Pages

Saturday, June 25, 2011

How fo Speak Pidgin

OMG this girl is so funny! :D

Mahalo to youtube user masterhide.

Wednesday, May 4, 2011

Theories on the origin of language diversity

There are two theories:

1. Monogenesis: belief in one original language, also known as the Mother Tongue theory. This belief may be grounded in scientific theory—it is often intertwined with the “Out of Africa” theory—though it is also favored by those with a belief in the divine origin of humanity and human language.

2. Polygenesis: belief in multiple original languages. It is also called the Candelabra theory, in which the world’s distinct language families are thought to have emerged from separate mother tongues.

There has been much rambling speculation on monogenesis, though claims which name individual languages are normally offshoots of cultural biases. Such assertions range from Greek, Sanskrit, German (a legacy of WWII-era Nazi propaganda, perhaps), and Hebrew. In a more amusing twist, the 17th-century Swedish philologist Andreas Kemke once satirically remarked that in the Garden of Eden God spoke Swedish, Adam spoke Danish, and the serpent spoke French.*

Perhaps a slightly more convincing claim (and I mean only very slightly more convincing) is that Basque was the world’s first language. Since it is a language isolate, it would be harder to discredit this assertion, though the chances are still very slim. Even though the claim cannot be scientifically negated, the initial assertion is itself unscientifically founded.

In the case of either scientific monogenesis or polygenesis, it would be impossible to identify the world's first language.

-------------------------------------------------------------------------
*This statement, of course, had political motives.

Saturday, April 16, 2011

Quentin Atkinson on human expansion and phoneme diversity

I believe in holding a fair amount of skepticism even when faced with a claim that could be legitimate to some degree. So I will have to personally look into the following claim before speculating on it any further:

"Human genetic and phenotypic diversity declines with distance from Africa, as predicted by a serial founder effect in which successive population bottlenecks during range expansion progressively reduce diversity, underpinning support for an African origin of modern humans. Recent work suggests that a similar founder effect may operate on human culture and language. Here I show that the number of phonemes used in a global sample of 504 languages is also clinal and fits a serial founder–effect model of expansion from an inferred origin in Africa. This result, which is not explained by more recent demographic history, local language diversity, or statistical non-independence within language families, points to parallel mechanisms shaping genetic and linguistic diversity and supports an African origin of modern human languages."

- Quentin D. Atkinson (source: http://www.sciencemag.org/content/332/6027/346.abstract)

I will update you when I have looked into this further. Though I am aware of a few isolated cases that support this theory (such as those mentioned only vaguely in this article: http://www.huffingtonpost.com/2011/04/15/human-language-africa_n_849809.html), I will reserve a few doubts because:

1. Even if there is some basic truth to it, this claim could be oversimplified-- not that I think Atkinson necessarilly meant it to be interpretted that way. You cannot expect this claim to be 100% accurate down to the very ratio of kilometers from Africa to the number of phonemes in a language, like a ripple effect in which a certain location in Africa is the origin and the expanding migration "rings" simply lose their phonemes as they keep going outwards. Migrations did not work that way. The same group could have, over hundreds and thousands of years, gone east, then north, and then divided themselves with one group going back west and the other continuing south.

2. What languages did Atkinson base his studies on? How do we know his research was not unbalanced?

3. There may be other factors involved than the actual distance from Africa (such as time or environment).

But despite a few doubts, I leave space in my imagination for this interesting possibility. If there is in fact a significant amount of accuracy to this claim, then the next question is why do languages lose phonemes when migrations expand outwards from Africa?

Monday, April 11, 2011

Sign Pidgins and Creoles

I have lately been reading about sign languages, something I had not really looked into much before. In the course of my readings, I became aware of not only linguistically-relevent processes of sign language development, but also on historical issues faced by the deaf and/or their communities. I have learned a little more also on how the two impact one another.

In the Nicaraguan case, deaf individuals were often sheltered, neglected, and isolated from other deaf persons. They were kept at home, and their families were embarrassed by them. This meant that: 1. they did not have a community to communicate with, which meant that 2. they did not develop grammar in early childhood when it is the opportune time to do so, which meant that 3. for the rest of their lives, they could no longer communicate fluently with anyone, not even other deaf people, except through crude mimickry.

After the Sandinista Revolution of the 1970s, a school for the deaf was established in Managua for the first time. Though Spanish-language immersion was not always a success, an astonishing new occurance was taking place: by simply bringing deaf people together, which was never the case in prior decades, deaf people began communicating with each other for the first time.

