning(s) of these ideograms, in a sense somewhat akin to super-charged gasoline or petrol which beefs up engine performance in cars - or by symbolic association, something along those lines. But I am forced to object to the designation of “intragrams” as surcharged in the textiles sector of the Minoan-Mycenaean economies, for the obvious reason that they are both literally and figuratively not surcharged at all. Again, the scribes never resorted to “intragrams”, unless they were absolutely critical to an actual attribute, whenever required in a particular case, such as the frequent designation of colour for textiles. Ask yourselves, why would any scribe in his right mind write out the full name of the default colour white for linen, when he did not have to? He simply would not. On the other hand, the Linear B scribes did make use of an attributive supersyllabogram when they knew perfectly well that it was critical to the economic class status of the cloth so designated. For instance, purple cloth, designated by the supersyllabogram PU for Mycenaean Linear B pupureyo – a royal colour par excellence – was much more refined and far more expensive than the heavier and coarser plain white linen cloth (rino) spun for the hoi polloi (the lower classes). So they had to mention that for the sake of the “wanaka” or King (of Knossos or Mycenae) to whom this distinction was all too important, given that neither he nor his Queen no any of the princes royal would ever be caught dead wearing cheap cloth. There is much more to this than meets the eye, as I shall clearly illustrate in the book, The Decipherment of Supersyllabograms in Linear B, which is to appear sometime in 2016, if all goes well. I would be truly remiss were I not to acknowledge the major contributions the French researcher, Marie-Louise Nosch, whom I have cited 15 times (!) in my bibliography, has made to fundamentally accurate definitions of supersyllabograms in the textile sector of the Minoan-Mycenaean economy. Although I happened upon all of her astonishingly insightful research articles only after I had deciphered 32 of the 34 supersyllabograms (the other two being beyond me, as well as her), the truly accurate and intrinsically logical conclusions she came to on her own back up my conclusions on the meanings of practically all the intragrams for textiles almost to the letter. This amazing co-incidence, if that is merely what it is, serves as solid circumstantial collateral evidence to substantiate my Theory of Supersyllabograms. Co-incidence? I rather doubt that. It is a given that researchers in any scientific field tend to strike their bearings in the same general direction in any age, including our own. Like Odysseus, we are all heading for the same shore. The most convincing conclusions which will eventually be drawn from the research we are all sharing in now are yet in the offing. But in my eyes one thing is certain. Everything we researchers in Mycenaean Linear B, as a community, are aiming for now is bound to make a ground-breaking, perhaps even profound, contribution in the near future to make the further decipherment of Linear B considerably much more accurate than any we have seen to date. The Bibliography to follow in two parts (1-69 & 70-138) in the next two posts. ADDENDUM: I shall be publishing this post & the next two in academia.edu very soon, prior to my presentation at the Conference in Pultusk, Poland, June 30 - July 2, 2015. Richard
Je suis Charlie – in French, English & Greek + 11 modern languages & 3 ancient Greek dialects!
