Writings Systems II
Assignment for November 7, 2006
Go to omniglot (http://www.omniglot.com/writing/definition.htm) and examine the writing systems from logographics to undeciphered systems. Look for information on the Rosetta Stone and its decipeherment. Based on that experience write a blog entry on how you would go about making sense of a system that you were seeing for the first time.
posted by TRB | 11:14 AM
13 Comments:
Assuming I was given no “Rosetta Stone” to begin decoding this unfamiliar script, I’d first try to figure out which language the script was supposed to represent. In an ideal situation, this would make the decoding process considerably easier, as I could look for spelling trends and other letter occurrences in the given language. However, not knowing the language of an undeciphered script is often most of the reason it remains undeciphered. If I couldn’t figure out what language it was, I’d then compare it to languages with which I was familiar (for example, if I had some knowledge of Hebrew, Greek, or Sanskrit), to find additional patterns. From there, I’d work on identifying the type of writing system the symbols represented; if the writing looked like a continuous script, it would most likely be some type of alphabet or syllabary; if the writing looked like pictures or individual shapes, it would most likely be a logographic or pictographic system. However, these two are not mutually exclusive, as Jean François Champollion, the translator of the Rosetta Stone, discovered. Champollion noted that the Egyptian hieroglyphs on the Stone were both ideograms and phonograms, meaning they both symbolized ideas and concrete objects, and represented individual sounds.
If the writing system were an alphabet or had characters that represented sounds, I would look to identify its vowels (if it had any) before tackling its consonants. If the writing system were logographic or pictographic, I would reference cultural points from the area in which the text was discovered, like agricultural terms, geographic features, or human characteristics. Next, just as Champollion did, I would determine if certain configurations of characters could give them a specific meaning; in his translation of the Rosetta Stone, Champollion realized that hieroglyphs written in cartouches spelled the names of pharaohs. (I doubt this exact situation would occur, but the location, size, or style of the characters, as well as the use of special markings and diacritics, could signify a special meaning). Another important step would be deciding in which direction the language was written: left to right, right to left, top to bottom, bottom to top, vertically, horizontally, as a boustrophedon, etc. If I discovered the type and characteristics of the writing system, and the direction in which it was written, I’d then be able to start attempting a translation. Obviously, something like the Rosetta Stone would make the process much easier, since I would at least have a solid starting point, but taking these other factors into consideration could still bring me closer to understanding the mysterious script.
Sources used:
http://www.clevelandart.org/archive/pharaoh/glyphs.html
http://en.wikipedia.org/wiki/Rosetta_stone
http://www.gregvilk.com/
If I came across a system I had never seen before, I suppose that the first thing I would have to do is a whole lot of research. I would want to know where the sample came from and when it was created. If it was an historic sample, I would want to know as much as I could about neighboring, contemporary systems, as well as systems from other eras in the same area. Hopefully that would get me grounded in the mindset that was probably used when the system was formed, i.e. if surrounding systems were abjads, I would guess not to look for a complete representation of vowels. I would also compare these systems to see if I can find any symbols or clusters that appear similarly in my target system. Any clue that would guide me to know what patterns I am looking for would be helpful. Essentially, this means anything that anyone has already discovered about my target system or its neighbors would be taken into consideration.
If, on the other hand, someone handed me an encrypted message that he or she had created, I would have to know some slightly different things. Firstly, what languages does this person know? In all likelihood, he or she would use a system based on what he or she knows very well. Also, how much time did this person put into this system, and how creative is this person? Knowing these, I could gauge the complexity of the system and how far it might deviate from the systems with which the author is acquainted.
With my clues assembled, I would be able to start confronting the sample directly. I would attempt to break it up into its smallest recognizable parts to see what building blocks I am dealing with. Then, I would look to see if there are any repeated patterns formed by these parts. For example, if I had deduced that an apple, a bird and a crown are each indivisible elements of the system, and I saw crown/apple/bird many times in the sample, I would conclude that crown/apple/bird is an important word or phrase in the document. Perhaps it a common article or preposition, or a key topic in the sample. And then if I see dog/crown/apple/bird many times, I would consider the possibility that it is related to the previous word, possibly in a different tense or number.