Each student brought with them from home his or her own signs from home. From that, the children began incorporating these into a signed pidgin.

Because this was still a makeshift language of crude signs, these first-generation deaf signers did not develop grammar either. But this was just a crucial first step for the generations that would follow. Where there was no grammar, no fluidity of signing, the younger children gave it both and as a result came up with their own language.

Other cases like Martha's Vineyard and the Bedouin village of Al-Sayyid, the developped in the same way, were different in that deaf people were an integral and accepted part of the otherwise hearing community. In both cases, an unusually high rate of hereditary deafness is/was present in society. All people, hearing and deaf alike, can and do sign.

As any of Al-Sayyid's 3,500 villagers will tell you, their ancestor married an Egyptian woman and settled in Palestine. They had five children, two of whom were deaf. All of the people of Al-Sayyid are descended from the deaf brothers.

Evidence for the fact that Al-Sayyid is a very recent new language: it only has two color words, "white" (white, reds, oranges, yellows) and "black" (black and cool colors). This is significant because, according to established linguistic research, there is a timeline for color-word development in languages. New languages have a white-black color categorizing system. The next step is normally a white-black-red system. And so more color words become standard as the language gets older. The fact that there are only two set color-words indicates that Al-Sayyid's sign language is a fairly recent language, as it's ancestral story also indicates.

In more recent years, the deaf children of Al-Sayyid have attended an Israeli school for the deaf, where they pick up words from Israeli Sign Language and bring them home; they even have words like "purple" and "green" that are unknown to older deaf generations. While conducting research in the village, linguists have had to pair hearing children with non-hearing children to observe their communication, so that there would be no interference from ISL (the hearing child, not having attended the school for deaf children, will only know the native sign language of his village, and the deaf child will then have to only use the native sign language to communicate with a hearing person).

We have discussed the immersion of signed creoles/fluent sign languages from signed pidgins. Now I will give an example of the opposite, a sort of ignorant repression of language.

Gallaudet University, openned in the 1800s, is the first school in the United States that was established for deaf students, and which employed deaf instructors. At some point, the university came under the control of hearing people who then subjected the school to Victorian biases about deafness and sign language. There were countless arguments that sign language was an inferior language, and many forced efforts to assimilate deaf people into hearing society. Deaf professors were replaced with hearing ones, who failed to communicate with their students.

The hearing professors of Gallaudet, upon observing their subjects, noticed the vivaciousn use of facial gestures. The professors tried to stop the deaf students from using face gestures while signing, deeming it innapropriate (they still did it on their own terms outside of the classroom). What these instructors just did was to remove all grammar from the language (while the hands sign words, the face carries grammar in many of the world's sign languages). Stripped of all grammar, the result was a rough language that had a lessened ability to communicate effectively. A creole had been irregularly demoted to a cruder form of signed pidgin* at Gallaudet University; despite their good intentions the instructors were doing very little to help their students.

Victorian-age prejudices were such that deaf people simply could have a real language. It was best if they learn to integrate and don't produce more deaf children.

In cases like Al-Sayyid, the attitudes towards deafness are absolutely reverse. It is not only common, but quite unremarkable for someone to be deaf. In a village where it is rare to marry anyone from outside of the village, and where cousin-marriage is a strategic way to preserve land ownership, inheritted deafness crops up frequently in families and stays in these families. Because of this accepted fact of life, deaf people have never been alienated in this society, and their language has become embraced rather than repressed.

(*: Remember, the term "pidgin" when used in a scientific (linguistic) context means a makeshift communication system that lacks grammar (often as a result of speakers of different languages needing to communicate with one another), whereas a "creole" differs in that it has a complete grammar system that had been deveolopped from and applied to the former pidgin. In the real world (non-linguistic/scientific language), these terms may be used a little more loosely, and there is nothing wrong with that. I just want to make the distinction between my use of words in a "real-world" or a scientific context, so as not to be horribly misunderstood.)

Sunday, March 27, 2011

Notes on Language and Brain

On a recent visit to the library, I came across How Language Works, by David Crystal. In chapter 27, How the brain handles language, Crystal simplifies how language is interpreted in the mind and how the mind constructs language. I will make notes on what he has to say in this chapter (the images are scanned and uploaded from the chapter as well):

Cerebrum: the largest region of the brain. It is divided into the left cerebral hemisphere and the right cerebral hemisphere.

Brain stem: connects the cerebral hemispheres to the spinal cord. The brain stem consists of the mid-brain, the pons, and the medulla oblongata.

