The last whisper

The last two posts have been about biological extinction; today's, to continue in the same elegiac tone, is about language extinction.

A study in Ethnologue found that of the Earth's current seven-thousand-odd languages, 3,143 -- around forty-four percent -- are in danger of going extinct.  Languages become endangered from a number of factors; when children are not taught as their first language, when there's government suppression, or when a different language has become the primary means of communication, governance, and commerce.  

Of course, all three of those frequently happen at the same time.  The indigenous languages of North and South America and Australia, for example, have proven particularly susceptible to these forces.  In all three places, English, Spanish, and Portuguese have superseded hundreds of languages, and huge swaths of cultural knowledge have been lost in the process.

In almost all cases, there's no fanfare when a language dies.  They dwindle, communication networks unravel, the average age of native speakers moves steadily upward.  Languages become functionally extinct when only a few people are fluent, and at that point even those last holdouts are already communicating in a different language with all but their immediate families.  When those final few speakers die, the language is gone -- often without ever having been studied adequately by linguists.

Sometimes, however, we can pinpoint fairly closely when a language died.  Curiously, this is the case with Ancient Egyptian.  This extinct language experienced a resurgence of interest in the early nineteenth century, due to two things -- the British and French occupations of Egypt, which resulted in bringing to Europe hundreds of priceless Egyptian artifacts (causing "Egyptomania" amongst the wealthy), and the stunning decipherment of hieroglyphics and Demotic by the brilliant French linguist Jean-François Champollion.

For a millennium and a half prior to that, though, Ancient Egyptian was a dead language.  And as weird as it sounds, we know not only the exact date of its rebirth, but the date it took its last breath.  Champollion shouted "Je tiens mon affaire!" ("I've figured it out!") to his brother on 14 September 1822.  

And the last inscription was made by the last known literate native speaker of Ancient Egyptian on 24 August 394 C.E.

It's called the "Graffito of Esmet-Akhom," and was carved on the wall of a temple in Philae.  It shows the falcon-god Mandulis, who was a fairly recent invention at the time, and like the Rosetta Stone, it has an inscription in hieroglyphics and Demotic:

A drawing of the Graffito of Esmet-Akhom. The hieroglyphic inscription is in the upper right, the Demotic in the lower right. [Image licensed under the Creative Commons Mickey Mystique, Graffito of Esmet-Akhom, drawing, 01, CC BY-SA 4.0]

The text reads:

Before Mandulis, son of Horus, by the hand of Esmet-Akhom, son of Nesmeter, the Second Priest of Isis, for all time and eternity.  Words spoken by Mandulis, lord of the Abaton, great god.

I, Esmet-Akhom, the Scribe of the House of Writings(?) of Isis, son of Nesmeterpanakhet the Second Priest of Isis, and his mother Eseweret, I performed work on this figure of Mandulis for all time, because he is fair of face towards me.  Today, the Birthday of Osiris, his dedication feast, year 110.

[Nota bene: The "year 110" is not, of course, by the Julian calendar; in Egypt, governmental records were dated using the number of years since the accession of the Roman emperor Diocletian in 284 C.E.  The "Birthday of Osiris" is what we now call the 24th of August.]

At this point, Greek and Latin had already superseded Egyptian in written records, so Esmet-Akhom was the last of a dying breed.  Ancient Egyptian lingered on for a while as a spoken language amongst the working classes; liturgical Coptic is a direct descendant.  But any final vestiges of Egyptian as a living language were eradicated in the seventh century when the Graeco-Roman state of Egypt fell to the Arabs.

So it may well be that the priests of Esmet-Akhom's family were the last people capable of reading and writing Egyptian, at least until Champollion came along.

I know change is the way of things, and given the interconnectedness of the world today, that widely-spoken languages will inevitably gain more and more of an edge over minority languages.  (Consider that a full forty percent of the Earth's people speak one of eight languages -- Mandarin, Spanish, English, Arabic, Hindi, Bengali, Portuguese, and Russian.)  But still, I can't help but find the loss of linguistic diversity, and the cultural information those dead languages once encoded, to be sad.

And you have to wonder how Esmet-Akhom himself felt, writing his defiantly confident inscription "for all time and eternity" at the end of the fourth century.  Did he know that Egyptian was, even at that point, moribund?  A means of communication that had existed for over four thousand years was on the road to extinction; what was left was only a whisper, and even that would soon be silenced.

Has to make you wonder what linguistic shifts will occur in the next thousand years.

****************************************

Root and branch

Linguists estimate that there are a little over seven thousand languages spoken in the world, sorted into around four hundred language families (including linguistic isolates, languages or language clusters that appear to be related to no other known languages).

As a historical linguist, one of the most common questions I've been asked is if, ultimately, all of those languages trace back to a common origin.  Or, perhaps, did disparate groups develop spoken language independently, so there is no single "pre-Tower-of-Babel" language (if I can swipe a metaphor from the Bible)?  The honest answer is "we don't know."  Determining the relationships between languages -- their common ancestry, as it were -- is tricky business, and relies on more than chance similarity between a few words.  My own area of research was borrow words in Old English and Old Gaelic (mostly from Old Norse), a phenomenon that significantly complicates matters.  English has an unfortunate habit of appropriating words from other languages -- a selective list of English vocabulary could easily lead the incautious to the incorrect conclusion that it originated from Latin, for example.  (In the preceding sentence, the words unfortunate, habit, appropriating, language, selective, vocabulary, incautious, incorrect, conclusion, originated, and example all come directly from Latin.  As do preceding, sentence, and directly.  So none of those are original to English -- they were adopted by scholars and clerics between the thirteenth and sixteenth centuries C.E.)