The whole process would have to involve a lot of the “guess and check” method. If there is something that I am curious about, I would set up a system where it is true and see if I run myself into any dead-ends that way. If not, I may have guessed correctly! If so, it’s back to the drawing board. The process seems that it certainly would not be clear-cut and methodical – I would probably end up constantly jumping back and forth between researching the system and attacking the system itself until I made major headway.
Looking at and attempting to translate a writing system that has not yet been deciphered is a daunting process. I cannot say I honestly know where I would begin. However, if I had a Rosetta Stone of some kind to get the process started, I would approach the problem in a manner similar to those who translated Egyptian hieroglyphics originally. For example, I would start by looking at similarities between the known languages and the unknown languages, such as names of places or people. From there, I would look for general trends or repeated patterns and make a series of educated guesses, narrowing down possibilities for translation, until a successful deciphering of the language could be made. Of course, this is example is a massive oversimplification, but, in my opinion, any sort of deciphering work is a massive “pencil and paper” project that requires time and effort through trial and error to be completed.
Without a Rosetta Stone, the task becomes far more difficult. My first step would be to learn what I could from archaeology and history about the language: where it originates, the lifestyle of the people, what other languages this language could be related to, etc. By learning about the lifestyle of the people who created the language system, I could potentially learn important clues about how they would relate to their writing. For example, writing systems that are boustrophedons are theorized as coming from early agricultural societies because the writing is designed to represent the back and forth movements of an ox plowing a field. Hopefully, I would also learn that the undeciphered language is related to a known translated language. If so, I would learn as much about the known language and then look for related patterns between the two writing systems which I could use to hopefully translate the unknown language. Some examples of patterns I would look for would be similar characters, similar sentence structure or punctuation marks, similarly spelled words, etc.
The process of decoding an unknown language is remarkably complex and more than slightly daunting. I would hope that if I were given the task of decoding a language, I would have the aid of a wonderful artifact similar to the Rosetta Stone in nature, otherwise, my wish would be for the next best thing available to me: quite a bit of luck.
After reading a bit about the history of the Rosetta stone and the various undeciphered languages in the world, I’m at a loss as to how to write this assignment. If confronting a language system for the first time, my approach would be to look for patterns. This is great if I’m trying to translate one to one with English or a language that I already know (I assume like the phrase Professor Beyer sent to us via email). I can look for the most common letters or symbols; these must correlate to the most common letters in my alphabet. If, for instance, I can solve which symbol stands for t in this new alphabet, I can radically limit the possible choices for a number of other letters. It seems like a tireless game of guesswork.
If I were translating a system that didn’t have a one to one correlation, I have no idea how I would begin. I suppose the work would begin again with looking for symbols. If the writing system were pictographic, it would make sense to look for the most commonly occurring sign and then look at the ones that come before it and after it. The context would also have to taken into consideration. If the text were found in a tomb, for instance, the content would probably cover the life and lineage of the deceased or the contents of the tomb. The process seems even more futile since you can decipher writing systems into a language that is no longer known.
For the code that Prof. Beyer sent us, I have tried substituting various letters for other letters. This is mainly trial and error work with very little logic supporting it. I keep substituting letters until I run into a problem, like a word that begins with “tn,” a consonant cluster that is not used in English. Then I see which letter or letters are most likely to be incorrect and try different combinations. I also looked at the small 2 and 3 letter words, which are few in number and should be easier to figure out. Two-letter words in English include: am, an, as, at, be, if, is, it, of, on, or, etc. The 2-letter words from the code ar “iw” and “ig.” I know that the first two letters are the same, so it will probably be a pair of words like at and as, or of and on. The three-letter words from the code are: ctp, oir, ekp, cwb. Our clue is to find patterns in 3-letter words in English that match the patterns in the code. I’m going to work on this some more and get back to y’all when I have an answer.
Going first with the assumption that I ran across a lengthy sample of unknown text in an unknown language without translations in other (possibly known) languages, I would, as Kevin said, research. I would want to know if any other known writing systems (whether deciphered or not) were similar to my text; I would try to find any other systems mine may have evolved into or from. Hopefully, I could connect my system to some other language. However, this step wouldn’t do me much good if the second system were unknown, or, if known, could not illumine my text. If my text were, for example, a syllabary, and the secondary system used an identical system of characters but was actually an alphabet, I would probably not get anywhere deciphering my text.