Cerebellum: responsible for body posture and the smooth coordination of all body movements. The cerebellum is located at the back of the pons.

(Cerebral) Cortex / Gray Matter: Nerve cells found on the surface of the cerebrum. The cortex is involved in the control of voluntary movements and intellectual functions, and the decoding of information from the senses.

White Matter: fiber tracts that transmit signals between the different parts of each hemisphere, and between the cortex and the brain stem.

Median Longitudinal Fissure: separates the cerebral hemispheres.

The brain is divided into four main lobes:
1. frontal
2. temporal
3. parietal
4. occipital


Theory of Cerebral Localization: "The idea that a single area of the brain can be related to a single behavioural ability, such as vision or speech."

Support for this theory came from the work of Paul Pierre Broca and Carl Wernicke (both were neurologists who found that damage done to certain parts of the brain results in reduced linguistic and/or motor capabilities; please see my previous blog post about aphasia, the loss of language due to brain damage).

Other areas of the brain which are associate with speech, listening, reading, writing, and signing are located in the image below:


- The front part of the parietal lobe, along the fissure of Rolando, is primarily involved in the processing of sensation, and may be connected with the speech and auditory areas at a deeper level.

- The area in front of the central fissure (aka the fissure of Rolando) is mainly involved in motor functioning, and is thus relevant to the study of speaking and writing.

- An area in the upper back part of the temporal lobe, extending upwards into the parietal lobe, plays a major part in the comprehension of speech (This is Wernicke's area).

- In the upper part of the temporal lobe is the main area involved in auditory reception, known as Heschl's gyri, after the Austrian pathologist R. L. Heschl.

- The lower back part of the frontal lobe is primarily involved in the encoding of speech (This is Broca's area).

- Another area towards the back of the frontal lobe may be involved in the motor control of writing. It is known as Exner's centre after the German neurologist Sigmund Exner.

- Part of the left parietal region, close to Wernicke's area, is involved with the control of manual signing.

- The area at the back of the occipital lobe is used mainly for the processing of visual input.

NEUROLINGUISTIC PROCESSING (Here is the amazing part):

- Speech production: Wernicke's area (utterance structure) --> Broca's area (encoding) --> adjacent motor area (articulation)

- Speech comprehension: speech --> auditory cortex (speech reception) --> Wernicke's area (interpretation)

- Reading aloud: written form --> visual cortex (reception) --> angular gyrus --> Wernicke's area (auditory representation) --> Broca's area (processing)

Saturday, March 5, 2011

Language Isolates

I hesitate to use the term “isolate” in this context. Past, present, future, any language that is an “isolate” could possibly (actually very likely) be/has been/become part of a larger group. Some “isolate” languages have only become so when all other family members die off. Some languages are “isolate” because linguists are not aware of any languages that are/were related to it. Does this then mean that such a language simply sprung out of nowhere? Presumably all language is derived from language, so the term “isolate” is actually misleading.

I will put so-called “isolated” languages into two categories:

A) A language for which there is no proof or disproof that any other languages are related to it, one that is completely unrelated to any known language in existence (past, present, future). Although such a language may be presumed to have had, at the very least, a predecessor, the fact that there is no evidence to support even that much makes it classified as an isolate.
B) Formerly had known language relatives that became obsolete over the course of history.

A few type A language isolates include Korean, Basque (Spain), Ainu (Japan), Etruscan (Italic Peninsula, extinct), Sumerian (Mesopotamia, extinct), and Zuni (SW US).

Pirahã (Brazil), is the last surviving member of the Muran Language Family (and so it is a type B language isolate), and only has about 200 active daily speakers. I will reference this language in another post which I am scheming.

“Japanese” was once considered a language isolate. That was before the realization that Ryukyuan languages (such as Okinawan) are actually different languages from Japanese rather than a dialects of it, though they are still related (I will later post about how and why languages that are different from one another come to be considered different dialects of the same language, and why different dialects are sometimes considered different languages). So the umbrella term of “Japanese” has become the Japonic Language Family, a.k.a. the Japanese-Ryukyuan Language Family, to include Japanese and the sub-branch of Ryukyuan languages.

Some linguists believe Korean and Japonic Languages to be part of the Altaic Language Family (which includes the Turkic, Mongolic, and Tungistic subgroups) but the general consensus is that they are isolates. This may also become another post for another time.

And as for the languages that have no known relatives in existence…where did they come from? From what were they derived? That is the mystery that continues to intrigue historical linguists but which no amount of research seems to resolve.