As you might expect, the longer two languages have been separate, the further they diverge, not only because they borrow words from (different) neighboring languages but because of random changes in pronunciation and syntax.  There's a good analogy here to biological evolution; the process is much like the effect that mutations have in evolution.  Closely-related species have very similar DNA; extremely distantly-related ones, like humans and apple trees, have very few common genes, and it's taken a great deal of detailed analysis to show that all life forms do have a single common ancestor.

That feat has not yet been accomplished with language evolution.  Finnish and Swahili may have a common ancestor, but if so, they've been separate for so long that all traces of that relationship have been erased over time.

Even with groups of languages with a more recent common ancestor, it can remarkably difficult to piece together what their relationship is.  For Indo-European languages, surely the most studied group of languages in the world, we're still trying to figure out their family tree, and aligning it with what is known from history and archaeology.  This was the subject of a study out of the University of Copenhagen that was published last week, and looked at trying to reconcile the language groups in southern and western Europe with what we now know from genetic studies of ancient bones and teeth.

[Nota bene: the Germanic and Slavic peoples were not part of this study; the current model suggests that 

Germanic groups are allied to the neolithic northern Corded Ware and Funnelbeaker Cultures, which appear to have originated in the steppes of what are now western Russia and Ukraine; the Slavs came in much later, probably from the region between the Danube River and the Black Sea.]

The study found a genetic correlation between speakers of the Italo-Celtic language cluster (Italian, Spanish, French, Portuguese, Catalan, Occitan, and Romanian; Irish, Scottish Gaelic, Manx, Cornish, Breton, and Welsh) and one between speakers of the Greco-Armenian cluster (Greek, Cypriot, Albanian, and Armenian).  The southern branch of the Corded Ware culture seems to have undergone two influxes from the east -- one from the Bell Beaker Culture, starting in around 2800 B.C.E. (so called because of the characteristically bell-shaped ceramic drinking vessels found at their settlement sites), which ended up migrating all the way to the Iberian Peninsula, and the other from the Yamnaya, which came from the Pontic steppe but never got past what is now Switzerland and eastern Italy (most of them didn't even get that far).

It's tempting to overconclude from this; just like my earlier example of Latin borrow words in English, the genetic correlation between the Italo-Celtic and Greco-Armenian regions doesn't mean that the differences we see in those two branches of the Indo-European language family come from the Bell Beaker people and the Yamnaya, respectively.  The lack of early written records for most of these languages means that we don't have a good "fossil record" of how and when they evolved.

But the current study provides some tantalizing clues about how migration of speakers of (presumably) two different dialects of Proto-Indo-European may have influenced the evolution of the western and eastern branches of today's Indo-European languages.

So it's one step toward finding the common roots of (most) European languages.  Even if we may never settle the question of how they're related for certain, it's cool that they're using the techniques of modern genetics to find out about where our distant ancestors came from -- and what languages they may have spoken.

****************************************

Onomatopoeia FTW

Given my ongoing fascination with languages, it's a little surprising that I didn't come across a paper published a while back in the Proceedings of the National Academy of Sciences earlier.  Entitled, "Sound–Meaning Association Biases Evidenced Across Thousands of Languages," this study proposes something that is deeply astonishing: that the connection between the sounds in a word and the meaning of the word may not be arbitrary.

It's a fundamental tenet of linguistics that language is defined as "arbitrary symbolic communication."  Arbitrary because there is no special connection between the sound of a word and its meaning, with the exception of the handful of words that are onomatopoeic (such as boom, buzz, splash, and splat).  Otherwise, the phonemes that make up the word for a concept would be expected to having nothing to do with the concept itself, and therefore would vary randomly from language to language (the word bird is no more fundamentally birdy than the French word oiseau is fundamentally oiseauesque).

That idea may have to be revised.  Damián E. Blasi (of the University of Zürich), Søren Wichmann (of the University of Leiden), Harald Hammarström and Peter F. Stadler (of the Max Planck Institute), and Morten H. Christiansen (of Cornell University) did an exhaustive statistical study, using dozens of basic vocabulary words representing 62% of the world's six thousand languages and 85% of its linguistic lineages and language families.  And what they found was that there are some striking patterns when you look at the phonemes represented in a variety of linguistic morphemes, patterns that held true even with completely unrelated languages.  Here are a few of the correspondences they found:

• The word for ‘nose’ is likely to include the sounds ‘neh’ or the ‘oo’ sound, as in ‘ooze.’
• The word for ‘tongue’ is likely to have ‘l’ or ‘u.’
• ‘Leaf’ is likely to include the sounds ‘b,’ ‘p’ or ‘l.’
• ‘Sand’ will probably use the sound ‘s.’
• The words for ‘red’ and ‘round’ often appear with ‘r.’
• The word for ‘small’ often contains the sound ‘i.’
• The word for ‘I’ is unlikely to include sounds involving u, p, b, t, s, r and l.
• ‘You’ is unlikely to include sounds involving u, o, p, t, d, q, s, r and l.