Je suis Charlie - in French, English & Greek + 11 modern languages & 3 ancient Greek dialects!I beg you, please be sure to RETWEET this, folks! As a polyglot Canadian, fluent in English and French, conversant with both modern languages and ancient, especially ancient Greek, with some 20 dialects under my belt, including Mycenaean Linear B & Arcado-Cypriot Linear C, I hope to reach not only everyone alive now, but as many of our ancestors as possible. I do this out of love for all the millions upon millions of people who have been slaughtered by warmongers, manaics, religious fanatics & terrorists, past, present and... God forbid... future! Je vous prie de tout mon coeur de faire des RETWEETs de ce message des plus urgents! Tout en étant canadien parfaitement bilingue, je suis également polyglotte, connaisseur de plusieurs langues modernes et anciennes, dont une vingtaine de dialèctes grecs tels que le mycénien en linéaire B et le chypro-arcadien en linéaire C. Dans ce but, j’espère communiquer ce message de solidarité bienveillante à tous ceux qui sont encore vivants autant qu’à tous nos ancêtres, dont d’innombrables millions qui ont perdu la vie, tous massacrés par des bellicistes, des maniaques, des fanatiques religieuses et des terroristes d’antan, de nos jours et... à Dieu ne plaise ... incontournablement à l’avenir. Richard Vallance Janke, Ottawa, Ontario, Canada
Happy New Year in Greek, Linear B, Linear C, English, Latin, French, Italian, Spanish, Romanian & German
Significant Phonetic Variations in the Pronunciation of the L & R + Vowel Series as Reflected in the Linear B & Linear C Syllabaries
Significant Phonetic Variations in the Pronunciation of the L & R + Vowel Series as Reflected in the Linear B & Linear C Syllabaries Comparison of the Mycenaean Linear B & the Arcado-Cypriot Linear C Syllabaries: Click to ENLARGEWhere the phonetic values of the syllabogram series R + vowel series (the L series missing in Linear B) do sound somewhat different in the two syllabaries, Linear B & Linear C, there is in fact a very sound phonetic reason for this. But first, let me tell you a little story. If there were such a thing as time-travel into the past or the future, modern Greeks from Athens travelling back to the city in 500 BCE would indeed be shocked at how Greek was pronounced then. Ancient Greeks thrust into the future would have the same reaction – utter disbelief. I myself put this hypothesis to test while I was in Greece in May 2012. I could read modern Greek fairly well even then. But when I tried to communicate with some folks I met in a restaurant with my own plausible version of the ancient Attic dialect (there are actually 3 or 4 possible versions), no-one could understand scarcely a word I spoke. And when I asked my colleagues to speak modern Greek to me, I was equally at a loss. But I could read the menu with no problem. Moving on then. Although the Mycenaean Greeks were apparently unable to pronounce the letter “L”, nothing in fact could be more deceptive to the unwary Occidental ear. It all comes down to a matter of our own ingrained linguistic bias in our own social-cultural context. To us, the Arcadians & Cypriots indeed appear to have already made the distinction between L & R, given that their syllabary contains syllabograms consisting of both of these consonants (L & R) followed by vowels. But I stress, to us, we cannot be sure how they pronounced L & R, or to what extent they had by then become phonetically separate. Looking at the Linear C Syllabary, you can see the distinction right away. See above. Now some of you may already be aware of the “fact” that to our ears in the West, the Japanese seem utterly incapable of pronouncing either L or R, but appear to be pronouncing something half-way between the two, which sounds like mumble-jumbo to our ears. Since the Linear B syllabary has no L + vowel series of syllabograms, the Mycenaeans too might have conflated L and R into one consonant in a manner similar to the way the Japanese pronounce it, at least in the early days of the Mycenaean dialect (possibly from 1450-1300 BCE). On the other hand, since the Arcadians and the Cypriots had apparently already made the distinction between L & R from as early as 1100 BCE, their immediate forebears, the Mycenaeans, might have already been well on the road to being able to pronounce L & R distinctly as consonants by 1300-1200 BCE, although they still may have been confusing them from time to time. However, they probably could see no point in adding an L + vowel series such as we see in Arcado-Cypriot Linear C, given that they had already used the Linear B syllabary for at least 2 and half centuries (ca 1450 – 1200 BCE). I am inclined to accept this hypothesis of the gradual emergence of L & R towards the end of Linear B's usage ca. 1200 BCE, in preference to a hypothetical Japanese-like pronunciation based on the assumption that the L+R concatenation is a merging of L & R as semi-consonants. Still, either scenario is perfectly plausible. Allophone English speakers invariably find the pronunciation of L & R in almost all other Occidental languages (French, Spanish, Italian etc.) much too “hard” to their ears. This is because the letters L & R in English alone are alveolars, mere semi-consonants or semi-vowels, depending on your perspective as an English speaker, which is in turn conditioned by the dialect you speak. There are some English dialects in which the letter R is still pronounced as a trilled consonant, but for the most part, allophone English speakers pronounce both L and R very softly – at least to the ears