Monday, February 21, 2011

A Linguist's Overview of Numeric Digit-Base Systems

From the title you may think this is a mathematics post, but don’t be fooled. This post is about the language of math, and more precisely the use of different digit-base systems in different languages; languages may differ not only in the words they use to count, but also in their fundamental system for counting and therefore conceiving of mathematical notions (though the final, ultimate understanding of course remains the same).

Most modern societies use a base-10, or “decimal” system:

0 1 2 3 4 5 6 7 8 9
10 …

A few languages that use(d) a decimal (base-10) system: Incan languages like Quechua and Aymara, Chinese languages, Vietnamese, Japanese, Korean, Thai (Japanese, Korean, and Thai decimal systems were adopted from the Chinese system).

The Hindu-Arabic numeral system, first developed in India, is a base-10 system, corresponding of course with languages that use a base-10 counting system.

Variations of Hindu-Arabic numerals are used in places that speak the following languages (but are not limited to them, especially in the modern age): Western languages, Arabic, Farsi, Indian languages, Burmese, Khmer, and Thai.

If we have a base-10 system in the English language, then why do “eleven” and “twelve” seem to be independently named? By the logic of the decimal system, shouldn’t eleven be called “oneteen” (one-ten) and twelve “twoteen” (two-ten)? They are, after all, followed by thirteen (three-ten), fourteen (four-ten), and fifteen (five-ten), right?

Although many European societies, such as those in the British Isles, used a duodecimal, or base-12 numeric system for measurements (one foot = 12 inches), monetary units, and time (12 months, 12 zodiac signs), the words eleven and twelve are actually thought to originate from the Proto-Germanic “ainlif” (one is left) and “twalif” (two are left). Still, in modern English this comes across as an irregularity.

French, although decimal-based, has the irregular use of base-20 for numbers 80 through 99. 80 is called quatre-vingt (4 x 20). 85 is quatre-vingt cing (4 x 20 + 5), 90 is quatre-vingt dix (4 x 20 + 10), and 95 is quatre-vingt quinze (4 x 20 + 15).

The Sumerians strategized a base-60, or sexagesimal system that was later passed on to the Babylonians. This system was used (and in many cases, is still used) to measure time (60 seconds in a minute, 60 minutes in an hour), angles, and geographic coordinates.

This base-60 system was not, however, derived from the Sumerian language itself, and the cuneiform numeric system was actually a decimal system.

The Duodecimal System (Base-12):

0 1 2 3 4 5 6 7 8 9
A B 10 11 12 13 …

Here, A=10, B=11, 10=12, etc…alphabetic letters represent numbers from 10 to the duodecimal base of 12. Numbers eleven and twelve have independent names in languages that employ the duodecimal system.

Languages that use a duodecimal (base-12) system are rare. Chepang (spoken in Nepal), Mahl (spoken on Mincoy Island, India), and Nigerian languages such as Janji, Gbiri-Niragu, and the Nimbia dialect of Gwandara all use a base-12 number system. J.R.R. Tolkein’s Elvish languages also use a duodecimal system.

The Mayans (and other Mesoamerican cultures) used a base-20 system:

0 1 2 3 4 5 6 7 8 9
A B C D E F G H I J
10 11 12 13 ….

In this system “10” represents 20. Alphabet letters are here used to represent the numbers that are between 9 and the Mayan digit-base number 20; what we call the “teen” numbers would not have been known as “sixteen” (six-ten) or “seventeen” (seven-ten) but would have had their own independent name.

Other digit-systems used in different languages include the following:

Base-4: Chumashan languages (native to southern California). Number names were structured according to multiples of 4 and 16.

Base-5 (quinary): Gumatj, Nunggubuyu, and Kuum Kopan Noot are all quinary languages indigenous to Australia. Saraveca, an extinct language spoken in what is now Bolivia, was also quinary.

Base-15: used in the Huli language of Papua New Guinea. In Huli, fifteen is called ngui, thirty is ngui ki (15x2), and 225 is ngui ngui (15x15).

Base-24: Kakoli (a.k.a. Umbu Ungu, of Papua New Guinea). Tokapu means 24. 48 is tokapu talu (24 x 2). 576 is tokapu tokapu (24 x 24)

Base-27: Telefol and Oksapmin, also in Papua New Guinea.

Base-32: Ngiti (Democratic Republic of Congo) is base-32 with base-4 cycles.