"These sound symbolic patterns show up again and again across the world, independent of the geographical dispersal of humans and independent of language lineage," said Morten Christiansen, who led the study.  "There does seem to be something about the human condition that leads to these patterns.  We don’t know what it is, but we know it’s there."

[Image licensed under the Creative Commons M. Adiputra, Globe of language, CC BY-SA 3.0]

One possibility is that these correspondences are actually not arbitrary at all, but are leftovers from (extremely) ancient history -- fossils of the earliest spoken language, which all of today's languages, however distantly related, descend from.  The authors write:

From a historical perspective, it has been suggested that sound–meaning associations might be evolutionarily preserved features of spoken language, potentially hindering regular sound change.  Furthermore, it has been claimed that widespread sound–meaning associations might be vestiges of one or more large-scale prehistoric protolanguages.  Tellingly, some of the signals found here feature prominently in reconstructed “global etymologies” that have been used for deep phylogeny inference.  If signals are inherited from an ancestral language spoken in remote prehistory, we might expect them to be distributed similarly to inherited, cognate words; that is, their distribution should to a large extent be congruent with the nodes defining their linguistic phylogeny.

But this point remains to be tested.  And there's an argument against it; if these similarities come from common ancestry, you'd expect not only the sounds, but their positions in words, to have been conserved (such as in the English/German cognate pair laugh and lachen).  In fact, that is not the case.  The sounds are similar, but their positions in the word show no discernible pattern.  The authors write:

We have demonstrated that a substantial proportion of words in the basic vocabulary are biased to carry or to avoid specific sound segments, both across continents and linguistic lineages.  Given that our analyses suggest that phylogenetic persistence or areal dispersal are unlikely to explain the widespread presence of these signals, we are left with the alternative that the signals are due to factors common to our species, such as sound symbolism, iconicity, communicative pressures, or synesthesia...  [A]lthough it is possible that the presence of signals in some families are symptomatic of a particularly pervasive cognate set, this is not the usual case.  Hence, the explanation for the observed prevalence of sound–meaning associations across the world has to be found elsewhere.

Which I think is both astonishing and fascinating.  What possible reason could there be that the English word tree is composed of the three phonemes it contains?  The arbitrariness of the sound/meaning relationship seemed so obvious to me when I first learned about it that I didn't even stop to question how we know it's true.

Generally a dangerous position for a skeptic to be in.

I hope that the research on this topic is moving forward, because it certainly would be cool to find out what's actually going on here.  I'll have to keep my eyes out for any follow-ups.  But now I'm going to go get a cup of coffee, which I think we can all agree is a nice, warm, comforting-sounding word.

  

****************************************

Lingua franca

Here's a question I wonder if you've ever pondered:

Why do the Spanish and French speak Romance languages and not Germanic ones?

It's not as weird a consideration as it might appear at first.  By the time the Western Roman Empire collapsed in the last part of the fifth century C.E., the entire western part of Europe had been completely overrun by Germanic tribes -- the Franks, the Burgundians, and especially the Visigoths.  This latter group ended up controlling pretty much all of southern France and nearly the entirety of Spain, and their king, Euric, ruled the whole territory from his capital at Toulouse.  It was Euric who deposed the last Western Roman emperor, poor little Romulus Augustulus, in 476 -- but showing unusual mercy, sent him off to a (very) early retirement at a villa in Campania, where he spent the rest of his life.  That he felt no need to execute the kid is a good indicator of how solidly Euric and the Visigoths were in control.

So the Germanic-speaking Goths more or less took over, and not long after that the (also Germanic) Franks and Burgundians came into northern France and established their own territories there.  The country of France is even named after the Franks; but their language, Franconian, never really took hold inside its borders.

Contrast this to what happened in England.  The Celtic natives, who spoke a variety of Brythonic dialects related to Welsh and Cornish, were invaded during the reign of the Emperor Claudius in the year 43 C.E., and eventually Rome controlled Britain north to Hadrian's Wall.  But when all hell broke loose in the fifth century, and the Roman legions said, "Sorry, y'all'll have to deal with these Saxons on your own" and hauled ass back home, the invaders' Germanic language became the lingua franca (pun intended) of the southern half of the island, with the exception of the aforementioned Welsh and Cornish holdouts.

All three places had been Roman colonies.  So why did France and Spain end up speaking Romance languages, and England a Germanic one?

The easier question is the last bit.  Britain never was as thoroughly Romanized as the rest of western Europe; it always was kind of a wild-west frontier outpost, and a great many of the Celtic tribes the Romans tried to pacify rebelled again and again.  When the Romans troops withdrew, there weren't a lot of speakers of Latin left -- exceptions were monasteries and churches.  Most of the locals had retained their original languages, and when the British Celts told the troops "Romani ite domum" (more or less), they just picked up where they'd left off.

The problem was, when the Angles and Saxons started arriving in huge numbers over the next two centuries, there wasn't a single dominant language there to stand up against them -- just a bunch of various dialects spoken by tribes that never were all that numerous, and didn't get along very well with each other anyhow.  So the West Germanic language the invaders spoke became the common language, eventually evolving into Old English.

The situation was different in France and Spain.  By the fifth century, those had both been solidly Roman for three hundred years.  The Celtic/Gaulish natives were by this time thoroughly subjugated, and many had even thrown their lot in with the conquerors, rising to become important figures.  (One example is first century B.C.E. writer and polymath Gnaeus Pompeius Trogus, who despite his Roman name was from the Celtic Vocontii tribe in the western foothills of the Alps.)  Business, record-keeping, and administration were all conducted in Latin; most of the cities were predominantly Latin-speaking.  