of allophones from other European nations. Practically all other modern European languages trill the letter R, making it a consonant in their languages. But not English. This is due to the “Great Vowel Shift” which occurred in the fourteenth and fifteenth centuries in English, whereby the earlier trilled R was abandoned in favour of the much softer semi-consonant or semi-consonant R we now use in English. For this reason and others besides, English alone of the Germanic languages is not guttural at all. At the same time, L, which had previously been pronounced just as in other European languages, became a semi-consonant. You can put this to the test by pronouncing either letter aloud, paying close attention to where you touch the interior of your mouth with your tongue. Your tongue will instantly feel that the semi-consonant L scarcely makes any contact with any firm area of the mouth, but is pronounced almost the same way the vowels are – almost, but not quite. On the other hand, R, which is a true semi-vowel, at least in Canadian and American English, does not make any contact with any firm part of the mouth, in other words, it is pronounced just as the vowels are – no contact. So it really is a vowel. But English scholars and linguists in the sixteenth century could see no point in changing R to a vowel, any more than the Mycenaeans could be bothered with a new series of syllabograms beginning with L. But it matters little, if at all, whether or not we pronounce L & R as semi-consonants or semi-vowels, as in English, or as separate consonants as in the other European languages, since we all pronounce them as distinct letters. This ingrained linguistic bias greatly colours our perceptions of how others pronounce the “same” letters, if indeed they are the same at all. So it all boils down to just one thing: it all depends on your socio-culturally conditioned perspective as a speaker of our own language. This state of affairs leaves me forced to draw the inescapable conclusion that to the Japanese it is we who have made a mess of things by separating the pronunciation of L & R, which sound identical to their ears, in other words as one consonant (which is neither a semi-consonant nor a semi-vowel). To assist you in putting this into perspective, consider the Scottish pronunciation of the letter R, which is also clearly a consonant, and is in fact the pronunciation of R before the Great Vowel Shift in Middle English. This is not to say that the Scottish pronounce their consonant R anywhere near the way the Japanese pronounce their single consonant, to our ears an apparent conflation of our two semi-consonants L & R. Again, the whole thing comes down to a matter of linguistic bias based on the socio-cultural conditions of two very distinct meta-cultures, Occidental and Oriental. In that context, the languages in their meta-classes (Occidental versus Oriental) are symbolic of two entirely different perspectives on the world. Need I say more? In conclusion, it is extremely unwise to draw conclusions for phonetic distinctions between the socio-culturally “perceived” pronunciation of any consonant or any vowel whatsoever from one language to another, especially in those instances where one of the languages involved is Occidental and the other is Oriental. The key word here is “perceived”. It is all a question of auditory perception, and that is always conditioned by the linguistic norms of the society in which you live. This still leaves us up in the air, so to speak. How can we be sure that the Mycenaean Greeks apparently could not pronounce their Ls “properly” (to be taken with a grain of salt). We cannot. Since we were not there at the time, our own linguistic, socio-cultural biases figure largely in our perception of what the so-called “proper” pronunciation was. If we mean by proper, proper to themselves, that is an altogether different matter. But what is proper to us was almost certainly not proper to them, of that we can pretty much rest assured. The same situation applies to every last ancient Greek dialect. What was proper pronunciation and orthography to the Dorian dialect most certainly was not for the Arcado-Cypriot dialect any more than it was for Attic Greek. To be perfectly blunt, we cannot ever be quite sure how anyone in ancient Greek pronounced their own dialect of the language, again because we weren't there. Richard
My Twitter account completely updated, new header new photo, and new, wider perspectives: Click to ENLARGE
My Twitter account completely updated, new header new photo, and new, wider perspectives: Click to Visit:I have just updated and completely revised not only the appearance but the contents of my Twitter account, to reflect my widely expanding interests as related, either directly or indirectly, to Mycenaean Linear B, Minoan Linear A, Arcado-Cypriot Linear C, ancient Greek etc. etc. I have posted a new header, which you see above, incorporating the Linear B word for Knossos, and part of the stunning dolphins fresco in the Queen’s Megaron at Knossos, which you can see here: Click to ENLARGE