Computers use a binary, or base-2 system to store data:

0 1 10 11 100 101 110 111 1000 1001

0 1 2 3 4 5 6 7 8 9

The top row represents the computer’s base-2 digit system, and the bottom row is there so you can compare it to the base-10 system.

Computer programmers, who are after all human, use a hexadecimal (base-16) system instead of the binary system:

0 1 2 3 4 5 6 7 8 9
A B C D E F 10 …

(10=16 here)

Web page colors are based on this system. Combinations of red, green, and blue (the primary colors of light) are represented as (R,G,B) = #RRGGBB. For example, #000000 = (0,0,0) = 0 Red, 0 Green, 0 Blue = WHITE; #FFFFFF =(256 Red, 256 Green, 256 Blue) = BLACK. Likewise, #FF0000 = red, #00FF00 = green, and # 0000FF = blue.

Those of us who are not computer programmers would have a difficult time with the hexadecimal system. Likewise, I imagine it must be difficult for a language learner trying to crossover from one digit-base system to another. I am sure this has happened in history where two or more cultures have collided. Would merchants have argued in an open marketplace over basic trade units? Would colonialist pedagogues have been puzzled by their native subjects’ seeming lack of comprehension in basic math? They may have been unaware that what appears to be a logical brick wall is in fact a linguistic barrier.

I hope this post has done more than make you realize that certain languages use different digit-base systems than our familiar base-10. Depending on what digit-base is used in your language’s counting system, individual numbers can take on a different meaning or level of significance. Think of how many times you refer to the number ten. “I’ll be there in ten minutes,” or saying to an obnoxious child “I will count to ten.” Or that demonical professor who wants a ten-page essay on quantum physics. Ten is just a handy, familiar number in our minds, and this is only because we speak a language whose number system is ultimately based on the number of fingers we have. If we were Huli speakers, 225 would be more significant than 100, and your professor would want a 15-page essay instead of ten pages. And if we spoke a base-8 language, we might celebrate our 64th birthday as though it were our 100th.

-------------------------------------------------------------------------------------

A book I came across in my research for this post:

Numbers, Language, and the Human Mind
By Heike Wiese

I found this on books.google.com, so if you are interested you may also type in the name/author of this book at that search engine.

Friday, February 11, 2011

Where do zombies come from?

Did you know that the English word zombie has African roots? The word-concept nzumbi (ghost, spirit) was first brought to the Caribbean by slaves who spoke Kimbudu (spoken in Angola). In the Carribean it developped into the creole word zonbi; the word also evolved into a superstition of a corpse that could be manipulated by supernatural power and made to obey commands.

nzumbi (Kimbudu) > zonbi (Carib. Creole) > zombie (English)

Incidentally, a "zombie" is also a voodoo snake deity of West African origin in Haiti, parts of the southern United States, and Brazil. I do not know if this is related to the Kimbudu-origin "nzumbi" (with lack of evidence, I tend to think not), but it is an interesting coincidence nontheless.

(http://kawann.k1.online.fr/atilye-de.htm)

Tuesday, February 8, 2011

Aphasia—Loss of Language

Aphasia is the loss of language due to damage to part or parts of the brain that are responsible for language. Damage may be brought about by head injuries, strokes, tumors, infection, or dementia.

Damage to the left or right temporal lobe (though normally in the left) results in Wernicke's Aphasia (fluent aphasia). A person with Wernicke's Aphasia will speak in long sentences that are hard for others to understand (it is also hard for such a person to understand others). They may add unnecessary or even invented words to their speech. Because the left temporal lobe is not close to any motor movement areas of the brain, Wernicke's Aphasia is not usually accompanied by weakness or paralysis of the body.


Broca's Aphasia, a type of non-fluent aphasia, occurs with damage to the frontal lobe. A person with Broca's Aphasia speaks in short, concise sentences that eliminate small words like "is" or "the." Though strained to speak, an affected person has no comprehension problem when listening to others. Broca's Aphasia is frequently accompanied by right-side weakness or paralysis, since the frontal lobe also controls motor movement.

Global Aphasia (non-fluent) is a result of damage in extensive areas of the brain's language areas, resulting in limited speech and comprehension abilities and extreme difficulties with communication.

Below are some examples of sentences that may be spoken by persons with aphasia. Remember, while Wernicke's is characterised by long, drawn-out speech with superfluous or made-up words, Broca's speech is identified by its brevity.

Wernicke's:

"You know that smoodle pinkered and that I want to get him round and take care of him like you want before.”

Broca's:

"Walk dog.” (meaning "I will take the dog for a walk.")
"Book book two table." (meaning "There are two books on the table.")