The Germanic tribes who swept through western Europe in the fourth and fifth centuries had an interesting attitude.  They didn't want to destroy everything the Romans had built; they just wanted to control it, and have access to all the wealth and land.  They didn't even care if the Roman town-dwellers stayed put, as long as they acknowledged the Goths' overlordship.  (Which almost all of them did, given that there were no other options.  Practical folks, the Romans.)

The invading Visigoths, Franks, and Burgundians had no written language we know of, so when they settled in to rule the place -- and most importantly, to do business with the local landowners -- their only real option was to learn Latin.  Latin became the prestige language, the language you learned if you wanted to go places, much the way English is now in many parts of the world.

The result was that Latin-derived Old French and Old Spanish were eventually adopted by the Germanic-descended ruling class, ultimately being spoken throughout the region, while the opposite pattern had happened across the Channel in England.  Interesting that the Franks gave their name to the country of France and its language, but the only modern language descended from Franconian is one spoken two countries northeast of there -- Dutch.

It's always fascinating to me to see how chance events alter the course of history.  You can easily see how it could have gone the other way -- the Visigoths might have been more determined to eradicate every trace of Romanness, the way so many conquerors have done.  Instead, they saw the value in leaving it substantially intact.  Not because they had such deep respect for other cultures -- they weren't so forward thinking as all that -- but because they recognized that they could use the Roman knowledge, language, and infrastructure for their own gain.  The result is that my Celto-Germanic ancestors spoke a language derived from Latin, even though by that time it was about the only Roman thing about them.

****************************************

The relic

The first thing I learned in my studies of linguistics is that languages aren't static.

It's a good thing, because my field is historical linguistics, and if languages didn't change over time I kind of wouldn't have anything to study.  There's an ongoing battle, of course, as to how much languages should change, and what kinds of changes are acceptable; this is the whole descriptivism vs. prescriptivism debate about which I wrote only last month.  My own view on this is that languages are gonna change whether you want them to or not, so being a prescriptivist is deliberately choosing the losing side -- but if lost causes are your thing, then knock yourself out.

Where it gets interesting is that the rates of language change can vary tremendously.  Some cultures are inherently protective of their language, and resist things like borrow words -- a great example is Icelandic, which has changed so little in a thousand years that modern Icelanders can still read the Old Norse sagas with little more difficulty than we read Shakespeare.

Speaking of Shakespeare, it bears mention that the language of Shakespeare and his contemporaries isn't (as I heard some students call it) "Old English."  Old English is an entirely different language, not mutually intelligible with Modern English, and by Shakespeare's time had been an extinct language for about four hundred years.  Here's a sample of Old English:

Fæder ure şu şe eart on heofonum, si şin nama gehalgod.  To becume şin rice, gewurşe ğin willa, on eorğan swa swa on heofonum.

I wonder how many of you recognized this as the first two lines of the Lord's Prayer:

Our Father, who art in heaven, hallowed be thy name.  Thy kingdom come, thy will be done on Earth as it is in heaven.

There's been a discussion going on in linguistic circles for years about which dialect of English has changed the least -- not since the time of Old English, but at least since Elizabethan English, the dialect of Shakespeare's time.  We have a tendency, largely because of some of the famous performances of Hamlet and Macbeth and Richard III, to imagine Shakespeare's contemporaries as speaking something like the modern upper-class in southeastern England, but that's pretty clearly not the case.  Analyses of the rhyme and rhythm schemes of Shakespeare's sonnets, for example, suggest that Shakespearean English was rhotic -- the /r/ in words like far and park were pronounced -- while the speech of southern England today is almost all non-rhotic.  Vowels, too, were probably different; today a typical English person pronounces words like path with an open back unrounded vowel /ɑ/ (a bit like the vowel in the word cop); in Shakespeare's time, it was probably closer to the modern American pronunciation, with a front unrounded vowel /æ/ (the vowel sound in cat).

Analysis of spoken English from dozens of different regions has led some linguists to conclude -- although the point is still controversial -- that certain Appalachian dialects, and some of the isolated island dialects of coastal North and South Carolina, are the closest to the speech of Shakespeare's day, at least in terms of pronunciation.  Vocabulary changes according to the demands of the culture -- as I said, there's no such thing as a static language.

[Image licensed under the Creative Commons Alumnum, Primary Human Languages Improved Version, CC BY-SA 4.0]

The reason all this comes up is that linguists have come upon another example of a dialect that preserves a relic dialect -- this one, from a great deal longer ago than Elizabethan English.  In the region of Trabzon in northern Turkey, there is a group of people who speak Romeyka -- a dialect of Pontic Greek that is thought to have changed little since the region was settled from classical-era Greece over two thousand years ago.

Since that time, Romeyka has been passed down orally, and its status as a cultural marker meant that like Icelandic, it has been maintained with little change.  Modern Greek, however, has changed a great deal in that same time span; in terms of syntax (and probably pronunciation as well), Romeyka is closer to what would have been spoken in Athens in Socrates's time than Modern Greek is.  "Conversion to Islam across Asia Minor was usually accompanied by a linguistic shift to Turkish, but communities in the valleys retained Romeyka," said Ioanna Sitaridou, of the University of Cambridge, who is heading the study.  "And because of Islamization, they retained some archaic features, while the Greek-speaking communities who remained Christian grew closer to Modern Greek, especially because of extensive schooling in Greek in the nineteenth and early twentieth centuries...  Romeyka is a sister, rather than a daughter, of Modern Greek.  Essentially this analysis unsettles the claim that Modern Greek is an isolate language."