As it now stands, in its short lifetime of less than two years, our Linear B (A & C) blog has become one of the primary Linear B resources on the entire Internet, with visits already running into the tens of thousands (an astounding figure for something as bizarre and esoteric as Linear B!). Soon approaching 40K, we expect at least 60K hits by our second anniversary, if not more. The reasons for this are obvious to anyone with even a passing interest in Linear B (A &C). Nothing is off-limits on our blog. Neither Rita Roberts, my research colleague, nor I, take anything for granted. We are both “doubting Thomases” to the core, casting doubt not only on translations of Linear B tablets by other Linear B researchers, but on one anothers as well, given that neither of us is in the least impervious to committing errors, sometimes egregious. Such errors must be drawn to our attention, come what may. If you are an expert in Linear B decipherment, and you do not like any translation either of us has made, feel free to give us a shout. The other principal concerns and issues our blog frequently focuses on are: 1. keeping the Linear B syllabary right up to date. The syllabary chart most commonly used on the Internet is way out-of-date, and must be replaced by this one: Click to ENLARGE
2. the introduction of the completely new theory of Supersyllabograms in Mycenaean Linear B, of which there are at least 30 from the store of 61 syllabograms. We have plenty of posts on our theory on our blog. Rita Roberts and I shall be publishing a major research article on supersyllabograms sometime in 2015 or 2016. If tenable, it should prove to be a revolutionary step forward in the decipherment of the remaining 10% or so of the Linear B syllabary, its homophones, logograms and ideograms as yet undeciphered over the past 62 years since Michael Ventris successfully and amazingly deciphered the other 90%. Our research will be widely available in PDF format on the Internet, and although copyrighted, will be free for use by any Linear B aficionados. Here is an example of just a few supersyllabograms, all dealing with sheep, rams & ewes, the primary concern of Linear B scribes by a long shot: Click to ENLARGE
3. Progressive Linear B Vocabulary and Grammar, another all-new approach to the study of Linear B, whereby I intend to re-construct as much of the lost grammar of Mycenaean Greek as I possibly can. I have already completely mapped the active voice of both Thematic and Athematic verbs in Mycenaean Greek. Nouns, adjectives, adverbs and prepositions with cases are to follow in 2015. To view all posts on this topic, visit our PINTEREST Board, Mycenaean Linear B Grammar and Vocabulary:
4. Rita Roberts and I shall be constructing an all new English-Linear B Lexicon sometimes between 2016 & 2018, which will be vastly superior to the currently available Mycenaean (Linear B) – ENGLISH Glossary on the Internet, of which the less said the better, as it is riddled with at least 100 errors! I strongly dis-advise anyone using it. If you must use a Mycenaean dictionary, be sure to avail yourselves of Chris Tselentsis’ far superior Linear B Lexicon. 5. the all new field of the feasibility of the possible application of the Linear syllabaries, especially B & C but also, to a lesser extent, Linear A, to the emerging field of extraterrestrial communication, by which I mean serious research as undertaken by NASA: Click to read the entire PDF
and other space administration, research facilities and professional online sites, and not crackpot nonsense such as UFOs, alien abductions and the like. Here are a few comic strips just to make it clear exactly what I think of extraterrestrial crackpots: Click to ENLARGE
followed by this famous quotation by Werner Karl Heisenburg: Click to read the Wikipedia article on him:
These are just the 5 major ventures we are undertaking on our blog, but we do not shy away from anything whatsoever which advances our knowledge of Linear B in general and in particular. My Twitter account has expanded its scope to include not only my primary pursuits, research into Linear A, B & C and ancient Greek, especially the archaic Greek of the Catalogue of Ships in book II of Homer’s Iliad, which I am in the process of translating in its entirety, as you can see here: Click to ENLARGE
but also the following areas of great interest to me: 1. posting of major research articles, not only in English, but in French and Italian as well, the latter two of which I shall translate into English whenever I deem it necessary for our blog readers; 2. ancient Greek vocabulary, but exclusively in the East Greek dialects, Mycenaean Greek, Arcado-Cypriot, Aeolian, Ionic and Attic; 3. Decipherment of ancient languages in general, insofar as these related, either directly or indirectly, to Linear syllabaries; 4. Cryptology, such as the Bletchley Circle project in World War Two, and the key rôle the brilliant genius, Alan Turing, the equal of Michael Ventris in intellect, played in the decipherment of the Enigma Code, especially as this astounding achievement relates to... 5. thorough investigation and in-depth analysis of the possible suitability of of syllabic scripts such as Linear A, B & C into extraterrestrial communication (NOT UFO’s, which are crackpot nonsense suitable only to... I will not fill in the blanks!); 6. astronomy, Mars, exoplanets etc. (not reflected on this blog, of course, except insofar as it may possibly relate to Linear syllabaries), linguistics in general, including translation from one language to another, especially between English & French, in which as a Canadian I am fluent, Latin & Greek and Italian, which I read very well & Spanish, fairly well. I have forgotten my Russian, which I learned 50 years ago, but I can still read the Cyrillic alphabet with no difficulty. Linguistics and translation posts on this blog must in some way be related to Linear syllabaries, but not on my Twitter account, where anything important about linguistics in general is just fine with me. Richard
The Homophone HA, used less often than AI, but equally significant… if not more so!
