Author Topic: Please help a non-linguist  (Read 2054 times)

Offline urbandekay

  • New Linguist
  • *
  • Posts: 2
Please help a non-linguist
« on: January 18, 2016, 01:39:17 PM »

PostPosted: Sun Jan 17, 2016 12:28 pm
Post subject: Help a non-linguist   Reply with quoteEdit/Delete this post
I am by no means a linguist. I do, however, need help in researching how a linguist might solve a particular problem. Imagine, if you will, that you have intercepted an analogue radio transmission of speech that you believe to be of non-terrestrial origin. You may assume you have unlimited computer power but no other people to hand and that the supposed alien tongue is compatible with our sensory organs and that the speech is generated with similar organs

1. Would you be able to discern whether or no the tongue was genuinely non-terrestrial and if so how?

2. Would you, in time, be able to translate said tongue and if so how?

3. If translation were not possible, would an attempt to communicate with the sender, presuming no common language, by radio transmission enable translation and if so how?

Offline Daniel

  • Administrator
  • Experienced Linguist
  • *****
  • Posts: 1582
  • Country: us
    • English
Re: Please help a non-linguist
« Reply #1 on: January 18, 2016, 04:49:41 PM »
Quote
1. Would you be able to discern whether or no the tongue was genuinely non-terrestrial and if so how?
Your scenario suggests that the sounds and perhaps general structure of the language is similar to our own. Therefore, it is possible that there is no way from the signal itself to determine that it is not made by humans. Perhaps these aliens are humanlike, at least in their speech abilities. There is no way to determine that a language is from earth unless the language shows non-humanlike properties. It certainly might. The easiest test would be for sounds that can't be produced by humans, although from the signal alone that could be challenging. Beyond that, theories of "Universal Grammar" (among HUMAN languages) would suggest some commonalities, but they are hypotheses about observed human languages, not predictions about what is non-human. So that would be a longshot as a test. The only other way would be for someone who knows a lot about many languages to figure out it just doesn't seem to belong to any groups, by elimination. But there are 6,000 or so languages today, so full elimination would be impossible.

Quote
2. Would you, in time, be able to translate said tongue and if so how?
Major breakthroughs in translation involve some kind of key to the meaning. The Rosetta stone allowed translation of Ancient Egyptian by having a Greek parallel text. Other cases where a historical relationship between two languages has been identified have been solved by comparison and guessing. A lot can be guessed, but there must be some kind of hint as to the meaning. Another way is through observation of correlated behavior. Take a look at this:
https://www.youtube.com/watch?v=sYpWp7g7XWU

Based on ONLY the sound itself, I don't think we could get too far. I don't know if this situation has ever existed for linguists in real life, but it has existed with writing systems. And in those cases it was almost impossible without some clues and context. That's when the breakthroughs happen.

Based on the signal alone, I think we could do some basic analysis, finding common words/sound groups, figuring out some basic grammatical forms, etc. But it would be entirely meaningless without some way to translate a few of the words at least. If it was similar enough to human languages here we might with enough data be able to make some guesses and start to fill in some blanks, but there's no reason to assume it would be. Do they have verbs? Proper names? Etc. In the best case we might be able to make a full logical grammatical description of the language with no actual understanding of what any of the words mean or why they are used-- these words get this ending, these get that one, and so forth. So we could generate random (nonsense) sentences in the language, but not actually work out the meaning.

Regarding context, also see the following:
https://www.youtube.com/watch?v=y1kXCh496U0

Quote
3. If translation were not possible, would an attempt to communicate with the sender, presuming no common language, by radio transmission enable translation and if so how?
Issues with the speed of light might make this very hard. But yes, if enough back and forth communication were possible, eventually the interaction could provide enough of a clue to the meaning. See the video above.
Welcome to Linguist Forum! If you have any questions, please ask.

Offline urbandekay

  • New Linguist
  • *
  • Posts: 2
Re: Please help a non-linguist
« Reply #2 on: January 19, 2016, 05:52:25 AM »
DJR33 thank you for your in depth reply, I wonder if I might trouble you with a follow-up question.

Suppose that the language in question, whilst not a known language, bears remarkable similarity to Hebrew, perhaps not in vocabulary but in grammatical structure and phonetic sound.

1. Would this be immediately apparent?
2. Were there to be occasional words, say no more than a handful, in common to Hebrew, would that be sufficient to enable translation?

Offline Daniel

  • Administrator
  • Experienced Linguist
  • *****
  • Posts: 1582
  • Country: us
    • English
Re: Please help a non-linguist
« Reply #3 on: January 19, 2016, 11:46:30 AM »
Quote
1. Would this be immediately apparent?
"Remarkable similarity"? Sure, to someone who is familiar with Hebrew.
Quote
2. Were there to be occasional words, say no more than a handful, in common to Hebrew, would that be sufficient to enable translation?
Maybe. But full translation would still require context.

The extreme case of difficulty in translating without additional information where only one instance of a word has been recorded is known as a "hapax legomenon". We might be able to guess with context, but without context, and without some other hint at the meaning, there's little we could do.

As an example of understanding the broad aspects of the language without knowing all the vocabulary, consider Lewis Carroll's Jabberwocky poem (easy to find online). We understand generally what's going on, but we don't really know what all the words mean. That's about what translation would look like if we did a good job with some insights and understood the grammatical structure but didn't have any reference for the meaning of individual words and no context.

Setting aside the "extra-terrestrial" aspect of the question, cases like this do exist, mostly with ancient languages that only exist in limited inscriptions, all with varying degrees of translation success. I don't have any references to name at the moment but I'm sure you can find out more. As I said earlier, though, almost always it seems the real success stories involve some sort of key to unlock something about the language.
Welcome to Linguist Forum! If you have any questions, please ask.

Offline Guijarro

  • Forum Regulars
  • Linguist
  • *
  • Posts: 97
  • Country: es
    • Spanish
    • Elucubraciones de José Luis Guijarro
Re: Please help a non-linguist
« Reply #4 on: January 21, 2016, 10:36:09 AM »
Curious!

Yesterday, Spanish papers had the news that some researchers have found a sort of rosetta to help decipher the guanche language (spoken by very ancient folks in the Canary Islands). It seems that there is no evident similarity with any known language --not even the African ones.

Maybe it is a relic from a Martian invasion!

Offline Daniel

  • Administrator
  • Experienced Linguist
  • *****
  • Posts: 1582
  • Country: us
    • English
Re: Please help a non-linguist
« Reply #5 on: January 21, 2016, 05:01:47 PM »
Interesting. Wikipedia lists Guanche (in the Canary Islands) as an extinct Berber language. I don't know more about it. Does the new information suggest otherwise?
Welcome to Linguist Forum! If you have any questions, please ask.