The problem facing the researchers is that like many minority languages, Romeyka is vanishing rapidly.  Most native speakers of Romeyka are over 65; fewer and fewer young people are learning it as their first language.  It's understandable, of course.  People want their children to succeed in the world, and it's critical that they be able to communicate in the majority language in schools, communities, and jobs.

But the loss of any language, especially one that has persisted virtually unchanged for so long, still strikes me as sad.

It's a consolation, though, that linguists like Ioanna Sitaridou are working to record, study, and preserve these dwindling languages before it's too late.  Especially in the case of a language like Romeyka, where there is no written form; without recordings and scholarly studies, once it's gone, it's gone.  How many other languages have vanished like that, without a trace -- when no more children are being raised to speak it, when the last native speaker dies?  It's the way of things, I suppose, but it's still a tragedy, a loss of the way of communication of an entire culture.

At least with Romeyka, we have people working on its behalf -- trying to find out what we can of a two-thousand-year-old linguistic relic from the time of Alexander the Great.

****************************************

Speaking beauty

My novel In the Midst of Lions, the first of a trilogy, has a character named Anderson Quaice, who is a linguistics professor.  He also has a strong pessimistic streak, something that proves justified in the course of the story.  He develops a conlang called Kalila not only as an entertaining intellectual exercise, but because he fears that civilization is heading toward collapse, and he wants a way to communicate with his friends that will not be understood by (possibly hostile) outsiders.

Kalila provides a framework for the entire trilogy, which spans over fourteen centuries.  I wanted the conlang to follow a similar trajectory as Latin did; by the second book, The Scattering Winds, Kalila has become the "Sacred Language," used in rituals and religion; by the third, The Chains of Orion, it has been relegated to a small role as a historical curiosity, something learned (and mourned!) only by academics, and which few speak fluently. 

But of course, in order to incorporate it into the narrative, I had to invent the conlang.  While I'm not a professor like Quaice, my master's degree is in historical linguistics, so I have a fairly solid background for comprehending (and thus creating) a language structure.  I've mostly studied inflected languages, like Old Norse, Old English, Latin, and Greek -- ones where nouns, verbs, and adjectives change form depending on how they're being used in sentences -- so I decided to make Kalila inflected.  (Interestingly, along the way English lost most of its noun inflections; in the sentences The dog bit the cat and The cat bit the dog you know who bit whom by word order, not because the words dog and cat change form, as they would in most inflected languages.  English does retain a few inflections, holdovers from its Old English roots -- he/him/his, she/her/hers, they/them/theirs, and who/whom are examples of inflections we've hung onto.)

One of the interesting choices I had to make centers on phonetics.  What repertoire of sounds did I want Kalila to have?  I decided I was aiming for something vaguely Slavic-sounding, with a few sound combinations and placements you don't find in English (for example, the initial /zl/ combination in the word for "quick," zlavo.)  I included only one sound that isn't found in English -- the unvoiced velar fricative (the final sound in the name Bach), which in accordance with the International Phonetic Alphabet I spelled with a letter "x" in the written form; lexa, pronounced /lekha/, means "hand."

Of course, in the end I used about one percent of all the syntax and morphology and lexicon and whatnot I'd invented in the actual story.  But it was still a lot of fun to create.

The topic comes up because of a really cool study that recently came out in the journal Language and Speech, by a team led by linguist Christine Mooshammer of Humboldt University in Berlin.  The researchers wanted to find out why some languages are perceived as sounding more pleasant-sounding than others -- but to avoid the bias that would come with actual spoken languages, they confined their analysis to conlangs such as Quenya, Sindarin, Dothraki, Klingon, Cardassian, Romulan, and Orkish.

The first stanza of a poem in Quenya, written in the lovely Tengwar script Tolkien invented [Image is in the Public Domain]

The results, perhaps unsurprisingly, rated Quenya and Sindarin (the two main Elvish languages in Tolkien's world) as the most pleasant, and Dothraki (from Game of Thrones) and Klingon to sound the most unpleasant.  Interestingly, Orkish -- at least when not being snarled by characters like Azog the Defiler -- was ranked somewhere in the middle.

Some of their conclusions:

• Languages with lower consonantal clustering were rated as more pleasant.  (On the extreme low end of this scale are Hawaiian and Japanese, which have almost no consonant clusters at all.)
• A higher frequency of front vowels (such as /i/ and /e/) as opposed to back vowels (such as /o/ and /u/) correlates with higher pleasantness ratings.
• Languages with a higher frequency of continuants (such as /l/, /r/, and /m/) as opposed to stops and plosives (like /t/ and /p/) were ranked as more pleasant-sounding.
• Higher numbers of unvoiced sibilants (such as /s/) and velars (such as the /x/ I used in Kalila) correlated with a lower ranking for pleasantness.
• The more similar the phonemic inventory of the conlang was to the test subject's native language, the more pleasant the subject thought it sounded; familiarity, apparently, is important.