The easiest way to decipher an unknown writing system would be in a Rosetta Stone-like situation, in which the text was written in both the unknown system and a known system. Proper nouns would be a good place to start, if it were possible to identify the proper nouns in the unknown system. Hieroglyphics make the job relatively easy by surrounding the royal names with cartouches. In an ideal situation (which, difficult as the process would be, my Rosetta Stone scenario is), the sounds representing the proper name in the known language would sound similar in the unknown language. A comparison between enough proper nouns could reveal the sounds at least a handful of characters or symbols represent; hopefully, then, one would find some words borrowed from one language to another and begin to work out what further characters or symbols mean.
First, I would like to say that I am extremely impressed by any successful efforts to decipher ancient texts describing languages that are no longer spoken. It seems to be an unthinkable task to decipher meaning from inscriptions when we know neither the meanings of the symbols, the nature of the language nor how it sounds, nor what the inscriptions are supposed to represent. Furthermore, to try and put sounds to the inscriptions seems even more impossible, and even after much thought, it seems one would need to hear a speaker of the language to even begin to do this.
After some thought and research, I have come to the conclusion that in order to begin, one must organize the language in some way. The first thing I would do, would be to search for patterns in the way the language is organized. What constitutes a complete idea, how does one break that idea up into parts. Then I would search for repetition, in doing this, one may be able to discover grammatical elements to the language, such as articles, numbers or cases. The placement of words might also be a useful element here. For example, one could examine if an element is consistently attached to another, placed before a group of elements, or at the end of it, or if several elements always appear together. Being able to recognize grammatical elements may help focus attention on what elements actually carry meaning. Again, one could look for repetition that is less frequent, perhaps indicating similar meanings.
Here is where I believe one cannot go farther without some sort of reference. The most basic for of this reference would be archeological evidence that might indicate what purpose an inscription serves. For example, is the writing on a tablet that may have been used to record transactions? The inscriptions and symbols used for this purpose probably contain a specific type of meaning, such as numbers and items typically traded in a transaction. Then, I would see if any of these symbols are used first in other inscriptions serving similar purposes. This might help to identify which symbols represent numbers, a more universal concept, and which represent the nouns, which would be specific to each type of transaction. Then, I would try and look for these same symbols in an inscription created for an entirely different purpose. One could then identify certain elements in our example of tablets used for transactions, one might be able to identify nouns, numbers or maybe even verbs. What would be important at this stage would be primarily the use of a word in a sentence or grouping, and in what order these fall in a grouping. Are nouns always in the same place, do numbers always fall near nouns, before or after, etc. Perhaps in this way, one could begin to predict what the function of unknown symbols might be.
In order to search for semantic meaning, again context would be useful, especially knowledge about the culture. For example if we knew that a community’s primary activity was sheep farming, one could search for symbols that could potentially be related to sheep farming. By comparing the different occurrences of these symbols one might be able to be more specific about their meanings.
If one has a translation of a text into another language, as with the rosetta stone, one could go about looking at an individual language to try and see which elements are grammatical and which contain meaning, then comparing the two languages to decipher the specific meanings of those words, based on location in the text, or similar symbollic elements.
If one knows a later descendant of an undeciphered language, on could use the similar elements to begin to apply some semantic context to a text. For example if an old symbol has now taken on a more grammatical use such as a plural, one could postualte as to its origional sense as a word that contained meaning. Visually, symbolls may have simplified. One might be able to identify their origional unsimplified versions.
Like the other posts have said, trying to decipher an unknown language is a very daunting task, no matter the situation. The possession of some sort of Rosetta Stone would certainly give the best clues, and is probably the only dependable way to accurately decode a language. If the language were alphabetic or syllabic, I would first identify any proper names to get some sense of any basic patterns in letter combinations. I would also do the same thing with any small words or particles that corresponded between the two (or more) systems. From there, I guess I would move up into larger and more complex words and concepts. I would attempt to identify if the language used articles or pronouns or the like, and whether any prepositions were separate words or morphological units tacked on to other words. If it were clearly a pictographic language, I would probably attempt to find any proper names, as before, but failing that I would probably go the most common symbols and look for any correlation between the known language. Either way it would be a very long and involved process with lots of trial and error.