[Source for all above information: National Institute on Deafness and Other Communication Disorders' website, at this url: http://www.nidcd.nih.gov/health/voice/aphasia.asp]

As you can see, not every aspect of language is controled by one single area of the brain, in fact different aspects of language are controlled by parts of the brain that are separated from one another. I believe this to be an advantageous development, since in the event of a head injury causing a form of aphasia the chance of losing all language capabilities whatsoever is lessened. That is my best guess, not actually being a neurologist myself.

After reading about this I would also be interested to know 1. How do the mechanics of the brain work in regards to language and communication, and 2. How and to what effect do the language-areas of the brain correspond with motor skills? If you are reading this and you know anything about this, please feel free to share your knowledge.

Saturday, February 5, 2011

Language and Dialect Quote

"One should not think that a dialect is a "badly spoken" language: it is indeed a full-fledged language, with a specific grammar and lexicon. If however it is said today that Italian, Spanish or French are languages, it is not because they are richer, handsomer or better structured than Leonese, Napolitan or Picard, but because they have acquired more prestige by becoming literary and official languages for constituted States. The idioms that were the starting points of these three languages, respectively the Tuscan dialect of Florence, the Castillan dialect and the dialect of the Ile-de-France, were themselves, originally, but dialects spoken on a reduced geographical area."

- Henriette Walter, in L'aventure des langues en Occident, p132.

Friday, February 4, 2011

Colorless Green Ideas: The Experiment

"Colorless green ideas sleep furiously."
- Noam Chomsky

This revolutionary sentence was composed by Noam Chomsky in 1957 to demonstrate that, contrary to popular opinion of the time, grammar is NOT the fundamental underlying structure of language. The sentence, though in agreement with the grammatical structure of Standard English, is nonsensical. In other words, meaning is not necessarily dependent on sentence structure.

I will break down Chomsky's sentence structure word-by-word: Adjective. Adjective. Subject. Verb. Adverb.

Now I will write a new sentence that makes more concrete sense but which follows the exact same grammatical structure:

"Colorful little fish swim quickly."

What I will do next is to mix up the word order of my sensical, structurally sound (by mainstream English standards) sentence and see if meaning may still be derived from it:

"Quickly swim fish little colorful." (Adv. V. Sub. Adj. Adj.)
"Fish colorful little swim quickly." (Sub. Adj. Adj. V. Adv.)
"Little fish colorful quickly swim." (Adj. Sub. Adj. Adv. V.)
"Swim little colorful fish quickly." (V. Adj. Adj. Sub. Adv.)
"Colorful swim quickly little fish." (Adj. V. Adv. Adj. Sub.)

What these sentences do is to make a mishmash of ideas (colorless green ones, of course) that form a picture in the mind of, well, colorful little fish swimming quickly. DEFINITE MEANING MAY STILL BE DERIVED FROM A NON-STRUCTURALLY SOUND SENTENCE, HOWEVER NOT FROM A STRUCTURALLY SOUND SENTENCE THAT HAS NO DEFINITE MEANING IN ITSELF.

If I were to alter the sentence "Colorless green ideas sleep furiously" in the same way, it would also make about as much sense as in its original form. :)

Thursday, February 3, 2011

Intro to Assyriology

For this post I will draw a few basic points from this interview with Pennsylvania State University's Assyriologist Gonzalo Rubio: http://www.livescience.com/history/dead-languages-sumerian-akkadian-assyriologist-q-a-101213.html

Sumerian and Akkadian were the languages of Ancient Mesopotamia, which is approximately the region of present-day Iraq. They were spoken during the Bronze Age in the "Cradle of Civilization." Ancient Mesopotamians had the first complex urban civilization in the world that gave rise to writing and advances in art, science, mathematics, and politics.

Sumerian was first spoken around 1300 BC (5000 years ago). Sumerian and Akkadian are both dead languages. They employed a writing system called cuneiform.

Sumerian is very similar in structure and grammar to other, more modern Semitic languages like Arabic and Hebrew. Akkadian, on the other hand, has a completely different structure; Rubio says that Akkadian sentence structure is comparable to the sentence structure of Japanese, Turkish, Finnish, Hungarian, and many Native American languages. This does not mean that Sumerian is related to any of those languages, it is only a comparison based on one single, very specified aspect of language.

Rubio and other Assyriologists analyze documents such as myths, legal texts, economic documents, poems, sales receipts, scientific and scholarly texts, and letters.