This last one introduces the bias I mentioned earlier, something that Mooshammer admits is a limitation of the study.  "One of our main findings was that Orkish doesn’t sound evil without the special effects, seeing the speakers and hearing the growls and hissing sounds in the movies," she said, in an interview with PsyPost.  "Therefore, the average person should be aware of the effect of stereotypes that do influence the perception of a language.  Do languages such as German sound orderly and unpleasant and Italian beautiful and erotic because of their sounds, or just based on one’s own attitude toward their speakers?"

I wonder how the test subjects would have ranked spoken Kalila?  If the researchers want a sample, I'd be happy to provide it.

It's a fun study, which I encourage you to read in its entirety.  It brings up the bigger question, though, of why we find anything aesthetically pleasing.  I'm fascinated by why certain pieces of music are absolutely electrifying to me (one example is Stravinsky's Firebird) while others that are considered by many to be masterpieces do nothing for me at all (I've yet to hear a piece of music by Brahms that elicits more than "meh" from me).  There's an emotional resonance there with some things and not others, but I'm at a loss to explain it.

So maybe I should end with a song by Enya, which is not only beautiful musically, but is sung in the conlang she invented, Loxian.  Give this a listen and see where you'd rank it.

I don't know about you, but I think that's pretty sweet-sounding.

****************************************

The language of Sark

The title of my master's thesis was The Linguistic and Cultural Effects of the Viking Invasions on England and Scotland.  I don't think many people read it other than me and my committee, but it did win the 1996 International Prize For Research With Absolutely No Practical Applications Whatsoever.  And it allowed me to learn valuable information such as the fact that there were two words in eleventh-century England for window -- one from Old English (eagþyrl, literally "eye-hole") and one from Old Norse (vindauga, literally "wind-eye") -- and for some reason the Old Norse one won and our word window comes from it rather than from Old English.

Which is a handy "fun fact" for me to bring out at cocktail parties, especially if I want everyone to back away slowly and then find other people to talk to for the rest of the evening.

In any case, I spent a good bit of my time in graduate school learning assorted random facts about western European linguistics, which was why I was a bit gobsmacked when I found out that there's a language in western Europe that I had never even heard of.  It's called Sarkese, and is only found on the tiny (1.5 by 3.5 kilometers) island of Sark, east of Guernsey in the Channel Islands.

The Channel Islands [Image licensed under the Creative Commons Aotearoa, Wyspy Normandzkie, CC BY-SA 3.0]

Sark is currently home to five hundred people, of whom only three learned Sarkese (known colloquially as patois) as their first language.  It's a Romance language -- the closest relative is French, but it's not mutually intelligible.  It came originally from medieval Norman French via the isle of Jersey; the ancestors of the people of Sark came over from Jersey in 1565 and it's been relatively isolated ever since.

The samples of Sarkese in the article I linked above illustrate how far the two have diverged in the close to a thousand years since it split from mainland French.  "Thank you very much," for example -- merci beaucoup in French -- is mérsî ben dê fê in Sarkese.  French has seventeen different vowel phonemes; Sarkese has over fifty.  Add to that the complication that the island is shaped like an hourglass, with a narrow isthmus (La Coupée) that is all but impassible during storms, and the two pieces (Big Sark and Little Sark) have different dialects.

Fortunately, a Czech linguist, Martin Neudörfl, is trying to document Sarkese, and has worked with the three remaining fluent speakers -- who are all over eighty years old -- and about fifteen semi-fluent individuals to produce a huge library of recordings, and reams of documents describing the morphology and syntax of Sarkese.  "We have hundreds of hours [of recordings] and our audio archive is outstanding," Neudörfl said.  "Even if I were to disappear, someone could revive the language just using the recordings.  We've only achieved this through years of exhaustive research.  It's all thanks to [the speakers] for sharing their knowledge."

It's always sad when a language goes extinct, and so many have done so without anyone ever recording them or writing them down.  In large part it's due to competition with more widely spoken languages; it's eye-opening to know that half of the world's individuals are native speakers of only fifteen different languages.  The other half speak one of the other seven-thousand-odd languages that currently exist in the world.  Sarkese is one of many languages that have fallen prey to the prevalence, convenience, and ubiquity of English.

On the one hand, I get why it happens.  If you want to be understood, you have to speak a language that the people around you can understand, and if you only spoke Sarkese you could communicate with eighteen other people on the island (and one Czech linguist).  But still, each language represents a trove of knowledge about the culture and history of a people, and it's a tragedy when that is lost.

So kudos to Martin Neudörfl, and the Sarkese speakers who are working with him to record this language before it's too late.  Makes me wish I'd tackled a project like this for my master's research.  I could be wrong, but I don't think Old Norse is coming back any time soon.

****************************************

The mystery of the Etruscans

One of the unresolved mysteries of European anthropology is where the Etruscans fit into the big picture.

The Etruscans lived in northwestern Italy, in the region now called Tuscany -- in fact, the name Tuscany comes from the Latin Tusci, one of several names they had for the people who lived there.  The Greeks called them the Τυρσηνοί -- the Tyrrhenians -- etymologically related both to Etruria (the region where they lived) and, obviously, the Tyrrhenian Sea that still bears their name.  They called themselves the Rasenna, a word which, like most of their language, is of uncertain origin.