If I didn’t have any kind of deciphering aid, it would probably all be about luck. I think I would first make a list of all grammatical and morphological concepts from languages I knew and languages that I could easily research, and then see if I could identify any of these same structures in the target language. If the script had noticeable word spacing and punctuation, then I may have a fighting chance to find similarities with known languages and go from there. If the writing was just solid line of characters like the Rongo Rongo script (or even like old Latin and Greek inscriptions if I were totally unfamiliar with the alphabet), I would be in trouble. I would look at the text to see if I could find any patterns that may be some form of verbal punctuation, like the small words that mark a sense break in Classical Greek. Trying to identify any common words that may correlate with something important to that culture is also a possibility, but even if I were able to identify some individual words/symbols, I don’t think that would bring me any closer to an understanding.
Trying to decipher a written language that is no longer used would be an extremely difficult challenge without a “Rosetta Stone.” Attempting to solve the meaning would be a challenge in itself, while trying to pronounce the language would be an entirely different task in itself.
However, assuming that I did have a “Rosetta Stone” to help me decipher the unknown text, I would first try to find the same symbols within the writing and assume that each is represented either a syllable or a letter. Knowing the language of the undeciphered script, or at least having a geographical idea of where it came from would also help as I would be able to compare it with other languages that are similar to it.
At first glance, trying to see the “shape” or “form” of the writing would help me link back to familiar languages such as either Latin, if the writing was in the Roman alphabet, or perhaps Chinese if the writing was mainly based on characters. We would be extremely lucky if we did find similarities between the unknown language and the languages that we are already familiar with as it would make the decoding much easier. However, if the text was written in entirely different symbols or characters, it would double the deciphering process as we would have to decide whether if vowels and consonants exist, or if a letter or symbol can have more than one meaning or reading. In addition, it would also be helpful to know the direction of the written language, if is written from top to bottom, or from left to right.
After the “primary analysis” of trying to understand how the language is placed together, I would then try to decode frequently appearing groups of letters or syllables because this might be the important nouns, which could also be assumed as the main point of the script. Next, I would try to decode the longer groups of letters that share similar endings, which might be conjugates or plurals. With some luck, I might be able to grasp as basic sense of grammar in the language as I could try to identify particles, nouns and perhaps even verbs. The placement of certain words may also help decide the sentence structure of the language.
Looking at the individual words themselves, I would also guess that more frequently appearing symbols or characters would be considered as vowels (if any) and would try to identify these before other consonants, as there would probably be more vowels than consonants.
However, all is easier said than being done. I remember thumbing through a friend’s Arabic textbook where every sentence in Arabic contained a direct sentence-to-sentence in English. I remember being very confused and found myself having the hardest time to even understand a simple sentence that read: “I am a student.” Similarly, I would imagine that it would be extremely difficult (almost impossible!) to decipher a written language with no form of translation in a known language, such as the “Rosetta Stone”.
If I encounter a writing system that I am not familiar with, I would first refer it to other language systems. Even extinct language systems would be helpful, because the writing system that I am working with may have similar patterns to writing systems that had been used in the past. If I am unsuccessful, I would take non-linguistic approaches to find solutions. I would want to estimate the time the writing system was created by using methods such as radiocarbon dating. If we do not have information about the location from where the writing system was found in and more importantly written in, studying the material of the medium the writing system was written on would be useful. Although these may seem trivial, determining the setting in which the written system was created can narrow down the search.
The first linguistic approach I would take is classifying the writing system into one of the different writing system types. It would be easy to determine whether the writing system is a pictogram, logogram, or logogram. Profuse characters or picture-like characters indicate that my writing system can be classified as one of these writing systems. Other language types (Abjads, Alphabets, and Syllabic Alphabets) have some kind of vowels (cases for Abjads) which are written in all phonetic languages.
Afterwards, I would pick out all the characters/letters from the given text. Hopefully, with the help of computers, I would be able to come up with a few possible pronunciation of the writing system. Since virtually all written systems are derived from their spoken form, successful pronunciation from the writing system would put me closer to decoding the writing system.
After I was successful in decipher the pronunciation of the writing system, I would compare these sounds of writing system with other languages of the same setting. It is highly likely that languages of neighboring nations/tribes would have significant similarities. I can take advantage of this similarity to derive possible meaning of the words in the text. After determining the definition of the words, grammar patterns and syntax can also be derived, which would complete decoding the newly discovered written system.
I have to say that I didn’t realize how difficult it is to decipher an unknown writing system until I saw the mysterious message on a piece of bark outside the professor’s door.