The Homophone HA, used less often than AI, but equally significant: Click to ENLARGE:This makes for entertaining reading, though possibly somewhat perplexing to some. Let no-one be under any illusion that the Linear B homophone HA is any less significant than AI, regardless of the fact that it appears less often in Linear B texts on extant tablets. The homophone HA is not a diphthong! This homophone (HA) takes an enormous leap forward, specifically and exclusively in the Linear B syllabary, by explicitly expressing initial or even internal aspirated A’s. This incredible achievement eclipsed even the ancient Greek alphabet, which, need I remind you, was always written in CAPS (uppercase) alone, and hence, was utterly incapable of expressing any aspirated, let alone, unaspîrated vowels. "What” I hear you indignantly explain, "Of course, they had aspirated and unaspirated vowels.” Yes, they did. But they never expressed them. Search any ancient alphabetical text in any dialect whatsoever for aspirated or unaspirated vowels, and you search in vain. Search Linear B, and voilà, staring us squarely in the face, is the aspirated A. Astonishing? Perhaps... perhaps not. But what this tells us unequivocally is that the ancient Greeks, even after the appearance of the alphabet, must have pronounced aspirated and unaspirated vowels, because in Mycenaean Greek, the aspirated A is squarely in the syllabary.
"But”, I hear you exclaim again, "If those Mycenaeans were so smart, why didn’t they also have a homophone for the aspirated E, which pops up all over the place in Medieval manuscripts in Classical Greek?” The answer is that Mycenaean Greek almost certainly had no use for the aspirated E, since all classical Greek words beginning with an aspirated E invariably begin with an aspirated A in Mycenaean Greek, as for instance, Mycenaean "hateros” versus classical Greek "heteros” (well, in most dialects, if not all). In other words, Mycenaean Greek grammar has no homophone for aspirated E, simply because they never used it, nor were they even aware of its existence. Still, the fact remains that, at least where the aspirated A is concerned, Linear B was one step ahead of ancient alphabetical Greek. Both aspirated and unaspirated initial consonants were a feature introduced into written classical Greek alphabet only in the Middle Ages, when monks & other scribes began making extensive use of lower case letters. And, sure enough, along with the aspiration and non-aspiration of initial vowels (most often A, E & U), they also introduced all those other crazy accents we all must now memorize: the acute, grave, circumflex and susbscripted iota, just to make reading ancient Greek wretchedly more complicated. Don’t you wish they had left well enough alone? I often do. But this was not to be, since from the Middle Ages, and especially from the Renaissance on, almost all Occidental languages (Greek & French being two of the worst offenders) used accents liberally. Apparently only the Romans never bothered with accents ... but even here we cannot be sure, as they too wrote only in CAPS (uppercase). Even English, which is the Western language most adverse to accents, always uses them in borrowed words from French, Italian, Spanish etc. So you just can’t win. Once again, amongst the ancient languages, at least as far as I know, Linear B alone was able to explicitly express the initial aspirated A, just as Linear B had the common sense to separate every word on the tablets from the next with a vertical line (|). After that, "something got lost in translation” (so to speak), and for at least 2 millennia, when all of a sudden everyone in the whole world went bonkers for accents. Such are the vagaries of linguistics. Richard
You must be logged in to post a comment.