The big question is whether the Etruscans were autochthonous (academia-speak for "they'd always been there") or allochthonous (migrants from somewhere else -- and if so, from where?).  Of course, the truth is that all Europeans are ultimately allochthonous, because we all started out in east Africa -- it's just that some of us have been in place for a lot longer than others.  We know the Etruscans were already in that region when the Romans got there, who encountered them in something like 500 B.C.E. and ultimately absorbed them completely.  (An occupation the Romans excelled at.)

The historian Thucydides said they were related to the Pelasgians, a bit of a catch-all term ancient Greeks used to describe the inhabitants of Greece prior to the arrival of the classical Greek-speaking Dorians, Ionians, Achaeans, and Aeolians.  The word Pelasgian was almost synonymous with barbarian -- the ancient Greeks and ancient Romans shared a rather off-putting self-congratulatory bent, summed up as "if you're not us, sucks to be you."  

Of course, they're hardly the only civilization to feel that way.  I could name a modern one or two that still haven't gotten over that attitude.

In any case, there's good evidence that the Etruscans had already been there a while when the Romans encountered them, and that they were not closely related to the people in the neighborhood.  Their language, for example, is still a mystery, and has only been partly deciphered by linguists.  The general consensus is that, like Euskara (the language of the Basque people), it is non-Indo European.  There are two other languages it seems to be related to -- the Rhaetic language, an extinct language once spoken by people in what is now eastern Switzerland and western Austria, and Lemnian, spoken on the distant island of Lemnos in the Aegean Sea prior to their being conquered by speakers of Attic Greek in the sixth century B.C.E. 

The latter suggests that Thucydides may have been right on the money in connecting the Etruscans to the Pelasgians.  Together, Etruscan, Rhaetic, and Lemnian seem to be related to no other known languages, and are tentatively classified as a linguistic isolate family (Tyrsenian).

None other than the Roman Emperor Claudius wrote a twenty-volume set on the history and language of the Etruscans -- apparently he himself was a fluent speaker, and was fascinated by their culture -- but tragically, no trace of that extensive manuscript remains.  It's one of a long list of works we only know by their titles, and through references in other books.

The Monteleone Chariot, bronze inlaid with ivory, from sixth century B.C.E. Etruria [Image is in the Public Domain]

A genetic study of Etruscan remains found that they seemed to be related to the central European Urnfield Culture -- so named because of their practice of cremation and burial in ceramic urns -- which probably originated on the steppes of eastern Europe.  But as this path was a pretty common one -- the ancestors of the Celts, Slavs, Hungarians, and Germanic peoples all came that way -- it might not tell us all that much about how or when the Etruscans arrived.

At least their later history was happier than that of many people who bumped into the Romans.  There was some warring and jockeying for power, which the Etruscans ultimately lost, but they were eventually subsumed into the Roman Republic, becoming full Roman citizens.  Many Etruscan towns went on to make large amounts of money as middlemen between the Romans and the conquered Celtic tribes to the north and west.  Several prominent families who were to rise to position of power in the Republic (and later Empire) had Etruscan roots, including the Caecinia, Urgulania, Tarquinia, and Volumnia families, all names that will be familiar to aficionados of Roman history.  Most of the people from modern Tuscany have Etruscan roots, indicating their ancestors have been living in the same place for over three thousand years.

In the end, though, we're left with a mystery.  A people who left behind buildings and works of art and an only partly-understood language, whose connections to other ancient peoples are lost to the shadows of time.  And a mystery is always fascinating -- even if we might never fully discover the answers to all the questions.

****************************************

One language to rule them all

The aphorism "No matter what you know, there's always more to learn" is something you'd be likely to see on one of those cheesy "motivational posters" that cheery type-A personalities like to pin up on office walls, but there's a lot of truth to it.  I rather prefer the formulation credited to Socrates -- "The more I know, the more I realize how little I know."

I ran into a fun example of this principle a while back, when a member of the online linguistic geekery group Our Bastard Language posted an article from The Public Domain Review called "Trüth, Beaüty, and Volapük," about a constructed language (or "conlang," in the lingo of the field) called Volapük that I had never heard of before.

My M.A. is in linguistics, but my field of study was historical/reconstructive linguistics (my thesis was about the effects of the Viking invasions on Old English and Old Gaelic, and should have won some kind of award for research that has absolutely no practical application).  But even though conlangs aren't my specialty, I've always had a fascination from them, and in fact I created a conlang called Kalila as a part of my soon-to-be-released novel In the Midst of Lions.  There are a remarkable number out there, from the familiar (Esperanto, Klingon, Elvish) to the obscure but fascinating (such as John Quijada's Ithkuil, which attempts to express concepts in a combinatory way from the smallest possible number of root words).

A sample of Tolkein's lovely Elvish script [Image is in the Public Domain]

But despite my interest in conlangs, I had never run across Volapük, which is strange because next to Esperanto, it's apparently one of the most studied constructed languages ever created.  It was the invention of a German priest named Johann Schleyer, who not only wanted to create a regularized speech that came from familiar roots (to Europeans, anyhow) and was easy to learn, but was also "beautiful sounding."  Schleyer had an inordinate fondness for umlauts, which he added because he thought that "A language without umlauts sounds monotonous, harsh, and boring."