I was completely lost and even thought that the original texts got distorted while traveling in the cyber space (It happens to Chinese sometimes in webmail). Having read about the decipherment of the Rosetta Stone, I feel that it’s almost impossible to decipher a written text if there were not a vast amount of writing samples from different arrays of life or the decipherers had no previous knowledge of a related language. As we know that “Champollion could read both Greek and Coptic language. He was able to figure out what the seven Demotic signs in Coptic were. By looking at how these signs were used in Coptic he was able to work out what they stood for. Then he began tracing these Demotic signs back to hieroglyphic signs. By working out what some hieroglyphs stood for, he could make educated guesses about what the other hieroglyphs stood for” the decipherment of the Rosetta Stone didn’t simply come from a daydream or imagination.
So if I am facing a writing system for the first time, firstly I have to make sure where the texts are found. Since I imagine the texts don’t just pop up by itself, the geographic location should be traced. If possible I will take advantage of the modern technology to analyze the original sample (such as the stone) and see in which time period that system was used. Then I will start to get familiar with the known languages that are spoken close to that area and ideally find someone who could read in those languages. Having had the knowledge of the neighbor writing systems, I will go find as many original texts of this system as possible and try to find patterns. There must be some graphs that appear over and over again. At this stage I probably could compare these symbols with the ones of the other writing systems close by and see if there is anything similar. During this process I can apply the knowledge of how languages mix and transform and come up with as many kinds of combinations as possible. Then…I'll just see if that illuminating moment is willing to come or not…
Not every effort gets paid off anyways.
Source:
http://en.wikipedia.org/wiki/Rosetta_Stone
Deciphering a writing system which is currently unknown presents itself as an enormously difficult and complicated task. It seems like the best place to start would be to find out as much as possible about the circumstances surrounding its creation. When might it date from? Was it created for a specific purpose? Do other writing systems exist from similar places or times? Any obtainable information might help lead to valuable clues about the system in question. If an already deciphered, similar writing system could be found, a comparison of that system to the new system would prove most helpful.
Looking for patterns seems like the next logical step to take. Are there any characters that appear commonly? If so, these could potentially be frequently used letters such as vowels or important, general words depending on whether the system was based upon an alphabetic or pictographic language. Do any aspects pop out as being similar in structure to systems which are already understood? Figuring out where individual words begin and end would be a big step in the right direction. Elements of grammar and their relations to individual words and sentence structure could then be examined.
The task seems nearly impossible. For a completely unknown system, it seems like it would be easier to assign meanings to characters than to figure out how the language is read and pronounced. The help of a Rosetta Stone seems like it would slightly reduce the daunting nature of such a task as examining similarities and differences between the systems could be very illuminating. Nevertheless, a huge amount of effort, patience and determination is required for any attempt to decipher an unknown writing
As with everyone else, it’s very hard to know where to start with an undeciphered language. I admire anyone who has the patience to give it a whirl. Languages can be difficult to interpret on several counts. Unfamiliar scripts, especially if one was trying to decipher a language like Chinese on one’s own, would pose innumerable problems that it is not even possible to imagine that a person could figure the language out. Other languages have noun cases that change the same word into anywhere from two to fifteen different forms. There are also polysynthetic languages, such as Mohawk. In Mohawk, Washakotya'tawitsherahetkvhta'se means "He ruined her dress" (strictly, "He made the thing that one puts on one's body ugly for her"). [Baker, The Atoms of Language] One could tear this word apart and find all of the elements of the sentence. However, if you weren’t familiar with the language, it would be very hard to know just where to pull it apart.
Like others, I would try to go off of a familiar language, particularly in script. If I found an Asian language with noticeable characters, I would try to use Japanese, Korean (much less likely), or Chinese characters to make sense of it. Sanskrit, Greek, and Latin would obviously be of major help as well. I would also look at the history of the area to help determine influences and possible meanings of words (always look at what was important to their culture).
The biggest problem lies in trying to get one language out of another. Welsh can never be derived from English, so to try to analyze Welsh in terms of English is really pointless. The same principle applies to all languages, and thus using one language as the “foundation” for an unknown does not give much of a sense of the unknown language. In the case of Professor Beyer’s encrypted message, it was much, much easier to decipher because it was a cryptex. Once you figured out that o=y and o will always be y, then you can figure out the other letters. This is obviously never what happens in the real world of language.
Post a Comment
<< Home