Which reminds me of the opening credits in Monty Python and the Search for the Holy Grail, especially the "A mööse once bit my sïster" part.  One of Schleyer's contemporaries couldn't resist poking some gøød-natured fün at him over his umlautophilia, and published the following limerick in the Milwaukee Sentinel:

A charming young student of Grük
Once tried to acquire Volapük
But it sounded so bad
That her friends called her mad,
And she quit it in less than a wük.

To my ears, it doesn't sound bad at all, and kind of has a Scandi-Slavic lilt to it. Here's a sample:

The author of the article in The Public Domain Review, Arika Okrent, attributes the relative failure of Volapük to its plethora of umlauts and the easier word roots of its competitor Esperanto, which currently has about two million fluent speakers (an estimated one thousand of whom learned it as their first language).  I'm a little doubtful about that; certainly umlautiness hasn't discouraged anyone from learning Finnish.  I think it's more that the idea of a universal language is one of those high-flown ideals that won't ever catch on because most people are going to be resistant to giving up their native tongue in favor of an invented system of speech, however easy it is to learn.  Language is such a deep part of culture that to jettison our own mode of communication runs counter to every social instinct we have.  (Note that one of the most common things conquerors do to conquered people is to outlaw the speaking of the native language -- it's a sure way to deal a death blow to the culture.)

Even so, I find the whole conlang thing fascinating, and was tickled to run across one I'd never heard of.  Back in my teaching days I every so often had students who participated in an independent study class I offered in introductory linguistics, and the final project was to invent the framework of a language -- a phonetic and phonemic structure, morphological scheme, and syntax, along with a lexicon of at least a hundred words.  They then translated a passage from English into their language.  (One of the best ones I've ever seen involved a charming translation of Eric Carle's The Very Hungry Caterpillar.)

The result of this project was twofold -- students found out how hard it is to create a realistic language, and they learned a tremendous amount about the structure of our own language.  And that's just from producing a rudimentary skeleton of a language.  For people like Schleyer, who created a rich and fully functional language, it was the result of many years of devotion, hard work, and love for language itself.

So it's kind of a shame that people didn't appreciate Volapük more.  Schleyer's dream of having a language that would bring the entire world together in a common mode of communication may be as far off as ever, but even so, it's a beautiful dream.  Even if it would mean making friends with the mäjestïc ümlaüt.

****************************************

Resurrecting a fossil

A year ago I wrote about linguistic isolates -- single languages, or small clusters of related languages, that have no apparent relation to any other language on Earth.  The problem, of course, is that being spoken by only a small number of people, these are some of the most endangered languages.  There are many of them for which the last native speakers are already elderly, and there's a high risk of their going extinct without ever being thoroughly studied.

And as I pointed out in my post, the sad part of that is that each one of those is a lens into a specific culture and a particular way of thinking.  Once lost, they're gone forever, or only exist in scattered remnants, like the fossils of extinct animals.  What you can reconstruct from these relics is perhaps better than nothing, but still, there's always an elegiac sense of what we've lost, and what we're still losing.

This topic comes up because of an article in Smithsonian sent to me by a friend and frequent contributor of topics to Skeptophilia, about some linguists who are trying to reconstruct the extinct indigenous Timucuan language of northern Florida.  Timucuan was a linguistic isolate, and seems to be unrelated to the languages spoken by neighboring groups (such as the Seminole, Muscogee, and Choctaw).  The Timucua people, which at the time of European contact in 1595 comprised an estimated 200,000 people in 35 chiefdoms, each of which spoke a different dialect, was decimated by war and by diseases like smallpox.  By 1700, there were only about a thousand Timucuans left, and the slave trade eradicated those few survivors.  There is currently a genetic study to see if some populations in Cuba might be the descendants of the Timucuans, but so far the results are inconclusive.

This would just be another in the long list of complete and irretrievable cultural loss, if it weren't for the efforts of linguists Alejandra Dubcovsky and Aaron Broadwell.  Working with a handful of letters written in Timucuan (using the Latin alphabet), and a rather amazing bilingual document by Spanish missionary Francisco Pareja called Confessions in the Castilian and Timucua Language, With Some Tips to Encourage the Penitent, they have assembled the first Timucuan dictionary and grammar, and reconstructed how a long-gone people spoke.

A page from the Confessions, with Spanish on the left and Timucuan on the right [Image courtesy of the John Carter Brown Library, Brown University]

Which is incredibly cool, but there's also a wryly amusing side to it, because with Dubcovsky's and Broadwell's knowledge of the Timucuan language, they're able to compare what Pareja wanted the translators to say with what they actually did say.  "Our favorite is the description of marriage," Dubcovsky said.  "The Spanish side asks very clearly, 'Have the man and a woman been joined together in front of a priest?'  And the Timucua version of that sentence is, 'Did you and another person consent to be married?'  The Timucua translation not only takes out any mention of gender, but it also removes any mention of a religious officiant.  A priest did not write this, because a priest does not forget to include himself in the story."

So the Confessions document is not only a Rosetta Stone for Timucuan, it gives us a fascinating window into how the Timucuan translators saw the Spanish Catholic culture that was being imposed upon them.

It's tragic that this language and its people were so thoughtlessly (and ruthlessly) eradicated; worse still that such tragedies are all too common.  So it's all the more important that people like Dubcovsky snd Broadwell work to resurrect these extinct languages from the scant fossils they left behind.  It can't ever repair the damage that was done, but at least allows us to glimpse the minds of an extinct culture -- and to honor their memory in whatever way we can.

****************************************